Introduction to IEEE STANDARDS and its different types.pptx
smart attendance system using signature verification 1.pptx
1. Smart Attendance Management and Analysis with
Signature Verification
Name - Chinmaya Kumar
Mohanty
Regd No. - 2002070126
Dept. - Electronics and
Telecommunication
1
2. Abstract
In many educational institutions, sufficient number of class attendance
is a requirement for earning a regular grade in a course. Automatic
signature verification is an active research area from both scientific
and commercial points of view as signatures are the most legally and
socially acceptable means of identification and authorization of an
individual.
This presentation proposes a novel automatic lecture attendance
verification system based on unsupervised learning. Here, lecture
attendance verification is addressed as an offline signature verification
problem since signatures are recorded offline on lecture attendance
sheets.
2
3. Table of Content
Serial Number Topic
1 Introduction
2 Block Diagram
3 Implementation
4 Case study
5 Discussion
6 References
3
4. Introduction
Classroom attendance management involves keeping track of
attendance of students in a classroom, laboratory, seminar, etc.
Attendance management is important because attendance affects
students‟ academic performance. In many institutions, if students do
not attend a minimum number of classes for a course, they either fail
the course or they are prevented from earning a regular grade in the
course.
Signatures are widely used to identify individuals in different
application domains such as education, legal, finance, security, and
business. Signatures are unique to individuals, which make it an
obvious reason to use for identification.
5. Why Signature
Signature is preferred over other biometric because of :
1. Ease of Adoption: The signature based system leverages
a familiar action - signing making it easy for user to adopt to the
new system.
2. Low hardware requirement : The Signature based system primarily
requires devices with touchscreen and styluses, which are widely
available and can be integrated into any modern devices.
3. Cost-effectiveness : It will be more cost effective compared to
some biometric systems that involve complex hardware installation
and maintenance.
6. Challenges
Signature verification involves verifying the identity of an individual
based on the analysis of their signature.
As unique features that exist in signatures are not all visible to the
human eyes, so it is very difficult for a human eye to verify the
signature.
Also manual verification requires a lot of time. So an automated
signature verification system is essential to enhance verification.
There are two approaches to automatic signature verification
systems and it is based on the method of acquisition of the
signature: 1. Online approach
2. Offline approach
7. Online Verification Offline Verification
1. Uses electronic tablet and pen to
take input
1. Signature is signed on paper and
photo is given as input to the system
2. System captures dynamic
information like pressure , velocity
and other unique features.
2. Features are extracted by
measuring and analysing
3. More accurate 3. Less accurate
4. More resistance to forgeries 4. Susceptible to forgeries
5. Depends on hardware for signing 5. Doesn’t depend on hardware for
signing
6. Verification process is slower 6. Due to automation verification
process faster and more consistent
8. Automatic signature verification has mainly utilized supervised learning,
commonly classification techniques, where features extracted from
signature samples are stored in a knowledge base, and subsequent
signatures supplied to the system are compared to samples in the
knowledge base.
But here we have a new verification technique is proposed i.e:hierarchical
clustering, which groups data objects into a tree of clusters in hierarchies
which is used for lecture attendance management.
Clustering is the process of grouping data into clusters or classes; such
that objects within a cluster are similar compared to objects in other
clusters.
9. Implementation using Hierarchical Clustering
Hierarchical clustering techniques group data objects into tree of
clusters in hierarchies. There are basically two hierarchical clustering
algorithms:
1. Agglomerative hierarchical clustering algorithms
2. Divisive hierarchical clustering algorithms
Agglomerative hierarchical
clustering algorithms
Divisive hierarchical clustering
algorithms
1. It starts with each data point as its own
cluster and gradually merges the most
similar clusters together until all data points
are in a single cluster.
1. It takes the opposite approach. It starts
with all data points in a single cluster and
then recursively divides clusters into smaller
subclusters.
2. It's a bottom-up approach. 2. It's a top-down approach.
10. 3. computationally expensive for large datasets 3. Computationally expensive for large datasets
4. Prior knowledge about data structure is not
needed
4. Prior knowledge about data structure is
needed
5. Produces better cluster quality 5. Cluster quality produces is lesser
Hierarchical clustering results are displayed
graphically using tree-like diagrams called
dendrograms.
Dendrogram trees displays clusters,sub-
clusters relationships as well as the order in
which clusters are merged.
12. The system consists of two modules.
The first module is used to accept the input of signature, do
comparison with existing signature and if recognized the name of
signature owner will be recorded in attendance list as present.
second module is used to add, delete or train the network with new
signature. The process in both modules are same.
The input signature is accepted, processed and then the output will be
displayed.
To use this system, first, the authorized person such as lecturers need
to log in to run the application. Users will be prompted with main
screen, which required users to users chose the mode, whether the
first mode (add, delete ortrain network), or the second module (take
attendance).
13. Once take attendance is selected, the system will do all these
operations
Signature
segmentation
and
preprocessing
Attendance
Sheet(s)
Preprocessed
signature
image Feature
Extraction
Signature
features
Verification
Estimated
number of times
Each student
attended lecture
Fig :- Framework of Automatic Lecture Attendance Management System
14. 1. Signature segmentation and preprocessing
Signature segmentation involves splitting that image into individual signature images.
Segmentation is straightforward because each signature cell on the attendance sheet is a
rectangle of known dimension.
Preprocessing was carried out to enhance the quality of the signature images.
2. Feature Extraction
Feature extraction is a very important aspect of signature verification because the system’s
efficiency, processing cost, and memory requirement are largely dependent on the feature set.
Also, the feature set considered should sufficiently reveal the characteristics of forged
signatures such as shaky lines, discontinuity of lines, texture density.
To show common and representative characteristics of the signature, plane features were
summarized using four statistical measures (Entropy, Mean, Median and Standard deviation).
Equations of these are given as
15. n = width of the signature image
p = number of black pixels in column i of the signature image
Verification
In this phase, the signatures in each row of the attendance sheet are examined to determine
how many are genuine, that is, how many times each student actually attended lectures.
The steps of verification are :
1. Construction of dendrogram
2. Identification of clusters
3. Determination of frequency of attendance
16. Case study
An experiment was conducted which validate the proposed automatic lecture attendance
management system
Experimental setup
The MATLAB R2010a simulation tool was used to implement the AUTOMATIC LECTURE
ATTENDANCE MANAGEMENT(ALAM) system.
Computer having specs :
1. a processor speed of 2.4 GHz
2. main memory capacity of 4GB
3. 64-bit Windows® 7 operating system
Agglomerative hierarchical clustering algorithm was implemented for the verification phase, and
it was experimentally determined a threshold value of 2.1 was suitable for identifying the clusters
in a dendrogram
17. Data Collection
Student signatures were collected on the lecture attendance sheets which have the following
columns; Students‟ registration number, students‟ name and weeks in the semester for a
duration of 10 weeks.
18. Results and Discussion
Figure in right shows how mean square error
(MSE) varies with the similarity threshold. As
can be observed from the figure, the optimal
similarity threshold value is 2.1, which results
in MSE of 0.96.
Relationship between similarity threshold and
mean square error
Predicted and actual frequencies of lecture attendance
Figure in left shows the actual and predicted
frequencies of lecture attendance for 50
students which resulted in the MSE of 0.96.
19. Conclusion
Although offline signature verification has traditionally been carried out using supervised machine
learning techniques, this work has shown how clustering, which is an unsupervised machine
learning technique can be used to effectively verify signatures in order to manage lecture
attendance.
Thus work can be improved also if individual similarity thresholds for identifying clusters in
dendrograms are determined for each student.
20. References
1. K. Huang, and H. Yan, “Off-line signature verification based on geometric feature extraction and
neural network classification,” Pattern Recognition, vol. 30,1997.
2. C. Kruthi, and D. C. Shet, “Offline Signature Verification Using Support Vector Machine,” In
Fifth International Conference on Signal and Image Processing, 2014.
3. O. Hilton, “The Detection of Forgery,” Journal of Criminal Law and Criminology, vol. 30.,1939.
4. K. Huang, and H. Yan, “Off-line signature verification based on geometric feature extraction and
neural network classification,” Pattern Recognition, vol. 30, 1997.
5. M.S. Arya, and V.S. Inamdar, “A Preliminary Study on Various Off-line Hand Written Signature
Verification Approaches,” International Journal of Computer Application, IJCA, vol. 1, 2010.
6. J. Coetzer, and J. Dupreez, “Off-line signature verification: a comparison between human and
machine performance,” In: Tenth International Workshop on Frontiers in Handwriting Recognition.
Suvisoft, 2006