Design Proposal


ENGR 103 - Spring 2013
Freshman Engineering Design Lab
Biometric Recognition for Control
Project Design Proposal

Date Submitted: April 9, 2013


Submitted to:
Chris Lester, clester@minerva.ece.drexel.edu
Group Members:
Dipesh Bhakta, dhb39@drexel.edu

Sachin Gandhi, sag@drexel.edu

Berk Ozoglu, bo73@drexel.edu

Aravind Ponukumati, asp79@drexel.edu


Abstract:

The motivation of this project is to create a system that would enhance the security and privacy for multiple users on their personal computers. The goal is to create an application that would utilize multiple biometric characteristics, such as facial features, voice, and fingerprints, to correctly identify an user and allow him/her to access his/her material on a computer. The technical challenges the group expects to face in the process of this project are creating a signal processing system that will make use of a webcam to identify facial features, a microphone to identify vocal patterns, and a fingerprint scanner to identify fingerprints. Another challenge is interfacing the signal processing systems together to function in identifying a person correctly. The final deliverable will be a MATLAB script that will be able to integrate a standard webcam, microphone, and fingerprint scanner to correctly identify a user and grant him/her access to his/her account on a computer.

Introduction


Biometric recognition is a rapidly-developing pattern recognition system used for a variety of purposes including verification, identification, and security as seen in Figure 1 [1]. Biometric recognition has applicability for individuals with restricted mobility such as those who cannot type or speak their password and acts as an additional safeguard for security purposes. Many biometric recognition systems “conduct a one-to-many comparison to establish an individual’s identity” [1]. It is also much more difficult for hackers and those with malicious intentions to forge or replicate biometric features. Many unique individual features can be used to distinguish one person from another such as fingerprints, facial features, irises, hand geometry, and even a person’s voice [1].



Figure 1. Sample biometric recognition flowchart for verification and identification of an individual.


The major tasks to be performed throughout the ten-week project are to develop a multi-faceted identification system that can cross-reference a person’s facial photograph and fingerprint with an existing database. There are inherent challenges when trying to accomplish this task, primarily due to the complexity of biometric features and the generation of a unique template for each individual. In the end, the desired result is to successfully identify, with minimum errors in identifying the wrong individual or failing to identify the proper individual, a person using multiple physiological characteristics.

Deliverables


At the conclusion of the project, a successfully working software model will be developed that is capable of identifying an individual using multiple biometric identifiers such as facial features and fingerprinting. To achieve this goal, an external fingerprint scanner will be purchased and the software will be developed to take webcam input into MATLAB software. The intended design will contain a database of various biometric templates for predetermined individuals, and simulations will be conducted to attempt to identify various individuals in the lab group and volunteers. All of these test cases will also be presented at the end of the project to assess the functionality of the design.
Schematics will be developed early in the term to plan out the design and setup of the project. Flowcharts and pseudocode will also be prepared to direct the progress of writing the code. Old developments of the code will be saved after completion of major milestones such as implementation of fingerprint analysis and digitization of various physiological features captured by the webcam.

Technical Activities

Literature Study and Signal Processing Introduction

This task will mainly include initial research and getting acquainted with the process in general. It will include planning how the other tasks will be completed as well as research that needs to be done in order to find a solution to the problem.


Digitization of Facial Features

This task will consist of setting up one of the main forms of signal processing in which the viewers face will be recognized. It will involve the use of a webcam built into many personal computers. In order to achieve this goal, code will be used to identify facial features based on an algorithm integrated into the program.

Voice Identification

Voice identification will be another method of recognizing individuals and add another layer of security in the identification process. This task will include design of a systematic algorithm that can identify a person by analyzing their speech patterns and tone.

Fingerprint Analysis


Fingerprint analysis will include setting up the fingerprint sensor and integrating it into MATLAB. The analysis will then make use of code to analyze the outputs of the sensor and a response from the system.

System Integration


This task will focus on the system as a whole as it forms and individual pieces will be incorporated. The design will be assembled and physical issues will be addressed.

Testing and Debugging


Testing will take place using the group members as well as other individuals that would like to participate. Correct identification of faces in the database will be the main concern, while other minor bugs such as issues with the output of the system will be addressed.

Final Report Preparation


This spot is reserved for preparing anything to do with the final report and presentation, from planning the presentation to editing the report.


Project Timeline


            The timeline for completion of the project is given in Table 1 below. Time for literature review and familiarization with biometric recognition has been allotted. In addition, some knowledge of image and signal processing is required for facial and voice recognition, respectively.

Table 1. Biometric recognition timeline over a 10-week period. Various weeks have been allotted for literature study, development of voice, fingerprinting, and facial feature identification as well as testing.



By around Week 2, it is expected that some sort of algorithm is being developed for the voice and facial feature identification. The fingerprinting analysis is added later to the project because it may take time for the hardware to arrive. However, it is expected that the system is being integrated by around Week 5 and testing and debugging occurs by Week 6. It is at this point that official test cases can be analyzed to ensure that the biometric recognition is working as expected. The final report is then developed by Week 8 through to the end of the term, and the testing should be completed by Week 9 at the latest.



Facilities and Resources


The group intends to make use of a computer webcam to obtain sample photographs of subjects being identified. This resource will be used to identify various facial features in the subject like distance between eyes, skin tone, and various other distinguishing characteristics [2]. Another resource that will need to be purchased is an external fingerprint scanner. Unlike fingerprint scanners that may be built-in to computers, the external fingerprint scanner can be attached to any computer with appropriate drivers and be yet another method of identification for biometric sensing [2]. Fingerprint recognition techniques are already prevalent in the market and therefore an obvious choice for preliminary biometric identification of an individual.



Expertise


            Developing a functional computer application capable of doing intensive image and signal processing requires significant prior experience with computer programming. Familiarity with MATLAB/Java is recommended for this project because of the well-written API and preexisting classes for processing external video and audio output. In addition, some knowledge of biomedical engineering is useful to understand which physiological features are unique amongst individuals. For example, it is important to know that features like fingerprints can be practically used to identify people while other features like the shape of the head may be too general for implementation. Some existing biometric technologies and their advantages and disadvantages are seen in Figure 2:

Figure 2. Pro-con analysis table of biometric technologies identifying various physiological features.



Some knowledge of the technology being used such as the webcam and fingerprint scanner would also be a boon while completing the project. Being able to interface the external hardware to MATLAB is a critical component of the project to be able to gather the input data. In addition, some knowledge of signal processing techniques should be learned during the literature study period to become more advanced in the subject.


Budget

The breakdown of the project budget is given in Table 2. Various hardware materials have been accounted for and materials that can be obtained free of charge has not been tabulated.

Table 2. Budget summary detailing hardware costs for biometric sensors.


Fingerprint Reader

The Eikon fingerprint reader has a capacitive sensor and is already built with the functionality to perform “biometric sensing in the device” [3]. The component will be used as a preliminary method of identification and an eventual check as to the identity of a person trying to access the computer in question.

7.2        Webcam





The Logitech Webcam has a sensor that takes in light. The webcam then converts this light into electrical signals, which are fed to a processor and then finally converted into a digital image. This component serves to assess the facial characteristics of an individual. This information will be fed into a script that determines the identity of a person.


7.3        Microphone

The built-in microphone in Apple head will receive audio input. This component will function alongside a MATLAB code to identity the vocal patterns of a user, and compare them with the ones stored in the database of user information.  





References

[1]        A. K. Jain. (2007). “Biometric recognition.” Nature. [Online]. 449, pp. 38-40.
[2]        S. Prabhakar. (2003, March). “Biometric recognition:  Security and Privacy
Concerns.” IEEE Security and Privacy. [Online]. 1(2), pp. 33-42.
[3]        Anonymous. “UPEK Eikon swipe fingerprint reader.” Internet. [2013, April 7].


No comments:

Post a Comment