Software Requirements

Specification
For

Palm Recognition System

Prepared by

Shubham Kumawat (BE/25006/13)

Khushal Agarwal (BE/25001/13)

Rahul Gupta (BE/25024/13)

Project Supervisor: Co-

Supervisor:
Mr Madan Mohan Agarwal Mr Subhash
Chand Gupta

Birla Institute of Technology, Jaipur

16 Aug 2016

Table Of contents
01.
Introduction..........................................................................
............3
1.1
Purpose ..................................................................................
............3
1.2 Problem
Description...........................................................................3
1.3
Objectives..............................................................................
.............3
1.4
Overview................................................................................
............3
02. Overall
Description .........................................................................4
2.1 Product
Perspective ...........................................................................
4
2.2
Methodology..........................................................................
.........4-5
2.3 Overview of Non-Functional
Requirements .....................................6
2.3.1
Input ......................................................................................
.........6
2.3.2
Output ...................................................................................
..........6
2.3.3 Internal
Data....................................................................................6
2.4 Constraints, Assumptions, Dependencies, Guidelines
..........6
2.4.1 Software
Constraints.......................................................................6
2.4.2 Hardware
Constraints .....................................................................6
2.4.3
Guidelines .............................................................................
.........6
2.4.4
Assumptions ..........................................................................
.........6
04. Specific
Requirements.....................................................................7
3.1 Functional
Requirements ..................................................................7
3.1.1
Enrollment..............................................................................
.........7
3.1.2
Verification.............................................................................
........7
3.1.3
Identification .........................................................................
.........7
3.3
Reliability...............................................................................
............7
3.4
Performance ..........................................................................
............8
3.5
Application.............................................................................
............8
04.
References
.8
List of Figures
Figure 1- Flowchart of Palmprint
Algorithm.......................5
Figure 2- Functions of Biometric
System............................................7

_______________________________________________________
____________CSE 2
Palmprint Recognition
01. Introduction
1.1 Purpose

This document describes the functional, non-functional and hardware

requirements and the importance of a palmprint recognition system with briefly
presenting its input and output functionalities. And also this would give an
overview to the client and would give detail specification for the developer.

1.2 Problem Description

Palmprint verification is one of the most reliable personal identification

methods, which is widely used today. Palmprint are popular for personal
identification due to the facts that they are easy to obtain, reliable, extremely
unique and amenability to automation. Although those automated systems are
much easier, they can be prone to noise. So Palmprint readers today are not
ideally suited to handle the large variation of populations that need to be enrolled.

Also there is a high demand for personal identification and

verification for reasons of security. The traditional secure measures,
passwords or ID cards, provide only limited protection for safety systems.
Thus, they cannot meet secure and automated requirements in the
modern world. Because ones unique characteristics cannot be stolen,
forgotten, duplicated, shared or observed, biometrics-based recognition is
emerging as the most reliable solution since it deals with physiological or
behavioural characteristics, which can be used to authenticate a persons
claim to identity or establish an identity. The palmprint recognition system
has advantages over the other physiological biometric systems. Some of
the advantages are fixed line structure, low intrusiveness, low cost
capturing device, low resolution imaging

1.3 Objectives

This project is aimed at,

To develop a system for Palmprint Recognition with increased

speed and accuracy

1.4 Overview

Section 2 here deals with the overall description about the product while
 describing the product perspective, functional and data requirements, input and
output data, general constraints and assumptions of the application briefly. Section
3 gives the detailed description about the functional and non-functional
requirements of the application.

_______________________________________________________
____________CSE 3
Palmprint Recognition

02. Overall Description

2.1 Product Perspective

The users are authenticated by the palmprint recognition systems.

These accept the users, who are authenticated, i.e. whose palmprint
match with a palmprint present in the database. If the user is not
authenticated, then the user is rejected. This process of accepting and
rejecting the user is done on the basis of the matching algorithm.

2.2Methodology
The palmprint recognition system includes pre-processing followed
by roi extraction. After ROI extraction, features are extracted using the
feature extraction algorithms. Then matching is done on the basis of the
extracted features. The palmprint is then accepted or rejected. All these
processes have been studied and reviewed. On the basis of the review,
the proposed system has been described. Each step has been studied in
detail. The proposed palmprint recognition system has been depicted in
figure 4 which is a flowchart of the palmprint recognition system.

Each step is further described below in detail.

Pre-processing :-To reduce the overhead, instead of directly using

the palmprint images, pre-processing needs to be done. Pre-
processing is used to remove distortion, align the palm prints and to
crop the region of interest. This cropped ROI is used for feature
 extraction.

This is done in five steps:

a. Formation of binary image of the palm image
b. Boundary tracking
c. Key points detection
d. establishing a coordination system
e. Extracting the central part

The third step can be accomplished by two approaches, tangent

based and finger based. The tangent based approach is preferred.
This approach considers the edges of the 2 finger holes on the
binary image to be traced. The common tangent of the two finger
holes is considered to be the axis. The key points for the
coordination system are calculated as the midpoint of the two
tangent points.

ROI Extraction The central part of the palm image is segmented

after the preprocessing. Different algorithms segment circular, half
elliptical or square regions for feature extraction. The square region
is the easiest and widely used. The cropped image is then passed
through a low pass filter (LPF), which blurs the image. In this
blurred image, the minor lines get suppressed. The major lines are
also affected, but they are prominent. These are then used for
feature extraction.

Feature Extraction and Matching : For matching the palmprints,

we need to extract some features first. The extracted features are
then used for matching.

_______________________________________________________
____________CSE 4
Palmprint Recognition

Some of the extraction and matching algorithms are line based,

subspace based, statistical and coding based approaches. 1) Line
based approach: This approach develops edge detectors and
makes use of the magnitude of the palm lines. The magnitudes of
the palm lines are projected in x and y coordinates forming
histograms. After this, the first and second order derivatives of the
palm images are calculated. The first order derivative is used to
identify the edge points and corresponding directions. The second
order derivative is used to identify the magnitude of lines. Then the
 Euclidian distance is used for matching. 2) Subspace based
approach: This approach makes use of Principal Component
Analysis (PCA), Linear Discriminant Analysis (LDA) and
Independent Component Analysis (IDA). The spatial coefficients are
considered as the features used for matching. This approach does
not need any prior knowledge of the palmprints. 3) Statistical
approach: These are of two types, local and global. The local
approach transforms the image in another domain. This
transformed image is then divided into several regions such as
mean and variance of each region. The global features include
moments, center of gravity and density. The global approach is
applied on the whole palmprint image. This is the only difference
between the local and global approach. The local approach is
applied on the segments of the palmprint image whereas the global
approach is applied on the whole image. 4) Coding approaches:
This approach uses a single Gabor filter to extract the local phase
information of palmprint. This extracted phase information is used
by the palmprint recognition systems to reduce the registered data
size and to deal with non-linear distortion between palmprint
images. This approach has very low memory requirement.

Accept / Reject The users are authenticated by the palmprint

recognition systems. These accept the users, who are
authenticated, i.e. whose palmprint match with a palmprint present
in the database. If the user is not authenticated, then the user is
rejected. This process of accepting and rejecting the user is done
on the basis of the matching algorithm. This matching is done on
the basis of the extracted features. Classification is the basis for the
palm images to be accepted or rejected.

Fig 1.1 Flowchart of Palmprint Algorithm

_______________________________________________________
___________CSE 5
Palmprint Recognition

2.3 Overview of Non-Functional Requirements

2.3.1 Input

Inputs are the palmprint images and personal details of the particular
person, in the purpose of identification.

2.3.2 Output

Database records with the personal details of the matched image against
the given image, or system generated alerts on strange data patterns and to
develop a system for Palmprint Recognition with increased speed and
accuracy.

2.3.3 Internal Data

The program maintain following internal data

Biometric Template (the distinguish measures) that consists with

the measured features of the images.

2.4 General Constraints, Assumptions, Dependencies, Guidelines

2.4.1 Software Constraints

Windows 2000, Windows XP, Windows 7,8.
MySQL as Database

2.4.2 Hardware Constraints

Palmprint scanning device, camera or any other image acquisition device.

Minimum of 1024MB RAM
2.4.3 Guidelines
Code is kept clean and simple for future upgrades and maintenance

2.4.4 Assumptions

User will provide palmprint image of his/her hand with a preferable quality.

_______________________________________________________
____________CSE 6
Palmprint Recognition

03. Specific Requirements

3.1 Functional Requirements

The main purpose of this project is to provide a better fitted engine for
palmprint recognition. The following functions are provided through an API.

3.1.1 Enrolment

In enrolment, samples of palmprint are captured from the sensor unit and
processed them in to a usable form and stored in the database for that individual.

And it involves
Sample acquisition
Feature extraction
Comparison
Decision Making

3.1.2 Verification

In verification, one or more samples of palmprint images are captured,

processed into a usable form, and then matched against a reference template to
check whether essential features are there. The image will either accept with the
reference template or not.

It involves
Comparison of the sample features against the reference template
A verification decision based on the match result with the reference
 template.

3.1.3 Identification

In identification, palmprint images of the identifiable person are captured,

processed into a usable form (template), and matched against a set of templates
stored in the database. The final result may be empty or contain only one
identifier that is best fitted to the captured image. System should be capable of
identifying persons among a large database with a reasonable speed. That would
be the final goal of the research.

Identification involves
Sample acquisition;
Feature extraction;
Comparison against templates stored in the database
An identification decision based matching result in one or more
attempts

Fig 1.2 Functions of Biometric System

_______________________________________________________
____________CSE 7
Palmprint Recognition

3.3 Reliability

The system should perform reliable identification with large databases, as

biometrical identification systems tend to accumulate False Acceptance Rate with
database size increase..

3.4 Performance

The system should response user within few minutes since the palmprint
recognition systems are real time. Therefore the engine we implement here should
have law response time. So the palmprint verification time should be low and it
requires high computational power. On the other hand this system will maintain a
large database which consists of millions of data. So the data processing time
while identifying a particular image on the database should also be very low.

3.5 Application
 A simple GUI application will be implemented for the demonstration
purpose of the verification engine.

04. References
[1] Tee Connie, Andrew Teoh Beng Jin, Michael Goh Kah Ong, David Ngo
Chek Ling, An automated palmprint recognition system, Image and Vision
Computing , Vol.23, pp.501515, 2005.

[2] Patprapa Tunkpien, Sasipa Panduwadeethorn, Suphakant Phimoltares,

Compact Extraction of Principle Lines in Palmprint Using Consecutive Filtering
Operations, Proceedings of the Second International Conference on Knowledge
and Smart Technologies, pp.39-44, 2010.

[3] Tee Connie, Andrew Teoh, Michael Goh, David Ngo, Palmprint
Recognition with PCA and ICA , Image and Vision Computing NZ, Palmerston
North, pp.227-232, 2003.

[4] I Ketut Gede Darma Putra, Erdiawan, High Performance Palmprint

Identification System Based On Two Dimensional Gabor TELKOMNIKA Vol. 8,
No. 3, pp.309-318, 2010.

[5] Kong W.K., Zhang, D., Li W.X., Palmprint feature extraction using 2-D
Gabor filters. Pattern recognition, Vol. 36, pp.23392347, 2003.

_______________________________________________________
____________CSE 8
Palmprint Recognition