The document describes a proposed algorithm for generating synthetic fingerprints that appear affected by common skin diseases like warts and atopic dermatitis. The algorithm modifies existing synthetic fingerprints generated using the SFinGe tool to simulate the visual effects of these diseases on fingerprint ridge structure and quality. It first analyzes real fingerprints affected by each disease to understand characteristic features. It then introduces methods to localize fingerprints and draw simulated warts or lines/patches onto synthetic fingerprints in a way that decreases their recognition scores and quality metrics when tested using commercial and open-source fingerprint analysis tools. The results confirm the algorithm successfully modifies fingerprints to appear disease-affected.
Paper multi-modal biometric system using fingerprint , face and speechAalaa Khattab
Biometric system is often not able to meet the desired performance requirements.
In order to enable a biometric system to operate effectively in different applications and environments, a multimodal biometric system is preferred.
In this paper introduce a multimodal biometric system which integrates fingerprint verification , face recognition and speaker verification.
novel method of identifying fingerprint using minutiae matching in biometric ...INFOGAIN PUBLICATION
Fingerprint is one of the best apparatus to identify human because of its uniqueness, details information, hard to change and long-term indicators of human identity where there are several biometric feature that can be recycled to endorse the individuality. Identification of fingerprint is very important in forensic science, trace any part of human, collection of crime part and proof from a crime. This paper presents a new method of identifying fingerprint in biometrics security system. Fingerprint is one of the best example in biometric security because it can identify personal information and it is much secure than any other biometric identification system. The experimental result exhibits the performance of the proposed method.
Face recognition is a widely used biometric approach. Face recognition technology has developed rapidly
in recent years and it is more direct, user friendly and convenient compared to other methods. But face
recognition systems are vulnerable to spoof attacks made by non-real faces. It is an easy way to spoof face
recognition systems by facial pictures such as portrait photographs. A secure system needs Liveness
detection in order to guard against such spoofing. In this work, face liveness detection approaches are
categorized based on the various types techniques used for liveness detection. This categorization helps
understanding different spoof attacks scenarios and their relation to the developed solutions. A review of
the latest works regarding face liveness detection works is presented. The main aim is to provide a simple
path for the future development of novel and more secured face liveness detection approach.
Multimodal biometric systems are those that utilize more than one physical or behavioural characteristic for enrolment , verification, or identification.
The foundations for biometrics or identification systems were laid long ago. Today these developments have contributed to the identification of people, access to private sites and all places that need security and order with the help of computerized computers that perform biometric facial recognition, exclusively based on images of human faces for their function. With the extraction of facial midst characteristics of each person provides information used for the detection of the face. This communication also addresses the different processes, stages and methods of feature extraction operated by facial recognition systems. Including the positive and negative aspects of the implementation of these, the advantages and disadvantages, peoples criteria in this respect. Tovbaev Sirojiddin | Karshiboev Nizomiddin "Image Based Facial Recognition" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-4 | Issue-4 , June 2020, URL: https://www.ijtsrd.com/papers/ijtsrd31330.pdf Paper Url :https://www.ijtsrd.com/computer-science/artificial-intelligence/31330/image-based-facial-recognition/tovbaev-sirojiddin
An SVM based Statistical Image Quality Assessment for Fake Biometric DetectionIJTET Journal
Abstract
A biometric system is a computer based system and is used to identify the person on their behavioral and logical characteristics such as (for example fingerprint, face, iris, keystroke, signature, voice, etc.).A typical biometric system consists of feature extraction and matching patterns. But nowadays biometric systems are attacked by using fake biometric samples. This paper described the fingerprint biometric techniques and also introduce the attack on that system and by using Image Quality Assessment for Liveness Detection to know how to protect the system from fake biometrics and also how the multi biometric system is more secure than uni-biometric system. Support Vector Machine (SVM) classification technique is used for training and testing the fingerprint images. The testing onput fingerprint image is resulted as real and fake fingerprint image by quality score matching with the training based real and fake fingerprint samples.
Paper multi-modal biometric system using fingerprint , face and speechAalaa Khattab
Biometric system is often not able to meet the desired performance requirements.
In order to enable a biometric system to operate effectively in different applications and environments, a multimodal biometric system is preferred.
In this paper introduce a multimodal biometric system which integrates fingerprint verification , face recognition and speaker verification.
novel method of identifying fingerprint using minutiae matching in biometric ...INFOGAIN PUBLICATION
Fingerprint is one of the best apparatus to identify human because of its uniqueness, details information, hard to change and long-term indicators of human identity where there are several biometric feature that can be recycled to endorse the individuality. Identification of fingerprint is very important in forensic science, trace any part of human, collection of crime part and proof from a crime. This paper presents a new method of identifying fingerprint in biometrics security system. Fingerprint is one of the best example in biometric security because it can identify personal information and it is much secure than any other biometric identification system. The experimental result exhibits the performance of the proposed method.
Face recognition is a widely used biometric approach. Face recognition technology has developed rapidly
in recent years and it is more direct, user friendly and convenient compared to other methods. But face
recognition systems are vulnerable to spoof attacks made by non-real faces. It is an easy way to spoof face
recognition systems by facial pictures such as portrait photographs. A secure system needs Liveness
detection in order to guard against such spoofing. In this work, face liveness detection approaches are
categorized based on the various types techniques used for liveness detection. This categorization helps
understanding different spoof attacks scenarios and their relation to the developed solutions. A review of
the latest works regarding face liveness detection works is presented. The main aim is to provide a simple
path for the future development of novel and more secured face liveness detection approach.
Multimodal biometric systems are those that utilize more than one physical or behavioural characteristic for enrolment , verification, or identification.
The foundations for biometrics or identification systems were laid long ago. Today these developments have contributed to the identification of people, access to private sites and all places that need security and order with the help of computerized computers that perform biometric facial recognition, exclusively based on images of human faces for their function. With the extraction of facial midst characteristics of each person provides information used for the detection of the face. This communication also addresses the different processes, stages and methods of feature extraction operated by facial recognition systems. Including the positive and negative aspects of the implementation of these, the advantages and disadvantages, peoples criteria in this respect. Tovbaev Sirojiddin | Karshiboev Nizomiddin "Image Based Facial Recognition" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-4 | Issue-4 , June 2020, URL: https://www.ijtsrd.com/papers/ijtsrd31330.pdf Paper Url :https://www.ijtsrd.com/computer-science/artificial-intelligence/31330/image-based-facial-recognition/tovbaev-sirojiddin
An SVM based Statistical Image Quality Assessment for Fake Biometric DetectionIJTET Journal
Abstract
A biometric system is a computer based system and is used to identify the person on their behavioral and logical characteristics such as (for example fingerprint, face, iris, keystroke, signature, voice, etc.).A typical biometric system consists of feature extraction and matching patterns. But nowadays biometric systems are attacked by using fake biometric samples. This paper described the fingerprint biometric techniques and also introduce the attack on that system and by using Image Quality Assessment for Liveness Detection to know how to protect the system from fake biometrics and also how the multi biometric system is more secure than uni-biometric system. Support Vector Machine (SVM) classification technique is used for training and testing the fingerprint images. The testing onput fingerprint image is resulted as real and fake fingerprint image by quality score matching with the training based real and fake fingerprint samples.
GANNON UNIVERSITY
ELECTRICAL AND COMPUTER ENGINEERING DEPARTMENT
FALL2015
GECE 572: DIGITAL SIGNAL PROCESSING
FINGER PRINT RECOGNITION USING MINUTIAE BASED FEATURE
FINAL PROJECT
Prepared by
THADASINA PRUTHVIN REDDY
[email protected]
SALMAN SIDDIQUI
[email protected]
Instructor:
Dr. Ram Sundaram
Table of contents
1. Abstract
2. Introduction
3. Fingerprint matching
4. Pre-processing stage
5. Minutiae extraction stage
6. Post-processing stage
7. Merits & Demerits
8. Applications & future scope
9. Conclusions
10.References
1. Abstract
Nowadays, conventional identification methods such as driver's license, passport, ATM cards and PIN codes do not meet the demands of this wide scale connectivity. Automated biometrics in general, and automated fingerprint authentication in particular, provide efficient solutions to these modern identification problems. Fingerprints have been used for many centuries as a means of identifying people. The fingerprints of individual are unique and are stay unchanged during the life time. Fingerprint matching techniques can be placed into two categories, minutiae-based and correlation based. Minutiae-based techniques first find minutiae points and then map their relative placement on the finger. However, there are some difficulties when using this approach. It is difficult to extract the minutiae points accurately when the fingerprint is of low quality the correlation-based method is able to overcome some of the difficulties of the minutiae-based approach. However, it has some of its own shortcomings. Correlation-based techniques require the precise location of a registration point and are affected by image translation and rotation.
2. Introduction
Biometric recognition refers to the use of distinctive physiological (e.g. fingerprint, palm print, iris, face) and behavioral (e.g. gait, signature) characteristics, called biometric identifiers for recognizing individuals.
Fingerprint recognition is one of the oldest and most reliable biometric used for personal identification. Fingerprint recognition has been used for over 100 years now and has come a long way from tedious manual fingerprint matching. The ancient procedure of matching fingerprints manually was extremely cumbersome and time-consuming and required skilled personnel.
Finger skin is made up of friction ridges and sweat pores all along these ridges. Friction ridges are created during fetal life and only the general shape is genetically defined. The distinguishing nature of physical characteristics of a person is due to both the inherent individual genetic diversity within the human population as well as the random processes affecting the development of the embryo. Friction ridges ...
Biometric system works on behavioral and physiological biometric parameters to spot a person. Every fingerprint contains distinctive options and its recognizing system primarily works on native ridge feature local ridge endings, minutiae, core point, delta, etc. However, fingerprint pictures have poor quality thanks to variations in skin and impression conditions. In personal identification, fingerprint recognition is taken into account the foremost outstanding and reliable technique for matching with keep fingerprints within the information. Minutiae extraction is additional essential step in fingerprint matching. This paper provides plan regarding numerous feature extraction and matching algorithms for fingerprint recognition systems and to seek out that technique is additional reliable and secure.
Fingerprint Minutiae Extraction and Compression using LZW Algorithmijsrd.com
For security and surveillance automated personal identification is major issue. We can see a lot varieties of biometric systems like face detection, fingerprint recognition, iris recognition, voice recognition, palm recognition etc. In our project we will only go for fingerprint recognition. Never two peoples have exactly same fingerprints even twins, they are totally unique. The sensors capture the finger prints of humans and convert them into images and a minutiae extraction algorithm extracts the location of minutiae points called termination and bifurcation. A database system stores these patterns and minutiae points of fingerprint. A large storage space required to store bifurcation and termination points for the fingerprint database. LZW compression algorithm has been used to reduce the size of data. With LZW applied on these extracted minutiae points, these minutiae points get encoded which add more security feature.
Biometrics technology is rapidly progressing and offers attractive opportunities. In recent years, biometric authentication has grown in popularity as a means of personal identification in ATM authentication systems. The prominent biometric methods that may be used for authentication include fingerprint, palmprint, handprint, face recognition, speech recognition, dental and eye biometrics. In this paper, a microcontroller based prototype of ATM cashbox access system using fingerprint sensor module is implemented. An 8-bit PIC16F877A microcontroller developed by Microchip Technology is used in the system. The necessary software is written in Embedded 'C' and the system is tested.
Touch is one of the most common forms of sign language used in oral communication. It is most commonly used by deaf and dumb people who have difficulty hearing or speaking. Communication between them or ordinary people. Various sign-language programs have been developed by many manufacturers around the world, but they are relatively flexible and affordable for end users. Therefore, this paper has presented software that introduces a type of system that can automatically detect sign language to help deaf and mute people communicate better with other people or ordinary people. Pattern recognition and hand recognition are developing fields of research. Being an integral part of meaningless hand-to-hand communication plays a major role in our daily lives. The handwriting system gives us a new, natural, easy-to-use communication system with a computer that is very common to humans. Considering the similarity of the human condition with four fingers and one thumb, the software aims to introduce a real-time hand recognition system based on the acquisition of some of the structural features such as position, mass centroid, finger position, thumb instead of raised or folded finger.
As we know the fingerprint is unique of every living objects. It is quite difficult to find out the prints.
Usually the Forensics use Fine powder and duct tapes to identify the prints of living object. As powder is
exceptionally muddled, so such molecule can cause loss of information after that examination the information is
coordinated with the system. The proposed system consists of an embedded device in which it consists of ultra
light to glow the fingerprints details. After that we can detect the fingerprint, analysis and it will checks on the
database, and it will return the output after matching. For matching and analysis of the Fingerprint, we will be
using the Algorithm for matching.
IRJET-Gaussian Filter based Biometric System Security EnhancementIRJET Journal
M.Selvi, T.Manickam, C.N.Marimuthu"Gaussian Filter based Biometric System Security Enhancement", International Research Journal of Engineering and Technology (IRJET), Volume2,issue-01 April 2015.e-ISSN:2395-0056, p-ISSN:2395-0072. www.irjet.net
Abstract
A novel software-based fake detection method that can be used in multiple biometric systems to detect different types of fraudulent access attempts. To ensure the actual presence of a real legitimate trait in contrast to a fake self-manufactured synthetic or reconstructed sample is a significant problem in biometric authentication, which requires the development of new and efficient protection measures. To enhance the security of biometric recognition frameworks, by adding liveness assessment in a fast, user-friendly, and non-intrusive manner, through the use of image quality assessment.
The proposed approach presents a very low degree of complexity, which makes it suitable for real-time applications, using 25 general image quality features extracted from one image (i.e., the same acquired for authentication purposes) to distinguish between legitimate and impostor samples. Multi-biometric and Multi-attack protection method which targets to overcome part of these limitations through the use of Image Quality Assessment (IQA).
Moreover, being software-based, it presents the usual advantages of this type of approaches: fast, as it only needs one image (i.e., the same sample acquired for biometric recognition) to detect whether it is real or fake, non-intrusive; user-friendly (transparent to the user), cheap and easy to embed in already functional systems and no hardware is required).
PREPROCESSING ALGORITHM FOR DIGITAL FINGERPRINT IMAGE RECOGNITIONijcsity
Biometrics is one of the most used technologies in the field of security due to its reliability and performance. It is based on several physical human characteristics but the most used technology is the fingerprint recognition, and since we must carry out an image processing to be able to exploit the data in each fingerprint we propose in this article an image preprocessing procedure in order to improve its quality before extracting the necessary information for the comparison phase.
Hand gesture recognition system has received great attention in the recent few years because of its manifoldness applications and the ability to interact with machine efficiently through human computer interaction. In this work Hand segmentation using color models is introduced for obtaining hand gestures or detecting user’s hand by color segmentation technique for faster, better, robust, accurate and real-time applications. There are many such color models available for human hand and human skin detection with relative advantages and disadvantages in the field of Image Processing. For the purpose of hand Segmentation mix model approach has been adopted for best results. For detection of Hand from an image. The proposed approach is found to be accurate and effective for multiple conditions
Fingerprints are imprints formed by friction
ridges of the skin and thumbs. They have long been used for
identification because of their immutability and individuality.
Immutability refers to the permanent and unchanging character
of the pattern on each finger. Individuality refers to the
uniqueness of ridge details across individuals; the probability
that two fingerprints are alike is about 1 in 1.9x1015. In despite of
this improvement which is adopted by the Federal Bureau of
Investigation (FBI), the fact still is “The larger the fingerprint
files became, the harder it was to identify somebody from their
fingerprints alone. Moreover, the fingerprint requires one of the
largest data templates in the biometric field”. The finger data
template can range anywhere from several hundred bytes to over
1,000 bytes depending upon the level of security that is required
and the method that is used to scan one's fingerprint. For these
reasons this work is motivated to present another way to tackle
the problem that is relies on the properties of Vector
Quantization coding algorithm.
7 multi biometric fake detection system using image quality based liveness de...INFOGAIN PUBLICATION
Biometric systems mostly popular in all over the world because of its user friendly and credible nature in security. In spite of this advantages, many attacks that done through synthetic , self manufactured, fake, reconstructed samples affected on the performance and accuracy of biometric system which becomes major problem in biometrics. Hence, new effective measures have to be taken to protect the biometric systems. In this paper, we propose novel software based multi-biometric fake detection system to detect various types of attacks. The main moto of this system is to enhance security level of biometric recognition systems through Image Quality Assessment (IQA) which is one of the liveness detection method.25 image quality measures calculated from test image which used to classify between real and fake trait using Linear Discriminative Analysis(LDA) classifier. The experimental results is done on the database of 2D face and fingerprint modalities, shows the proposed system is ease in implementation in real time application as complexities is very less because of one input image. Also this system is fast, user-friendly, non-intrusive which is more competitive with any other state of the art approaches, classifies between real and fake traits.
Biometrics Authentication of Fingerprint with Using Fingerprint Reader and Mi...TELKOMNIKA JOURNAL
The idea of security is as old as humanity itself. Between oldest methods of security were
included simple mechanical locks whose authentication element was the key. At first, a universal–simple
type, later unique for each lock. A long time had mechanical locks been the sole option for protection
against unauthorized access. The boom of biometrics has come in the 20th century, and especially in
recent years, biometrics is much expanded in the various areas of our life. Opposite of traditional security
methods such as passwords, access cards, and hardware keys, it offers many benefits. The main benefits
are the uniqueness and the impossibility of their loss. The main benefits are the uniqueness and the
impossibility of their loss. Therefore we focussed in this paper on the the design of low cost biometric
fingerprint system and subsequent implementation of this system in praxtise. Our main goal was to create
a system that is capable of recognizing fingerprints from a user and then processing them. The main part
of this system is the microcontroller Arduino Yun with an external interface to the scan of the fingerprint
with a name Adafruit R305 (special reader). This microcontroller communicates with the external database,
which ensures the exchange of data between Arduino Yun and user application. This application was
created for (currently) most widespread mobile operating system-Android.
Ijaems apr-2016-1 Multibiometric Authentication System Processed by the Use o...INFOGAIN PUBLICATION
The present day authentication system is mostly uni-model i.e having only single authentication method which can be either finger print, iris , palm veins ,etc. Thus these models have to contend with a variety of problems such as absurd or unusual data, non-versatility; un authorized attempts, and huge amount of error rates. Some of these limitations can be reduced or stopped by the use of multimodal biometric systems that integrate the evidence presented by several sources of information. This paper converses a multi biometric based authentication system based on Fusion algorithm using a key. Our work mainly focuses on the fusion algorithm, i.e fusion of finger and palm print out of which the greatest features from the above two traits are taken into account. With minimum possible features the fusion of the both the traits is carried out. Then some key is generated for multi biometric authentication. By processing the test image of a person, the identity of the person is displayed with his/her own image. By the fusion algorithm, it is found that it has less computation time compared to the existing algorithms. By matching results, we validate and authenticate the particular individual.
Attendance Monitoring System of Students Based on Biometric and GPS Tracking ...IJAEMSJORNAL
This paper is a study of a fingerprint recognition system based on minutiae based fingerprint algorithms used in various techniques. This line of track mainly involves extraction of minutiae points from the model fingerprint images and fingerprint matching based on the number of minutiae pairings among to fingerprints. This paper also provides the design method of fingerprint based student attendance with help of GSM. This system ignores the requirement for stationary materials and personnel for keeping of records. The main objective of this project is to develop an embedded system, which is used for security applications. The biometrics technology is rapidly progressing and offers attractive opportunities. In recent years, biometric authentication has grown in popularity as a means of personal identification in college administration systems. The prominent biometric methods that may be used for authentication include fingerprint, palmprint, and handprint, face recognition, speech recognition, dental and eye biometrics. In this paper, a microcontroller based prototype of attendance system using fingerprint sensor and face recognition module is implemented. The tracking module is used here to identify the location of the missing person.
Francesca Gottschalk - How can education support child empowerment.pptxEduSkills OECD
Francesca Gottschalk from the OECD’s Centre for Educational Research and Innovation presents at the Ask an Expert Webinar: How can education support child empowerment?
June 3, 2024 Anti-Semitism Letter Sent to MIT President Kornbluth and MIT Cor...Levi Shapiro
Letter from the Congress of the United States regarding Anti-Semitism sent June 3rd to MIT President Sally Kornbluth, MIT Corp Chair, Mark Gorenberg
Dear Dr. Kornbluth and Mr. Gorenberg,
The US House of Representatives is deeply concerned by ongoing and pervasive acts of antisemitic
harassment and intimidation at the Massachusetts Institute of Technology (MIT). Failing to act decisively to ensure a safe learning environment for all students would be a grave dereliction of your responsibilities as President of MIT and Chair of the MIT Corporation.
This Congress will not stand idly by and allow an environment hostile to Jewish students to persist. The House believes that your institution is in violation of Title VI of the Civil Rights Act, and the inability or
unwillingness to rectify this violation through action requires accountability.
Postsecondary education is a unique opportunity for students to learn and have their ideas and beliefs challenged. However, universities receiving hundreds of millions of federal funds annually have denied
students that opportunity and have been hijacked to become venues for the promotion of terrorism, antisemitic harassment and intimidation, unlawful encampments, and in some cases, assaults and riots.
The House of Representatives will not countenance the use of federal funds to indoctrinate students into hateful, antisemitic, anti-American supporters of terrorism. Investigations into campus antisemitism by the Committee on Education and the Workforce and the Committee on Ways and Means have been expanded into a Congress-wide probe across all relevant jurisdictions to address this national crisis. The undersigned Committees will conduct oversight into the use of federal funds at MIT and its learning environment under authorities granted to each Committee.
• The Committee on Education and the Workforce has been investigating your institution since December 7, 2023. The Committee has broad jurisdiction over postsecondary education, including its compliance with Title VI of the Civil Rights Act, campus safety concerns over disruptions to the learning environment, and the awarding of federal student aid under the Higher Education Act.
• The Committee on Oversight and Accountability is investigating the sources of funding and other support flowing to groups espousing pro-Hamas propaganda and engaged in antisemitic harassment and intimidation of students. The Committee on Oversight and Accountability is the principal oversight committee of the US House of Representatives and has broad authority to investigate “any matter” at “any time” under House Rule X.
• The Committee on Ways and Means has been investigating several universities since November 15, 2023, when the Committee held a hearing entitled From Ivory Towers to Dark Corners: Investigating the Nexus Between Antisemitism, Tax-Exempt Universities, and Terror Financing. The Committee followed the hearing with letters to those institutions on January 10, 202
More Related Content
Similar to Generation of Skin Diseases into Synthetic Fingerprints
GANNON UNIVERSITY
ELECTRICAL AND COMPUTER ENGINEERING DEPARTMENT
FALL2015
GECE 572: DIGITAL SIGNAL PROCESSING
FINGER PRINT RECOGNITION USING MINUTIAE BASED FEATURE
FINAL PROJECT
Prepared by
THADASINA PRUTHVIN REDDY
[email protected]
SALMAN SIDDIQUI
[email protected]
Instructor:
Dr. Ram Sundaram
Table of contents
1. Abstract
2. Introduction
3. Fingerprint matching
4. Pre-processing stage
5. Minutiae extraction stage
6. Post-processing stage
7. Merits & Demerits
8. Applications & future scope
9. Conclusions
10.References
1. Abstract
Nowadays, conventional identification methods such as driver's license, passport, ATM cards and PIN codes do not meet the demands of this wide scale connectivity. Automated biometrics in general, and automated fingerprint authentication in particular, provide efficient solutions to these modern identification problems. Fingerprints have been used for many centuries as a means of identifying people. The fingerprints of individual are unique and are stay unchanged during the life time. Fingerprint matching techniques can be placed into two categories, minutiae-based and correlation based. Minutiae-based techniques first find minutiae points and then map their relative placement on the finger. However, there are some difficulties when using this approach. It is difficult to extract the minutiae points accurately when the fingerprint is of low quality the correlation-based method is able to overcome some of the difficulties of the minutiae-based approach. However, it has some of its own shortcomings. Correlation-based techniques require the precise location of a registration point and are affected by image translation and rotation.
2. Introduction
Biometric recognition refers to the use of distinctive physiological (e.g. fingerprint, palm print, iris, face) and behavioral (e.g. gait, signature) characteristics, called biometric identifiers for recognizing individuals.
Fingerprint recognition is one of the oldest and most reliable biometric used for personal identification. Fingerprint recognition has been used for over 100 years now and has come a long way from tedious manual fingerprint matching. The ancient procedure of matching fingerprints manually was extremely cumbersome and time-consuming and required skilled personnel.
Finger skin is made up of friction ridges and sweat pores all along these ridges. Friction ridges are created during fetal life and only the general shape is genetically defined. The distinguishing nature of physical characteristics of a person is due to both the inherent individual genetic diversity within the human population as well as the random processes affecting the development of the embryo. Friction ridges ...
Biometric system works on behavioral and physiological biometric parameters to spot a person. Every fingerprint contains distinctive options and its recognizing system primarily works on native ridge feature local ridge endings, minutiae, core point, delta, etc. However, fingerprint pictures have poor quality thanks to variations in skin and impression conditions. In personal identification, fingerprint recognition is taken into account the foremost outstanding and reliable technique for matching with keep fingerprints within the information. Minutiae extraction is additional essential step in fingerprint matching. This paper provides plan regarding numerous feature extraction and matching algorithms for fingerprint recognition systems and to seek out that technique is additional reliable and secure.
Fingerprint Minutiae Extraction and Compression using LZW Algorithmijsrd.com
For security and surveillance automated personal identification is major issue. We can see a lot varieties of biometric systems like face detection, fingerprint recognition, iris recognition, voice recognition, palm recognition etc. In our project we will only go for fingerprint recognition. Never two peoples have exactly same fingerprints even twins, they are totally unique. The sensors capture the finger prints of humans and convert them into images and a minutiae extraction algorithm extracts the location of minutiae points called termination and bifurcation. A database system stores these patterns and minutiae points of fingerprint. A large storage space required to store bifurcation and termination points for the fingerprint database. LZW compression algorithm has been used to reduce the size of data. With LZW applied on these extracted minutiae points, these minutiae points get encoded which add more security feature.
Biometrics technology is rapidly progressing and offers attractive opportunities. In recent years, biometric authentication has grown in popularity as a means of personal identification in ATM authentication systems. The prominent biometric methods that may be used for authentication include fingerprint, palmprint, handprint, face recognition, speech recognition, dental and eye biometrics. In this paper, a microcontroller based prototype of ATM cashbox access system using fingerprint sensor module is implemented. An 8-bit PIC16F877A microcontroller developed by Microchip Technology is used in the system. The necessary software is written in Embedded 'C' and the system is tested.
Touch is one of the most common forms of sign language used in oral communication. It is most commonly used by deaf and dumb people who have difficulty hearing or speaking. Communication between them or ordinary people. Various sign-language programs have been developed by many manufacturers around the world, but they are relatively flexible and affordable for end users. Therefore, this paper has presented software that introduces a type of system that can automatically detect sign language to help deaf and mute people communicate better with other people or ordinary people. Pattern recognition and hand recognition are developing fields of research. Being an integral part of meaningless hand-to-hand communication plays a major role in our daily lives. The handwriting system gives us a new, natural, easy-to-use communication system with a computer that is very common to humans. Considering the similarity of the human condition with four fingers and one thumb, the software aims to introduce a real-time hand recognition system based on the acquisition of some of the structural features such as position, mass centroid, finger position, thumb instead of raised or folded finger.
As we know the fingerprint is unique of every living objects. It is quite difficult to find out the prints.
Usually the Forensics use Fine powder and duct tapes to identify the prints of living object. As powder is
exceptionally muddled, so such molecule can cause loss of information after that examination the information is
coordinated with the system. The proposed system consists of an embedded device in which it consists of ultra
light to glow the fingerprints details. After that we can detect the fingerprint, analysis and it will checks on the
database, and it will return the output after matching. For matching and analysis of the Fingerprint, we will be
using the Algorithm for matching.
IRJET-Gaussian Filter based Biometric System Security EnhancementIRJET Journal
M.Selvi, T.Manickam, C.N.Marimuthu"Gaussian Filter based Biometric System Security Enhancement", International Research Journal of Engineering and Technology (IRJET), Volume2,issue-01 April 2015.e-ISSN:2395-0056, p-ISSN:2395-0072. www.irjet.net
Abstract
A novel software-based fake detection method that can be used in multiple biometric systems to detect different types of fraudulent access attempts. To ensure the actual presence of a real legitimate trait in contrast to a fake self-manufactured synthetic or reconstructed sample is a significant problem in biometric authentication, which requires the development of new and efficient protection measures. To enhance the security of biometric recognition frameworks, by adding liveness assessment in a fast, user-friendly, and non-intrusive manner, through the use of image quality assessment.
The proposed approach presents a very low degree of complexity, which makes it suitable for real-time applications, using 25 general image quality features extracted from one image (i.e., the same acquired for authentication purposes) to distinguish between legitimate and impostor samples. Multi-biometric and Multi-attack protection method which targets to overcome part of these limitations through the use of Image Quality Assessment (IQA).
Moreover, being software-based, it presents the usual advantages of this type of approaches: fast, as it only needs one image (i.e., the same sample acquired for biometric recognition) to detect whether it is real or fake, non-intrusive; user-friendly (transparent to the user), cheap and easy to embed in already functional systems and no hardware is required).
PREPROCESSING ALGORITHM FOR DIGITAL FINGERPRINT IMAGE RECOGNITIONijcsity
Biometrics is one of the most used technologies in the field of security due to its reliability and performance. It is based on several physical human characteristics but the most used technology is the fingerprint recognition, and since we must carry out an image processing to be able to exploit the data in each fingerprint we propose in this article an image preprocessing procedure in order to improve its quality before extracting the necessary information for the comparison phase.
Hand gesture recognition system has received great attention in the recent few years because of its manifoldness applications and the ability to interact with machine efficiently through human computer interaction. In this work Hand segmentation using color models is introduced for obtaining hand gestures or detecting user’s hand by color segmentation technique for faster, better, robust, accurate and real-time applications. There are many such color models available for human hand and human skin detection with relative advantages and disadvantages in the field of Image Processing. For the purpose of hand Segmentation mix model approach has been adopted for best results. For detection of Hand from an image. The proposed approach is found to be accurate and effective for multiple conditions
Fingerprints are imprints formed by friction
ridges of the skin and thumbs. They have long been used for
identification because of their immutability and individuality.
Immutability refers to the permanent and unchanging character
of the pattern on each finger. Individuality refers to the
uniqueness of ridge details across individuals; the probability
that two fingerprints are alike is about 1 in 1.9x1015. In despite of
this improvement which is adopted by the Federal Bureau of
Investigation (FBI), the fact still is “The larger the fingerprint
files became, the harder it was to identify somebody from their
fingerprints alone. Moreover, the fingerprint requires one of the
largest data templates in the biometric field”. The finger data
template can range anywhere from several hundred bytes to over
1,000 bytes depending upon the level of security that is required
and the method that is used to scan one's fingerprint. For these
reasons this work is motivated to present another way to tackle
the problem that is relies on the properties of Vector
Quantization coding algorithm.
7 multi biometric fake detection system using image quality based liveness de...INFOGAIN PUBLICATION
Biometric systems mostly popular in all over the world because of its user friendly and credible nature in security. In spite of this advantages, many attacks that done through synthetic , self manufactured, fake, reconstructed samples affected on the performance and accuracy of biometric system which becomes major problem in biometrics. Hence, new effective measures have to be taken to protect the biometric systems. In this paper, we propose novel software based multi-biometric fake detection system to detect various types of attacks. The main moto of this system is to enhance security level of biometric recognition systems through Image Quality Assessment (IQA) which is one of the liveness detection method.25 image quality measures calculated from test image which used to classify between real and fake trait using Linear Discriminative Analysis(LDA) classifier. The experimental results is done on the database of 2D face and fingerprint modalities, shows the proposed system is ease in implementation in real time application as complexities is very less because of one input image. Also this system is fast, user-friendly, non-intrusive which is more competitive with any other state of the art approaches, classifies between real and fake traits.
Biometrics Authentication of Fingerprint with Using Fingerprint Reader and Mi...TELKOMNIKA JOURNAL
The idea of security is as old as humanity itself. Between oldest methods of security were
included simple mechanical locks whose authentication element was the key. At first, a universal–simple
type, later unique for each lock. A long time had mechanical locks been the sole option for protection
against unauthorized access. The boom of biometrics has come in the 20th century, and especially in
recent years, biometrics is much expanded in the various areas of our life. Opposite of traditional security
methods such as passwords, access cards, and hardware keys, it offers many benefits. The main benefits
are the uniqueness and the impossibility of their loss. The main benefits are the uniqueness and the
impossibility of their loss. Therefore we focussed in this paper on the the design of low cost biometric
fingerprint system and subsequent implementation of this system in praxtise. Our main goal was to create
a system that is capable of recognizing fingerprints from a user and then processing them. The main part
of this system is the microcontroller Arduino Yun with an external interface to the scan of the fingerprint
with a name Adafruit R305 (special reader). This microcontroller communicates with the external database,
which ensures the exchange of data between Arduino Yun and user application. This application was
created for (currently) most widespread mobile operating system-Android.
Ijaems apr-2016-1 Multibiometric Authentication System Processed by the Use o...INFOGAIN PUBLICATION
The present day authentication system is mostly uni-model i.e having only single authentication method which can be either finger print, iris , palm veins ,etc. Thus these models have to contend with a variety of problems such as absurd or unusual data, non-versatility; un authorized attempts, and huge amount of error rates. Some of these limitations can be reduced or stopped by the use of multimodal biometric systems that integrate the evidence presented by several sources of information. This paper converses a multi biometric based authentication system based on Fusion algorithm using a key. Our work mainly focuses on the fusion algorithm, i.e fusion of finger and palm print out of which the greatest features from the above two traits are taken into account. With minimum possible features the fusion of the both the traits is carried out. Then some key is generated for multi biometric authentication. By processing the test image of a person, the identity of the person is displayed with his/her own image. By the fusion algorithm, it is found that it has less computation time compared to the existing algorithms. By matching results, we validate and authenticate the particular individual.
Attendance Monitoring System of Students Based on Biometric and GPS Tracking ...IJAEMSJORNAL
This paper is a study of a fingerprint recognition system based on minutiae based fingerprint algorithms used in various techniques. This line of track mainly involves extraction of minutiae points from the model fingerprint images and fingerprint matching based on the number of minutiae pairings among to fingerprints. This paper also provides the design method of fingerprint based student attendance with help of GSM. This system ignores the requirement for stationary materials and personnel for keeping of records. The main objective of this project is to develop an embedded system, which is used for security applications. The biometrics technology is rapidly progressing and offers attractive opportunities. In recent years, biometric authentication has grown in popularity as a means of personal identification in college administration systems. The prominent biometric methods that may be used for authentication include fingerprint, palmprint, and handprint, face recognition, speech recognition, dental and eye biometrics. In this paper, a microcontroller based prototype of attendance system using fingerprint sensor and face recognition module is implemented. The tracking module is used here to identify the location of the missing person.
Similar to Generation of Skin Diseases into Synthetic Fingerprints (20)
Francesca Gottschalk - How can education support child empowerment.pptxEduSkills OECD
Francesca Gottschalk from the OECD’s Centre for Educational Research and Innovation presents at the Ask an Expert Webinar: How can education support child empowerment?
June 3, 2024 Anti-Semitism Letter Sent to MIT President Kornbluth and MIT Cor...Levi Shapiro
Letter from the Congress of the United States regarding Anti-Semitism sent June 3rd to MIT President Sally Kornbluth, MIT Corp Chair, Mark Gorenberg
Dear Dr. Kornbluth and Mr. Gorenberg,
The US House of Representatives is deeply concerned by ongoing and pervasive acts of antisemitic
harassment and intimidation at the Massachusetts Institute of Technology (MIT). Failing to act decisively to ensure a safe learning environment for all students would be a grave dereliction of your responsibilities as President of MIT and Chair of the MIT Corporation.
This Congress will not stand idly by and allow an environment hostile to Jewish students to persist. The House believes that your institution is in violation of Title VI of the Civil Rights Act, and the inability or
unwillingness to rectify this violation through action requires accountability.
Postsecondary education is a unique opportunity for students to learn and have their ideas and beliefs challenged. However, universities receiving hundreds of millions of federal funds annually have denied
students that opportunity and have been hijacked to become venues for the promotion of terrorism, antisemitic harassment and intimidation, unlawful encampments, and in some cases, assaults and riots.
The House of Representatives will not countenance the use of federal funds to indoctrinate students into hateful, antisemitic, anti-American supporters of terrorism. Investigations into campus antisemitism by the Committee on Education and the Workforce and the Committee on Ways and Means have been expanded into a Congress-wide probe across all relevant jurisdictions to address this national crisis. The undersigned Committees will conduct oversight into the use of federal funds at MIT and its learning environment under authorities granted to each Committee.
• The Committee on Education and the Workforce has been investigating your institution since December 7, 2023. The Committee has broad jurisdiction over postsecondary education, including its compliance with Title VI of the Civil Rights Act, campus safety concerns over disruptions to the learning environment, and the awarding of federal student aid under the Higher Education Act.
• The Committee on Oversight and Accountability is investigating the sources of funding and other support flowing to groups espousing pro-Hamas propaganda and engaged in antisemitic harassment and intimidation of students. The Committee on Oversight and Accountability is the principal oversight committee of the US House of Representatives and has broad authority to investigate “any matter” at “any time” under House Rule X.
• The Committee on Ways and Means has been investigating several universities since November 15, 2023, when the Committee held a hearing entitled From Ivory Towers to Dark Corners: Investigating the Nexus Between Antisemitism, Tax-Exempt Universities, and Terror Financing. The Committee followed the hearing with letters to those institutions on January 10, 202
Macroeconomics- Movie Location
This will be used as part of your Personal Professional Portfolio once graded.
Objective:
Prepare a presentation or a paper using research, basic comparative analysis, data organization and application of economic information. You will make an informed assessment of an economic climate outside of the United States to accomplish an entertainment industry objective.
Introduction to AI for Nonprofits with Tapp NetworkTechSoup
Dive into the world of AI! Experts Jon Hill and Tareq Monaur will guide you through AI's role in enhancing nonprofit websites and basic marketing strategies, making it easy to understand and apply.
Acetabularia Information For Class 9 .docxvaibhavrinwa19
Acetabularia acetabulum is a single-celled green alga that in its vegetative state is morphologically differentiated into a basal rhizoid and an axially elongated stalk, which bears whorls of branching hairs. The single diploid nucleus resides in the rhizoid.
Normal Labour/ Stages of Labour/ Mechanism of LabourWasim Ak
Normal labor is also termed spontaneous labor, defined as the natural physiological process through which the fetus, placenta, and membranes are expelled from the uterus through the birth canal at term (37 to 42 weeks
Read| The latest issue of The Challenger is here! We are thrilled to announce that our school paper has qualified for the NATIONAL SCHOOLS PRESS CONFERENCE (NSPC) 2024. Thank you for your unwavering support and trust. Dive into the stories that made us stand out!
2024.06.01 Introducing a competency framework for languag learning materials ...Sandy Millin
http://sandymillin.wordpress.com/iateflwebinar2024
Published classroom materials form the basis of syllabuses, drive teacher professional development, and have a potentially huge influence on learners, teachers and education systems. All teachers also create their own materials, whether a few sentences on a blackboard, a highly-structured fully-realised online course, or anything in between. Despite this, the knowledge and skills needed to create effective language learning materials are rarely part of teacher training, and are mostly learnt by trial and error.
Knowledge and skills frameworks, generally called competency frameworks, for ELT teachers, trainers and managers have existed for a few years now. However, until I created one for my MA dissertation, there wasn’t one drawing together what we need to know and do to be able to effectively produce language learning materials.
This webinar will introduce you to my framework, highlighting the key competencies I identified from my research. It will also show how anybody involved in language teaching (any language, not just English!), teacher training, managing schools or developing language learning materials can benefit from using the framework.
The French Revolution, which began in 1789, was a period of radical social and political upheaval in France. It marked the decline of absolute monarchies, the rise of secular and democratic republics, and the eventual rise of Napoleon Bonaparte. This revolutionary period is crucial in understanding the transition from feudalism to modernity in Europe.
For more information, visit-www.vavaclasses.com
Digital Tools and AI for Teaching Learning and Research
Generation of Skin Diseases into Synthetic Fingerprints
1. Milan Barta & Martin Drahansky
International Journal of Image Processing (IJIP), Volume (10) : Issue (5) : 2016 229
Generation of Skin Diseases into Synthetic Fingerprints
Milan Barta milanbarta@gmail.com
Faculty of Information Technology
Brno University of Technology
Bozetechova 2, 612 66 Brno, Czech Republic
Martin Drahansky drahan@fit.vutbr.cz
Faculty of Information Technology
Brno University of Technology
Bozetechova 2, 612 66 Brno, Czech Republic
Abstract
People suffering from skin diseases on their fingers have often problems with fingerprint biometric
recognition. Therefore, it is necessary to train the recognition algorithms using databases of
fingerprints acquired from such affected patients. However, such databases are hard to obtain.
Therefore, we propose an algorithm of synthetic fingerprints modification, so that they appear as
fingerprints affected by selected skin diseases. Two most common diseases have been chosen -
warts and atopic dermatitis.
The proposed solution has been tested using NEUROtechnology VeriFinger and NIST NFIQ 2.
The experimental results confirmed that the proposed solution modifies synthetic fingerprints in
such a way that the recognition score of VeriFinger drops significantly and several NFIQ 2 quality
features report a decrease in quality of the fingerprints.
Keywords: Synthetic Fingerprint, SFinGe, Skin Disease, Wart, Atopic Dermatitis.
1. INTRODUCTION
Fingerprint recognition is one of the most often used biometric technologies all over the world.
Nowadays, users find it even on their mobile phones. It has been designed so that everyone can
use it. However, a problem rises in case of people with some kind of skin disease, disallowing
them to use fingerprint scanning technology in order to authenticate themselves.
Skin diseases represent an important issue in fingerprint recognition. Unfortunately, this problem
is often not taken into consideration when designing such a device. Although precise numbers
are hard to calculate, it is estimated that 20–25% of people suffer from some kind of skin disease
[1]. With such a large number of potential users of fingerprint technology, scanning technologies
should be ready to deal with certain effects that the diseases have on the fingerprint ridge
structure.
This requirement leads to a larger demand on testing of recognition algorithms. However, large
databases of fingerprints are required in order to perform such tests. It is also a very time-
consuming task, as many people need to participate in order to acquire enough samples. In order
to overcome these problems, a database of synthetic fingerprints can be used. With already
existing tools, a large quantity of synthetic fingerprints can be generated in a short time. However,
fingerprint images produced by synthetic fingerprint generating tools are often too perfect to be
used for any meaningful testing of existing algorithms.
Therefore, we aim to alter the generated synthetic fingerprints, so that they appear as fingerprints
of people suffering from various selected skin diseases. Using a database of altered fingerprint
images, existing algorithms might be better prepared for users that suffer from such diseases.
2. Milan Barta & Martin Drahansky
International Journal of Image Processing (IJIP), Volume (10) : Issue (5) : 2016 230
The rest of the paper is organized as follows: The following section 2 introduces synthetic
fingerprint generation methods. Selected diseases are described and fingerprint images analysed
in section 3. In section 4, a method of fingerprint-affecting disease simulation is proposed and the
results are presented. Section 5 deals with evaluation and verification of changes implemented
into synthetic fingerprint and finally section 6 concludes the results achieved.
2. SYNTHETIC FINGERPRINT
With the ever-growing progress of fingerprint recognition systems adoption in many different
areas, methodical and accurate performance evaluations of fingerprint recognition algorithms are
needed. The evaluation is usually based on their recognition accuracy on test data. The
evaluation process consists of three main steps [2]:
1. fingerprint images acquisition,
2. features extraction and matching in order to generate match scores,
3. match scores analysis in order to compute error rates.
Due to very small error rates to be estimated, a reliable evaluation method requires large
databases of fingerprints. However, collection of a large database of fingerprints is [3]:
1. expensive in terms of money,
2. time-consuming,
3. problematic due to privacy legislation that applies to biometric features.
Synthetic fingerprint database has been used for example at a performance evaluation event
FVC2006 [4] (Fingerprint Verification Competition 2006) along with three other databases
containing real fingerprints. [5]
In order to overcome the difficult and time-consuming process of fingerprint collection, synthetic
fingerprint generators can be used to produce large databases at very low cost in terms of money
and time. For our solution, we have chosen a well-established synthetic fingerprint generator
SFinGe
1
.
2.1 SFinGe
One of the first approaches to realistic synthetic fingerprint image generation has been introduced
by R. Cappelli from the University of Bologna in 2004 [3]. The generated synthetic fingerprints
emulate real images acquired with on-line sensors such as capacitive or optical ones.
FIGURE 1: Basic steps of SFinGe method [3].
1
http://biolab.csr.unibo.it/sfinge.html
3. Milan Barta & Martin Drahansky
International Journal of Image Processing (IJIP), Volume (10) : Issue (5) : 2016 231
The basic steps to create a synthetic fingerprint image are [3]: generate a fingerprint shape, a
directional map and density map separately and combine the three features afterwards. The
process of fingerprint generation process can be seen in figure 1. SFinGe method uses a master-
fingerprint in order to derive several synthetic images of the same fingerprint.
SFinGe method is a mature synthetic fingerprint algorithm with a large variety of settings that can
be modified. It produces high-quality images of realistic fingerprints. Authors provide two versions
of the software tool, a free version limited to generating only one fingerprint at a time, and a full
version capable of generating whole databases of hundreds of fingerprints.
3. DISEASE-AFFECTED FINGERPRINTS
Although it is well-known that fingerprints do not change in time, images of the same captured
finger can vary a lot over a period of time due to many factors. These include injuries and bruises,
peeling of the skin on the finger, current dryness of the skin, and also developing a skin disease
affecting person’s skin on fingers. [6]
Skin diseases represent an important factor of fingerprint acquirement process. However, it is
often left out of consideration. The fact that this subject is ignored is supported by missing
research in this area of fingerprint biometrics. Only two recent academic works conducted by
Drahansky et al. [1] and Lee et al. [7] explore the effects of skin diseases on the fingerprint
acquisition process. The work of Lee et al. concentrates on patients with hand dermatitis while
the work of Drahansky et al. considers several various diseases and at the same time presents a
way of enhancing the disease-affected fingerprint image to improve the success rate of enrolment
and matching.
The following subsections deal with two selected most common skin diseases affecting fingers —
warts and atopic dermatitis. First, the disease is described briefly and then an analysis of
fingerprints affected by the disease is conducted.
3.1 Warts
Warts are caused by human papillomavirus (HPV) which belongs to a group of papovaviruses.
There are more than a hundred of types of HPV and gene sequences of HPVs throughout the
world are similar. Most of them cause specific types of warts and favour certain anatomic
locations, such as plantar warts, common warts, genital warts, and so on. [8]
Primarily, our work focuses on common warts which are the most-spread variant of warts
(affecting approximately 10% of the population [9]). They usually cause frustration to the patient.
Social activities can be affected, lesions can be uncomfortable or bleed, and treatment is often
painful and frustratingly ineffective [10].
FIGURE 2: Common warts on hands and fingers [8].
Common warts are usually located on the hands, favouring the fingers and palms (see figure 2).
Fissuring may lead to bleeding and tenderness. Lesions range in size from pinpoint to more than
1 cm, most averaging about 5 mm. They grow in size for weeks to months and are usually
present as elevated, rounded papules with a rough, grayish surface. In some instances, a single
wart (mother wart) appears and grows slowly for a long time, and then suddenly many new warts
4. Milan Barta & Martin Drahansky
International Journal of Image Processing (IJIP), Volume (10) : Issue (5) : 2016 232
erupt. On the surface of the wart, tiny black dots may be visible, representing thrombosed, dilated
capillaries. Warts do not have fingerprint folds, as opposed to calluses, in which these lines are
accentuated. [8]
3.2 Warts-affected Fingerprint Analysis
Let us now compare three different fingerprint images affected by warts (figure 3) that have been
captured using a single sensor, Sagem MSO 300.
Fig. 3a shows a fingerprint with clean ridge structure except for the part where the wart is located.
The wart is located near the right border of the image and on the image it is represented by a
white circle-shaped object with several black dots irregularly spread over its surface. When the
wart is relatively small, usually it contains little or no black dots at all.
In Fig. 3b, a single large wart is located near the core of the whorl. Black dots on top represent
the hard and scaly skin of the wart. The wart is irregularly shaped and its border is well-defined.
The ridge structure around the wart is mildly deformed and the ridges are compressed. However,
except for the close surroundings of the wart, the ridge structure of the fingerprint is unaffected.
Fig. 3c shows a fingerprint that was affected by warts in a large area of its surface. There are at
least three large white oval-like objects with irregular borders near each other. The warts affect
the ridge structure near their edges similarly as the wart described previously.
FIGURE 3 (a,b,c): Different fingerprints affected by warts acquired by Sagem MSO 300.
From the available subset of STRaDe
2
research group fingerprint database, it has been found
that the size of warts on fingers varies from very small ones to ones as large as half of the
hypothetical radius of the fingerprint. The location of warts on fingerprint is random and often a
single wart often produces other so-called satellite warts in its close surroundings.
3.3 Atopic Dermatitis
Atopic dermatitis (AD) [8] is a chronic, inflammatory skin disease that is characterized by pruritus
and a chronic course of exacerbations and remissions. The prevalence of AD increased
dramatically in the last half of the twentieth century, becoming a severe health problem in many
countries. Rates of AD are around 15–20% worldwide with up to 20% of children affected by the
disease [11].
The skin, in general, is dry and somewhat erythematous. Lichenification and prurigo-like papules
are common. Papular lesions tend to be dry, slightly elevated, and flat-topped. They are nearly
always excoriated and often coalesce to form plaques. [8]
The hands, including the wrists, are frequently involved in adults, and hand dermatitis is a
common problem for adults with a history of AD. Hand eczema (figure 4) is the most common
occupational skin condition, accounting for more than 80% of all occupational dermatitis. [8]
2
http://strade-fs.fit.vutbr.cz/cms/en/
5. Milan Barta & Martin Drahansky
International Journal of Image Processing (IJIP), Volume (10) : Issue (5) : 2016 233
FIGURE 4: Hand eczema [8].
3.4 Atopic Dermatitis-affected Fingerprint Analysis
Let us now analyse three different fingerprint images affected by atopic dermatitis (see figure 5)
that were captured using Sagem MSO 300 sensor.
The first figure 5a shows clearly wide and long white lines running throughout the whole
fingerprint. The lines are mostly horizontally oriented. The finger is dry and ridge structure is in
some areas of the fingerprint image less visible. On the other hand, other parts of the fingerprint
show unusually dark areas with a damaged ridge structure.
Figure 5b is similar in the structure of the abnormal white lines to the previously described one.
The lines run predominantly in horizontal direction with their length as large as the width of the
fingerprint. Other thinner and shorter white lines can be seen in both, horizontal and vertical
directions. The ridge structure is clearer than on the previous fingerprint image, however the
patches are present as well.
Fingerprint on figure 5c contains many large white-only patches with no ridge structure
whatsoever. In the centre of the image, there is a wide line running from the bottom-left corner
through the centre of the fingerprint to the upper-right corner of the image. Other thinner lines can
be seen in the left half of the fingerprint. As the ridge structure is mostly strongly damaged, this
fingerprint image can hardly be used in an authentication system.
FIGURE 5 (a,b,c): A set of fingerprints affected by atopic eczema acquired by Sagem MSO 300.
To sum up, the two types of damage by atopic dermatitis are abnormal white lines and light and
dark patches. According to Lee et al., the patches represent dystrophy of the skin and the median
percentage of the surface area of dystrophy in their study was 22.80% [7].
The abnormal white lines usually run in horizontal or vertical direction and their length ranges
from very short up to lines running throughout the whole fingerprint. According to the study of Lee
et al., the median number of white lines per fingerprint was 12 and short horizontal lines prevailed
(with occurrence in 73.0%), followed by short vertical lines (56.5%), long horizontal lines (52.5%),
and long vertical lines (18.0%) [7].
6. Milan Barta & Martin Drahansky
International Journal of Image Processing (IJIP), Volume (10) : Issue (5) : 2016 234
4. FINGERPRINT DISEASE GENERATOR
In the following subsections 4.1 and 4.2, a design of methods for synthetic fingerprint modification
is described in detail for each of the selected diseases. Finally, resulting fingerprint images are
presented in subsection 4.3.
4.1 Warts-affected Fingerprint Generation
Based on the analysis of existing fingerprints affected with warts, design of a method for
generating similar synthetic fingerprint images is proposed in this section. The algorithm consists
of the following steps:
1. localise the fingerprint area on the image;
2. determine the wart size and locate its centre point in the fingerprint;
3. draw the wart into an image buffer:
(a) create an empty image buffer;
(b) generate a number of small circles around the centre point of the wart;
(c) draw the generated circles into the buffer;
(d) draw dark dots inside the wart;
(e) determine the final colour of each wart pixel;
(f) blur the wart in the buffer.
4. draw the wart from the image buffer into the fingerprint image;
5. generate possible secondary warts.
In order to generate warts into the synthetic fingerprint image, the fingerprint has to be localised
in the input image first. This is done in step 1. First, an adaptive thresholding is applied in order to
clearly separate the fingerprint structure from background. Then the image is blurred so that the
fingerprint ridges connect and contours can be localised in the image. The contour of the largest
area is then selected as the fingerprint contour (see figure 6). This contour then defines the
border of the fingerprint.
FIGURE 6: Fingerprint area localisation (a) original fingerprint, b) blurring applied, c) fingerprint contour).
In step 2, a centre of the new wart is localised. The point coordinates are randomly generated
and are used only if they comply with the requirements (location inside of the fingerprint, minimal
distance from the fingerprint border). Also in this step, the size of the generated wart is randomly
determined within boundaries set.
FIGURE 7: Wart drawing (a) wart drawn in distinctive colours, b) dots added, c) drawing in final colours).
7. Milan Barta & Martin Drahansky
International Journal of Image Processing (IJIP), Volume (10) : Issue (5) : 2016 235
Each new wart is first drawn into its own image buffer as not to interfere with the rest of the
fingerprint. This is done in step 3 (see figure 7). First, an empty buffer of a size large enough for
the new wart to fit in is created (step 3a). In the following step 3b, a number of small circles of
varying radius is generated with their centres being distributed with exponentially large distance
from the wart centre point. In the next step 3c, the previously generated circles are drawn into the
buffer in distinctive colour (e.g. red or green) with their border drawn in a different distinctive
colour (figure 7a). The final drawing step is step 3d in which the dots with randomly generated
coordinates are drawn onto the warts surface. The dots are drawn with the same colour as the
border of the small circles in the previous step (figure 7b).
FIGURE 8 (a,b): Wart drawing (steps 4 and 5).
With the wart shape drawn in the buffer, the algorithm proceeds with step 3e where the final
colour of each pixel of the wart is determined (figure 7c). Depending on if the pixel is drawn by
colour for border or the colour of the inside of the wart, the colour of the neighbouring pixels in the
original fingerprint image is acquired (in this case dark pixels for border and light ones for the
inside of the wart). The final pixel colour is then determined by one of the two following methods.
First method picks random neighbouring pixel and copies its colour. The second method
computes the mean colour of all the neighbouring pixels and then the mean colour is computed
and applied to the pixel. Afterwards the buffer image is blurred slightly in step 3f in order to better
fit into the original fingerprint image.
Finally, in step 4 the buffer is drawn into the original fingerprint image taking in consideration the
transparency of the pixels in the buffer and blending them into the original image appropriately
(figure 8a).
Eventually, secondary warts are drawn into the fingerprint if required in step 5 following the same
steps of the algorithm as for the main wart (figure 8b). The only difference is an added
requirement not to overdraw already existing warts in the image.
4.2 Atopic Dermatitis-affected Fingerprint Generation
Based on the analysis of existing fingerprints affected with atopic eczema, design of a method for
generating similarly damaged synthetic fingerprint images is proposed in this section. The
algorithm consists of the following steps:
1. localise the fingerprint area on the image;
2. create an empty image buffer;
3. draw eczema patches into the buffer:
(a) determine the centre and size of the patch;
(b) draw the patch of determined type (light, dark).
4. determine the final colour of each pixel of the patches;
5. blur the patches in image buffer;
6. draw eczema white lines into the buffer:
(a) determine the starting point, direction, and length of the line;
(b) generate line points in given direction and length;
(c) interpolate the generated line points;
8. Milan Barta & Martin Drahansky
International Journal of Image Processing (IJIP), Volume (10) : Issue (5) : 2016 236
(d) draw the lines in determined thickness.
7. blur the lines in image buffer;
8. draw the buffer into the fingerprint image.
Step 1 of the algorithm is identical to the first step of the algorithm for generating warts. The result
of this step is a contour of the fingerprint on the input image. Knowing precisely where the
fingerprint is located in the image is necessary in order to draw onto the fingerprint area only and
not outside of it.
In step 2, a new image buffer is created. The size of the buffer is the same as the size of the input
image. The patches and white lines shall be drawn into the buffer separately as not to interfere
with the original image.
First, the light and dark patches are drawn into the buffer in step 3. The number of patches is
generated randomly within set boundary values. Afterwards the type (light, dark), size, and a
centre point for each patch is determined in step 3a. If the centre point lies within the fingerprint
boundaries, the algorithm proceeds to drawing the patch into the buffer. This is step 3b (see
figure 9a). In this step, pixels in distance generated with exponential distribution from the centre
point are drawn into the buffer in a distinctive colour (e.g. red or blue). It is randomly chosen if the
pixels of the patch will later be in light colour or dark one.
When all the patches are drawn into the buffer in a distinctive colour, the final colour of each pixel
is determined in step 4 (see figure 9b). First, the neighbouring pixels of each pixel in patch are
collected from the input image. Then, based upon the selected algorithm, the final pixel’s colour is
either one of a randomly chosen neighbouring pixel or mean of all its neighbour’s colours. After
this, the patches in buffer are blurred in step 5 (see figure 9c).
FIGURE 9 (a,b,c): Eczema patches drawing (steps 3, 4, and 5).
The second significant part of the algorithm takes place in step 6 where white lines are drawn into
the buffer. Each part of the process is described in the following paragraphs.
In step 6a, parameters of each line are determined. The length of the line is determined within set
boundary values and the line direction (either vertical or horizontal) is set. The starting point for
line generation is found using random coordinates generation. The starting point must be
sufficiently far from all other starting points of all other lines of the same type.
Line points are generated in step 6b. Beginning with the starting point, other leading points are
generated based on the length of the line, the direction of the line, and a random generated angle
within a pre-defined range (see figure 10a).
To make the lines look more realistic, in step 6c, the line leading points count is doubled and
spline interpolation of the first order is applied. This makes the line appear less edgy and smooths
it (see figure 10b).
Finally, each line is drawn into the buffer in step 6d (see figure 10c). The thickness of the line is
set and the line is drawn in several steps starting with the whole length drawn in the smallest
9. Milan Barta & Martin Drahansky
International Journal of Image Processing (IJIP), Volume (10) : Issue (5) : 2016 237
thickness. Then the first and last leading points are removed and the line is drawn over with a
higher thickness. This process repeats until the final set thickness is reached. This ensures that
the line’s width decreases towards line ends.
FIGURE 10 (a,b,c): Eczema lines drawing (into buffer) (steps 6b, 6c, and 6d).
In the last two steps, the buffer is once again blurred in step 7 and then in the following step 8,
the buffer is drawn into the original fingerprint image taking in consideration the transparency of
the pixels in the buffer and blending them into the original image appropriately.
4.3 Results
In this subsection, the resulting fingerprint images are presented. Figure 11 represents a selected
set of fingerprint images to show the product of warts simulation. Subfigure 11a represents
original synthetic fingerprint and next to it, on subfigure 11b, there is the same image with disease
marks implemented in the Fingerprint disease simulator.
FIGURE 11: Results of fingerprint warts simulation (a) synthetic fingerprint, b) simulation result).
Similarly, the following figure 12 represents a selected set of fingerprint images to show the
product of atopic eczema simulation. Subfigure 12a represents the original synthetic fingerprint
while the same image with disease marks implemented in the Fingerprint disease simulator can
be seen on subfigure 12b.
FIGURE 12: Results of fingerprint atopic eczema simulation (a) synthetic fingerprint, b) simulation result).
10. Milan Barta & Martin Drahansky
International Journal of Image Processing (IJIP), Volume (10) : Issue (5) : 2016 238
5. EXPERIMENTAL RESULTS
In order to evaluate the output of the Fingerprint disease simulator, the resulting damaged
fingerprints must undergo verification against unmodified synthetic fingerprint images and their
quality has to be measured. Thirteen sets, each containing 250 fingerprint images were
generated with various parameters for evaluation. A detailed description of the datasets can be
found in section 5.1.
Two methods of evaluation were selected. VeriFinger[12] by NEUROtechnology was used for the
fingerprint verification. The output is a matching score of a damaged fingerprint expressed in a
relation to a matching score of the original fingerprint that has been verified against itself. The
results for both warts datasets and eczema datasets can be found in section 5.2.
NFIQ 2.0[13] fingerprint quality assessment tool was used to determine the quality of the
damaged fingerprints compared to the quality of unmodified synthetic fingerprints. The output is a
score based on several quality features computed from the fingerprint image. The results for the
warts and eczema datasets are presented in section 5.3 in detail.
5.1 Description of Datasets
For the purpose of verification and quality assessment of the synthetic fingerprints with disease
marks generated by the Fingerprint disease simulator, a total number of thirteen datasets were
generated, each containing 250 unique fingerprint images affected by warts or atopic eczema.
Each dataset was created with different set of parameters of the algorithm so that it can be
evaluated which parameters affect the quality of the fingerprint the most.
The parameters of warts datasets are presented in table 1. Four datasets of warts-affected
fingerprints were created in total. Warts dataset 1 and 2 are generated with warts size set to 5–
10% of hypothetical radius of the fingerprint, while the size for dataset 3 and 4 is set to 10–15%.
The maximum number of secondary warts (Max. SW cnt) is two for datasets 2 and 4. In case of
datasets 1 and 3, they were disabled completely.
Dataset Wart size [%] Max. SW cnt
Warts 1 5–10 0
Warts 2 5–10 2
Warts 3 10–15 0
Warts 4 10–15 2
TABLE 1: Warts datasets parameters.
The number of eczema datasets generated for verification and quality assessment is nine and the
parameters for each dataset can be found in table 2. The parameters include range of horizontal
white lines count (HL cnt), vertical lines count (VL cnt), line length represented by percentage of
hypothetical fingerprint radius (L. length), line width (L. width), patches count (P. cnt) and their
size in percent relative to hypothetical fingerprint radius (P. size).
Datasets 1–4 combine various line types and their length, while eczema datasets 5 and 6 add
different line thickness. Datasets 7 and 8 contain different amount of patches with no lines and
the last dataset 9 combines the vertical and horizontal lines with patches together.
Dataset HL cnt VL cnt L. length [%] L. width P. cnt P. size [%]
Eczema 1 4–12 0 50–100 8 0 0
Eczema 2 4–12 0 100–200 8 0 0
Eczema 3 0 2–6 50–100 8 0 0
Eczema 4 0 2–6 100–200 8 0 0
11. Milan Barta & Martin Drahansky
International Journal of Image Processing (IJIP), Volume (10) : Issue (5) : 2016 239
Eczema 5 4–12 2–6 50–200 8 0 0
Eczema 6 4–12 2–6 50–200 5 0 0
Eczema 7 0 0 0 0 2–10 20–80
Eczema 8 0 0 0 0 10–20 20–80
Eczema 9 4–12 2–6 50–200 8 4–20 20–80
TABLE 2: Eczema datasets parameters.
5.2 NEUROtechnology VeriFinger
For verification of generated disease-affected fingerprint, VeriFinger, a fingerprint identification
tool by NEUROtechnology, was used.
The tool allows the user to enrol fingerprints into a database and verify or identify fingerprints
against the existing database templates. During the process, a matching score is printed out. This
score was used to evaluate the degree of the damage generated by the Fingerprint disease
simulator into the synthetic fingerprint.
The methodology of the verification process is as follows:
1. enrol the original unmodified synthetic fingerprints into the database;
2. verify the same unmodified synthetic fingerprints and save their scores (this constitutes a
baseline score — 100% score);
3. for each fingerprint from a given testing set, verify the damaged fingerprint against the
original unmodified one and record the matching score;
4. evaluate the percent value of the recorded matching score — the normalized median
value of the matching score;
5. calculate the median value for each set.
5.2.1 Warts
The results of warts-affected fingerprint verification are presented in table 3. The table contains
the median of the set matching score and its normalized value. Graphical representation of the
result values can be found on figure 13.
Dataset Original Warts 1 Warts 2 Warts 3 Warts 4
Median 1116.50 1037.50 1029.00 944.50 918.00
Norm. median [%] 100.00 93.47 92.74 85.80 83.42
TABLE 3: VeriFinger score: warts datasets.
From the data in table 3 it is clear that presence of warts in a fingerprint negatively affects the
value of the fingerprint matching score. The Warts 1 and Warts 2 datasets are generated with
only relatively small warts in size, while the other two datasets, Warts 3 and Warts 4 contain
noticeably larger warts. Because of this, the score dropped by almost 10% when warts were
relatively small (size parameters set to 5–10%) and by about 15% for larger warts (size
parameter set to 10–15%).
Also the presence of secondary warts in datasets Warts 2 and Warts 4 decreases the final score
considerably (by 0.73% in case of Warts 1 compared to Warts 2; and by 2.38% for Warts 3 and
Warts 4 datasets).
The most probable reason for this is that the wart in a fingerprint creates new minutiae while
covering those present before the disease has been generated into the fingerprint. This fact
results in lower matching score as some ridge structures cannot be properly matched by the
algorithm any more.
12. Milan Barta & Martin Drahansky
International Journal of Image Processing (IJIP), Volume (10) : Issue (5) : 2016 240
FIGURE 13: VeriFinger score: warts datasets.
5.2.2 Atopic Eczema
The results of atopic eczema-affected fingerprint verification are presented in table 4 containing
the median of the set matching score as well as its normalized value. Graphical representation of
the results can be found on figure 14.
Dataset Original Ecz. 1 Ecz. 2 Ecz. 3 Ecz. 4
Median 1116.50 836.50 841.50 855.50 857.50
Norm. median [%] 100.00 76.11 75.94 77.15 77.80
Dataset Ecz. 5 Ecz. 6 Ecz. 7 Ecz. 8 Ecz. 9
Median 809.00 869.50 842.00 761.50 706.00
Norm. median [%] 73.79 78.33 76.25 68.20 64.30
TABLE 4: VeriFinger score: eczema datasets.
Judging from the data acquired, when comparing datasets Eczema 1 and Eczema 2 with
datasets Eczema 3 and Eczema 4, it can be said that the type of eczema lines in the fingerprint
has only a little effect on the matching score (approximately 24% decrease for horizontal lines
versus circa 23% decrease). The damage caused by the horizontal and vertical lines is principally
the same. The white lines disrupt the flow of ridges and thus create new false minutiae in places
of line crossing.
The length of the lines has a negligible effect on the final matching score. The difference of
medians of datasets Eczema 1 and Eczema 2 is only 0.17% and in case of datasets Eczema 3
and Eczema 4, the difference is 0.65%.
The small difference might be caused by the fact that a significant part of the line can be
generated outside of the fingerprint. The reason for this is that while the starting point of the line is
13. Milan Barta & Martin Drahansky
International Journal of Image Processing (IJIP), Volume (10) : Issue (5) : 2016 241
generated to be inside of the fingerprint, the direction of the generated line is decided randomly.
Therefore if the starting point is located near the fingerprint border and the direction is determined
to point out of the fingerprint, the generated line might be in fact shorter than expected.
Examining the influence of line thickness on the matching score by comparing datasets Eczema 5
and Eczema 6 shows that thicker lines have a greater damaging effect on the fingerprint. The
median of matching score for lines of maximal thickness 8 is only 73.79%, while the median of
score for lines of maximal thickness 5 is 78.33%. This effect can be explained by the fact that
thinner ridge disruptions might be repaired by the matching algorithm, while thicker lines
discontinue the ridges more effectively.
Comparing the damage of each type of eczema marks (white lines versus patches), the one
causing the greatest damage to the fingerprint are patches. Their effect is most significant when a
large amount of patches is generated into the fingerprint. This demonstrates the comparison of
dataset Eczema 7 (2–10 patches per fingerprint) with median of matching score 76.25% versus
dataset Eczema 8 (10–20 patches per fingerprint) with median of matching score 68.20%. The
patches in fingerprint make the ridge structure effectively less clear and bring more noise into it.
This makes the minutiae recognition process for the matching algorithm harder to do, thus
lowering the matching score.
The most effective damaging results are brought by both types of damage (white lines and
patches) combined. As represented by dataset Eczema 9, the median of matching score declined
to 64.30%.
FIGURE 14: VeriFinger score: eczema datasets.
5.3 NIST NFIQ 2.0
The second method of verification of damaged synthetic fingerprints by the Fingerprint disease
simulator was done with the help of NFIQ 2.0 fingerprint quality assessment tool released by
NIST (National Institute of Standards and Technology).
In order to assess the quality of a fingerprint, NFIQ 2.0 computes a set of quality features from
the input image, and uses them to predict the fingerprint image quality. The output of the
algorithm is a score value in range of [0–100], where 0 represents an image of no utility and 100
is the highest utility value. The NFIQ 2.0 algorithm bases its score computation on fourteen
selected quality features which together constitute the final score for a given input image.
14. Milan Barta & Martin Drahansky
International Journal of Image Processing (IJIP), Volume (10) : Issue (5) : 2016 242
The base score was established for each fingerprint by evaluating the quality of the unmodified
original synthetic fingerprint image. Then for each damaged fingerprint, its score was evaluated.
A median of scores for each dataset was found and expressed in percent in relation to the base
score median.
5.3.1 Warts
The results of warts-affected fingerprint quality assessment can be seen in table 5. The table
contains the median of a set quality score and its normalized value. Graphical representation of
the result values can be found on figure 15.
Dataset Original Warts 1 Warts 2 Warts 3 Warts 4
Median 64.00 66.00 66.00 67.00 67.00
Norm. median [%] 100.00 101.92 101.92 104.20 104.20
TABLE 5: NFIQ2 score: warts datasets.
As data in table 5 shows, the NFIQ2 score actually increased for all testing datasets of warts-
affected fingerprints when compared to the unmodified original synthetic fingerprint dataset.
FIGURE 15: NFIQ2 score: warts datasets.
In order to explain this, the NFIQ2 score calculation has to be taken into account. The value of
the score is based on fourteen various quality features computed from the input fingerprint image.
Because the decision logic of the application is based on trained random forest learning, not all
quality features have the same weight. Detailed information on the weight of each quality feature
can be found in the NFIQ 2.0 documentation
3
(page 29).
3
http://biometrics.nist.gov/cs_links/quality/NFIQ_2/nfiq2_report.pdf
15. Milan Barta & Martin Drahansky
International Journal of Image Processing (IJIP), Volume (10) : Issue (5) : 2016 243
Therefore, it can be assumed that the quality features affected by changes implemented by the
Fingerprint disease simulator do not have weights large enough to influence the final NFIQ2
score. In fact, as in this case, some other quality features might be enhanced by the changes so
that the final score rises above the score of the control set.
Detailed examination of several individual quality feature results can be found in section 5.4.
5.3.2 Atopic Eczema
Quality assessment results of fingerprints affected by atopic eczema can be found in table 6. The
table contains the median of a set quality score as well as its normalized value. Graphical
representation of the result values can be found on figure 16.
Dataset Original Ecz. 1 Ecz. 2 Ecz. 3 Ecz. 4
Median 64.00 65.00 64.00 63.00 64.00
Norm. median [%] 100.00 101.71 100.00 100.00 100.00
Dataset Ecz. 5 Ecz. 6 Ecz. 7 Ecz. 8 Ecz. 9
Median 66.00 64.00 65.00 68.00 68.00
Norm. median [%] 102.03 100.00 101.79 104.67 104.67
TABLE 6: NFIQ2 score: eczema datasets.
Also in the case of eczema datasets, the NFIQ2 quality score is equal or higher than the score of
the control dataset. The reasons were described in the previous section and are also the same in
this case.
FIGURE 16: NFIQ2 score: eczema datasets.
5.4 NIST NFIQ 2.0: Selected Quality Features
In order to explain the higher value of NFIQ2 score for all datasets of damaged fingerprints, while
the exact opposite was expected, the NFIQ2 score computation method has to be studied. As it
was stated above, the score value is calculated from fourteen unique quality features with various
weights. Thus, in order to verify that the changes made by the Fingerprint disease simulator bring
significant damage marks into the fingerprint, let us investigate the scores of several selected
16. Milan Barta & Martin Drahansky
International Journal of Image Processing (IJIP), Volume (10) : Issue (5) : 2016 244
quality features evaluated by the NFIQ 2.0 algorithm. Only quality features that expressed a
noticeable change in the score were selected into the set of examined quality features. Others
stayed rather constant when compared to control dataset of unmodified synthetic fingerprints,
therefore there is no need to study them further.
The NFIQ 2.0 employs a customized version of FingerJet FX OSE minutia extractor
4
for
determining the amount of minutiae detected in the whole image (Minutiae cnt) and an arithmetic
mean of all minutiae quality values. Two different methods for computing the quality of the
minutiae are used.
The first method calculates the quality using an arithmetic mean of pixel values in the input image
(MU M. Quality). The second method of minutiae quality assessment computes the quality as the
Orientation Certainty Level of blocks of pixels centred at the minutia location (OCL M. Quality).
The other quality features presented are ROI Relative Orientation Map Coherence Sum
(OM Coherence) which represents the average coherence values over all image blocks in the
fingerprint ROI.
The last of the studied quality features is the Ridge Valley Uniformity (Uniform Image) feature. It
measures the consistency of the ridge and valley widths. For a finger image with clear ridge and
valley separation it is expected that the ratio remains rather constant and thus the standard
deviation of the ratios is used as an indication of the fingerprint quality.
5.4.1 Warts
The evaluation results of selected quality features of the NFIQ 2.0 algorithm for warts-affected
fingerprints datasets are presented in table 7.
Dataset Original Warts 1 Warts 2 Warts 3 Warts 4
Minutiae cnt 36 38 37 43 45
MU M. Quality [%] 85.00 83.00 84.00 75.00 71.00
OCL M. Quality [%] 77.00 71.00 69.00 60.00 55.00
OM Coherence [%] 75.00 74.00 74.00 73.00 72.00
Uniform Image [%] 53.72 53.49 53.42 53.23 53.06
TABLE 7: NFIQ2: selected quality features: warts datasets.
From the summarized results it can be deduced that changes brought by the Fingerprint disease
simulator generating warts, increase the number of detected minutiae in a fingerprint. This means
that the newly detected minutiae must be false ones. Increasing the size of warts also increases
the number of new false minutiae.
The quality of minutiae measured by both methods previously described also decreases with
increasing size of warts generated. Original quality score values of 85% and 77% go down as low
as to 0.71% and 0.55% for dataset Warts 4 containing generated warts of a large size (10–15%)
and secondary warts generating enabled.
As far as OM Coherence and Uniformity of Image quality features are concerned, their score
values differ insignificantly from the original control dataset.
To sum up, for warts-affected fingerprints, new false minutiae are detected by the minutiae
extractor and their quality is significantly lower than the quality of minutiae found on the original
fingerprint image. Other score values change only negligibly.
4
https://github.com/FingerJetFXOSE/FingerJetFXOSE
17. Milan Barta & Martin Drahansky
International Journal of Image Processing (IJIP), Volume (10) : Issue (5) : 2016 245
5.4.2 Atopic Eczema
The results of selected quality features evaluation of the NFIQ 2.0 algorithm for datasets
containing fingerprints affected by atopic eczema are presented in table 8.
Dataset Original Ecz. 1 Ecz. 2 Ecz. 3 Ecz. 4
Minutiae CNT 36 36 36 36 35
MU M. Quality [%] 85.00 79.00 79.00 82.00 80.00
OCL M. Quality [%] 77.00 75.00 76.00 76.00 76.00
OM Coherence [%] 75.00 75.00 75.00 75.00 75.00
Uniform Image [%] 53.72 49.50 49.66 49.69 49.87
Dataset Ecz. 5 Ecz. 6 Ecz. 7 Ecz. 8 Ecz. 9
Minutiae CNT 36 36 36 37 40
MU M. Quality [%] 78.00 82.00 79.00 71.00 68.00
OCL M. Quality [%] 72.00 76.00 69.00 46.00 43.00
OM Coherence [%] 74.00 75.00 73.00 67.00 67.00
Uniform Image [%] 49.00 49.61 43.20 36.70 37.58
TABLE 8: NFIQ2: selected quality features: eczema datasets.
Judging from the summary of results presented the minutiae count does not change significantly
for eczema-affected fingerprints. The only exception for this rule is the last dataset Eczema 9
which contains fingerprints damaged by all types of available eczema marks. In comparison to
other datasets which keep the same number of detected minutiae as the control dataset, the
median value of minutiae detected in dataset Eczema 9 is 40.
The only considerable quality change of the minutiae of fingerprints affected by eczema can be
observed in Eczema 8 (71% and 46% versus 85% and 77% for the control dataset) and
Eczema 9 datasets (68% and 43% versus 85% and 77% for the control dataset). Other datasets
do not show such a significant change in quality. The most probable reason for this is that the last
two datasets contain a large number of eczema patches generated into the fingerprints. This
decreases the quality of minutiae measured by both of the methods of fingerprint minutiae quality
assessment.
As far as ROI Relative Orientation Map Coherence Sum (OM Coherence) quality feature is
concerned, the only significant change can be seen in Eczema 8 and Eczema 9 datasets (the
score of both is 67%). The score of control dataset is 75% and the other dataset scores range
between 73% and 75%. Again, this is caused by a large number of eczema patches present in
fingerprints contained in Eczema 8 and Eczema 9 datasets.
For the Uniformity of Image quality feature, the score of patches-containing datasets decreases
noticeably while datasets containing fingerprints with eczema lines record only a small decrease
in score value (approximately 4% decrease). Even a small number of patches brings the score
value down to 43.20% for dataset Eczema 7. Larger decrease of the score value can be seen in
Eczema 8 and Eczema 9 datasets (36.70% and 37.58% versus the control dataset value of
53.72%).
All in all, except for the minutiae count, all other quality features are negatively affected by a large
number of eczema patches present in a fingerprint image. On the other hand, eczema lines do
not have such a significant effect on the score value of selected quality features.
18. Milan Barta & Martin Drahansky
International Journal of Image Processing (IJIP), Volume (10) : Issue (5) : 2016 246
6. CONCLUSIONS
The aim of our work was to design an algorithm for modifying a synthetic fingerprint image in a
way similar to the way a real disease would do. In order for the goal to be fulfilled, a number of
subjects concerning fingerprint biometry had to be studied.
From the STRaDe database of disease-affected fingerprints, two most common diseases were
selected: warts and atopic eczema. For each of the disease, further study was carried out. Based
on the samples available from the database, an analysis of selected fingerprint images was
conducted and disease-specific features described. Based on the analysis, design of a method
for the specific disease generation was proposed. The resulting outputs of the proposed solution
were presented.
The simulation results are datasets of synthetic fingerprint images with varying disease-specific
marks on them. In case of warts, typical circle-shaped objects are generated into the fingerprint
with black dots inside. Simulation output images for eczema are represented by various white
lines and colour patches generated into the synthetic fingerprint image.
Conducted experiments were described in chapter 5. In order to verify the damaging effect
caused to the synthetic fingerprints by simulating diseases marks to them, thirteen datasets (four
warts and nine eczema datasets) each containing 250 fingerprint images were generated in total.
Each dataset was created using various parameters so that it could be later examined which
parameters affect the fingerprint recognition process in what way. Two methods were used to
verify the fingerprints and assess their quality: VeriFinger by NEUROtechnology and NFIQ 2.0
algorithm by NIST.
When evaluated with the VeriFinger algorithm, warts datasets showed decrease in the matching
score depending on the generated warts size. The score for the dataset of warts with the size
parameter set to 10–15% and secondary warts generation enabled went as low as 83.42% of the
control dataset score.
Even more significant was the decrease of matching score for datasets of eczema-affected
fingerprints. For datasets with fingerprints containing white eczema lines only, the score dropped
by approximately 23–24% to 76–77% of the control dataset score. It was also found that more
damaging are thicker lines (score of 78.33% for thickness 5 versus 73.79% for line thickness 8).
By far, the most damaging effect on the fingerprint ridge structure have eczema patches. The
matching score of datasets generated with the patches enabled dropped to 68.20% of the control
group score. When patches and eczema lines were combined, the matching score recorded went
down to only 64.30%.
As far as the quality assessment is concerned, the final score of the NFIQ 2.0 quality assessment
tool did not reflect changes made by the Fingerprint disease simulator as expected. The reason
for this is that the final score is computed based on fourteen various quality features, each of
them having various weights in the algorithm. Therefore, several specific quality features have
been selected instead and their results are discussed further.
For warts datasets, there has been an increase in minutiae count compared to the control dataset
(up to 45 versus 36 detected minutiae per fingerprint) meaning that the disease marks simulated
by the Fingerprint disease simulator create new false minutiae in the fingerprint. Along with that,
the quality of minutiae decreased significantly, measured by two different methods (by up to 14
percent points and by up to 22 percent points). Other quality features changed only slightly.
Eczema datasets showed almost no new false minutiae creation (except for the last dataset
combining the eczema patches and white lines, where the minutiae count rose up to 40
compared to 36 minutiae per fingerprint for the control dataset). However, the quality of the
minutiae dropped significantly, mainly in datasets with large number of generated patches. The
quality dropped by up to 9 percent points by one measuring and by up to 32 percent points
measured by the second method. Also the other two quality features (ROI Relative Orientation
19. Milan Barta & Martin Drahansky
International Journal of Image Processing (IJIP), Volume (10) : Issue (5) : 2016 247
Map Coherence Sum and Uniformity of Image) dropped significantly for the datasets with large
number of generated patches (from 75% down to 67% for the OM Coherence and from 53.72%
down to 37.58% for the Uniformity of Image).
All in all, the changes to the synthetic fingerprints made by the Fingerprint disease generator are
well-measurable. In case of warts, the main aspect influencing the damage extent is the size of
the generated warts and the amount of them. As far as atopic eczema is concerned, the main
influencing aspect is the count of eczema patches, while eczema white lines have a minor effect.
The impact of this research is that recognition algorithms will no longer perform significantly
worse for patients suffering from various skin diseases compared to unimpaired users thanks to
the introduction of our method. The method proposes a way of generating large databases of
synthetic fingerprints with marks of skin diseases such as warts or atopic eczema. These
databases can then be used to train fingerprint recognition algorithms to perform better for users
with such skin diseases.
7. FUTURE RESEARCH
For this research, two most common skin diseases have been chosen to simulate: warts and
atopic eczema. The Fingerprint disease simulator was designed so that it can be extended with
new modules simulating other fingerprint diseases as well. For a good understanding of the
studied diseases, further ongoing cooperation with dermatologists will be required.
Another way of utilising the research outcomes is in designing and testing of a tool for automatic
recognition of skin disease from fingerprint image.
All in all, incorporating the proposed method, matching algorithms could profit from the virtually
unlimited amount of fingerprints generated in order to adapt themselves and improve the
fingerprint recognition of fingerprints with various diseases.
8. ACKNOWLEDGEMENTS
This work was supported by The Ministry of Education, Youth and Sports of the Czech Republic
from the National Programme of Sustainability (NPU II); project IT4Innovations excellence in
science – LQ1602.
9. REFERENCES
[1] M. Drahansky, M. Dolezel, J. Urbanek, E. Brezinova and T. Kim. “Influence of skin diseases
on fingerprint recognition”. Journal of Biomedicine and Biotechnology, vol. 2012, 14 pages,
2012.
[2] Q. Zhao, A. K. Jain, N. G. Paulter and M. Taylor. “Fingerprint image synthesis based on
statistical feature models,” in Proc. BTAS, 2012.
[3] R. Cappelli. “SFinGe: an approach to synthetic fingerprint generation,” in Proc. Int. Workshop
on Biom. Technol., 2004, pp. 147–154.
[4] R. Cappelli, M. Ferrara, A. Franco and D. Maltoni. “Fingerprint verification competition 2006.”
Biom. Technol. Today, vol. 15, pp. 7–9, 2007.
[5] R. Cappelli, D. Maio and D. Maltoni. “Synthetic fingerprint-database generation,” in Proc.16th
International Conference on Pattern Recognition, vol. 3, 2002, pp. 744–747.
[6] K. Jain, S. Prabhakar and A. Ross. “Fingerprint matching: Data acquisition and performance
evaluation.” Dep. of Comput. Sci., Mich. State Univ., East Lansing, Tech. Rep. MSU-CPS-
99–14.
20. Milan Barta & Martin Drahansky
International Journal of Image Processing (IJIP), Volume (10) : Issue (5) : 2016 248
[7] Lee, C. Chang, A. Johar, O. Puwira and B. Roshidah. “Fingerprint changes and verification
failure among patients with hand dermatitis”. JAMA Dermatol, vol. 149, pp. 294–299, 2013.
[8] W. D. James, T. G. Berger and D. M. Elston. Andrew’s Diseases of the Skin: Clinical
Dermatology, 11th Edition. Elsevier Health Sciences, 2011.
[9] U. Schellhaas, W. Gerber, S. Hammes and H. M. Ockenfels. “Pulsed dye laser treatment is
effective in the treatment of recalcitrant viral warts”. Dermatol. surg., vol. 34, pp. 67–72, 2008.
[10]A. Ciconte, J. Campbell, S. Tabrizi, S. Garland and R. Marks. “Warts are not merely
blemishes on the skin: A study on the morbidity associated with having viral cutaneous
warts”. Australas. j. of dermatol., vol. 44, pp. 169–173, 2003.
[11]J. Ring, A. Alomar, T. Bieber, M. Deleuran, A. Fink-Wagner, C. Gelmetti, U. Gieler,
J. Lipozencic, T. Luger, et al. “Guidelines for treatment of atopic eczema (atopic dermatitis)
part I”. J of the Eur. Acad. of Dermatol. and Venereol., vol. 26, pp. 1045–1060, 2012.
[12]“NEUROtechnology, Verifinger SDK”. “Internet:
http://www.neurotechnology.com/verifinger.html, [Sept., 15, 2016].
[13]National Institute of Standards and Technology. “Development of NFIQ 2.0” Internet:
http://www.nist.gov/itl/iad/ig/development_nfiq_2.cfm, [Sept., 15, 2016]