This document discusses using thumb and iris detection for security purposes. Thumb detection works by scanning finger ridges and valleys, preprocessing the image, extracting global and local structures, and matching to a database. Iris detection captures images, segments the iris, extracts raw data, and matches patterns to codes in a database. The system would fuse iris and thumb models, use distance classifiers to determine similarities, and provide a recognition index. The goal is to recognize identity through iris and thumbprint patterns for identification and authentication. Hardware would include an ARM board or Raspberry Pi, while software would use Python with libraries for iris and thumbprint detection.
3. Thumb Detection
• The finger’s ridges and valleys are scanned.
• Acquisition : Image is obtained from hardware or a file.
• Pre-processing: Noise reduction, thinning, image
enhancing and error correction.
• Extraction: Global and local structures are found.
• Matching: Fingerprints are compared against a database.
4. Iris Detection
• Capturing the image using hardware.
• Segment and localize iris from the acquired image.
• Extraction is where raw data is divided and reduced to
manageable groups.
• Matching iris pattern with help of iris codes from
database.
5. Fusion of Iris and Thumb model
• Fusion:- Combining data from multiple sources.
• Distance classifier:- Tells similarities and dissimilarities in two
datas.
• Recognition index:- Measure of accuracy and effectiveness.
• Parameters:- These tells about different needs for an input
data.
6. Objective of the project
Goal :- To recognize human identity through the
textural characteristics of one’s iris and muscular thumb
print patterns to provide safe and convenient
identification and authentication with a human touch.
7. Scope of the project
• In Indian biometric market, market value was expected to grow to $823.46 million in 2014. In 2020 it was expected to
reach $2.6 billion.
• CAGR – Compound Annual Growth Rate
8. • Indian has started to witness biometric devices being deployed in several fields like local security , immigration sectors
and more.
• Advancements in other fields like defense, consumer electronics, transport & logistics is also made.
9. Details of Hardware and Software Used
Hardware
Either ARM board or Raspberry Pi will be used to dump the code.
Software
Language used – Python
Libraries used :–
For iris detection
• NumPy
• skimage.util
• Matplotlib.pyplot
• cv2
For thumbprint detection
• NumPy
• Os
• cv2
11. Contribution of the project
• Biometric systems make access control more convenient to
users eliminating need for keys, cards, etc.
• Reduced fraud and theft as biometric data cannot be easily
stolen or replicated especially iris.
• Improved healthcare as it can be used to ensure that the
patient receives appropriate treatment. This reduces
medical errors.
• Enhanced security compared to traditional methods like
passwords or PINs. Prevents unauthorized access.
12. Conclusion
• Multi biometric systems offer several advantages over single
biometric systems including increased security, accuracy and
reduced vulnerability to spoofing attacks.
• It provides a better performance in real world scenarios.
• As technology continues to advance, biometric systems will
become more sophisticated and accurate. We may also see
emergence of new biometric modalities such as gait (Features of
human motion are automatically obtained) recognition which could
further improve security and accuracy of these systems.
13. Presented by :-
Afreen Roshan D (4NI20EC005)
Bhavana K S (4NI20EC019)
Bhavana M S (4NI20EC020)
Deepthi M N (4NI20EC033)