ITC571 Assignment-2

(Project Proposal & Plan)

BIOMETRIC SYSTEMS

(Face Recognition)

Prepared by

AVINASH KANAPARTHI

(11559587)

For the lecturer

ATHER SAEED

Table of Contents

RATIONALE

DEFINING PROBLEM

PURPOSE AND JUSTIFICATION

RESEARCH QUESTIONS

PREVIOUS WORK

METHODOLOGY

Research and System Development Method

Data Collection Methods

Ethical Issues

Requirement OF Compliance

Data Analysis

PROJECT PLAN

Deliverables

Work Breakdown Structure

Risk analysis

Duration

Gantt chart

REFERENCES

BIOMETRIC SECURITY: FACE RECOGNITION

RATIONALE

DEFINING PROBLEM

Biometric is the new technology that is being used mainly for authentication purpose by identify the human being from their physical characteristics such as fingerprints, retina scan, face recognition etc. Biometric security in the present world is being considered the best technique in ensuring safety. Among all the different biometric methods human face recognition has gained a lot of popularity in the past few years and is mainly used by many security agencies to identify the individuals. The added advantage that is there associated with facial recognition is that in facial recognition there is no need for the person to be in contact with the system as the image of the person can be captured from a distance.In face recognition the various patterns of the face are identified with the image of the face that is there in the database and then if the person is identified(AchmadFirdausy, 2012). Mostly this system was developed to provide security but over the years face recognition is also been used in various other applications.

The issues that are associated with this technology are being discussed below:

  • Highly dependent over the surroundings: Face recognition process is highly dependent on the surrounding of the user and in situation where the lighting is not proper it can cause problems in recognizing the person. Even if the person is wearing a cap or is wearing glasses the system can encounter certain drawbacks.
  • Privacy: Privacy is another area of concerns that is associated with facial recognition as any person can use facial recognition software to capture the face of a person with his or her knowledge and can then access personal information or other accounts of the person.
  • Dependent on positioning of the face: Face recognition is highly dependent on the position of the face and only works properly when the user is fully facing the system or is not more than 20 degree off from the system. Even the distance can be a concern in situation where the person is too close to the system(Agrawal & Sharma, 2016).
  • Difference in the facial expression: Another issue that can affect the working of the biometric face recognition system is that in situations where the facial expression of a person are different from the previous instance. Even if the system is asked to predict the expression of a person it can wrongly interpret the expression(Arca, Campadelli, & Lanzarotti, 2006).
  • Adaptability: Adaptability is another issue that is associated with a biometric system as biometric system may fail to adapt to the facial changes that a person is bound to have over the years due to his age, illness or in situation of an injury.
  • Maintenance: Maintenance of the biometric face recognition system is another are of concern as it can be a tough and a highly expensive task. Organizations may be able to spend a lot of money to install a face recognition system but may lag in maintain the system due to the amount of cost that is required for its maintenance.
  • Efficient Threat models: For facial recognition system the threat models should be well considered as this system is used to provide security and there can be instances where this system will attacked by intruders or other malicious parties(Chen, Liao, Lin, & Han, 2001).
  • High implementation cost: The cost that is levied over the implementation of face recognition system is very high thus it can only be afforded by companies or organizations that are big and have adequate resources.

PURPOSE AND JUSTIFICATION

Biometrics face recognition is an emerging technology which has gained a lot of popularity and importance over the year in ensuring security and providing authorization only to the authorized in using certain services. The purpose of this research is to gain complete knowledge over the topic and deliver the best and adequate knowledge about the face recognition.

The reason for using this topic for research is that face recognition is considered to be the future in providing security and personal identification. It is now being used in many applications like Facebook where face recognition is used to tag people over the photographs. It is considered to be the faster and most reliable method in authentication(FirdausyAchmad, 2011).

RESEARCH QUESTIONS

The research questions that are associated with biometric face recognition technique are:

  • What is face recognition technology?
  • What are the areas of its application?
  • What are the limitations of this technology?
  • What are face recognition techniques?

PREVIOUS WORK

In this paper the author (Lin, 2000) have provided a framework explaining the face recognition system and the various issues that are there associated with this new emerging technology. In the recent years face recognition has fathered much attention in various fields and is being implemented by various organizations. The areas that have benefitted most from this technology are network security, content indexing and video compression as in this the people are the main point of attraction. Face recognition not only makes it impossible for the intruders to gain access of the user’s information but also provides user friendly interface. The author in this paper has also discussed the various face recognition algorithms.

METHODOLOGY

Research and System Development Method

The research and system development method that is deployed for the development of the biometric face recognition is performance evaluation. This method is deployed as it is best describes the face recognition system that deploy human verification and identification model. The development model that is proposed for the face recognition system would implement pre-processing, representation and identification module (Moon, Seo, & Pan, 2016). For the purpose of identification of the development method literature review were done so that complete information of the topic can be gathered. The literature review provides complete in depth knowledge about the topic, its areas of use and the area of concerns that are there associated with it.

Data Collection Methods

The various data collection methods that are used in this research are:

  • Questionnaires: Questionnaires are set of questions that are prepared so that they can be asked from researchers. The questionnaires are mailed to the researchers so that they can answer the questions asked in the questionnaire later on. The questions that are asked through the questionnaires are very simple and thus produce effective results.
  • Interviews: Interviews of the researchers are conducted to get instant answers and quick results. Interviews will allow collecting the information very quickly and is very cost effective method for gathering information (Frick, 2009).
  • Previous work: All the online generals, articles and books that have been written in this regard are searched so that complete information regarding face recognition system can be accumulated.

Ethical Issues

The ethical issues that are associated with the research on face recognition are described below:

  • Conflict of interest: Conflict of interest between the researchers can be an ethical issue that is related with the research over facial recognition system.
  • Tampering the information: Another ethical issue that is there is related to the information that is gathered and can be tampered or changed when written.

Requirement OF Compliance

The compliance requirements that are associated with this research work are:

  • The data that is collected should be from reliable resources and the integrity of the data should be maintained.
  • The data gathered should be self-explanatory and accurate.

Data Analysis

The data analysis method that is used to analyze the information is qualitative data analysis method. This method allows analyzing useful information from large sources of data. This method is about interpretations and impressions made by key researchers in their work(Hock Koh, Ranganath, & Venkatesh, 2002).

PROJECT PLAN

Deliverables

The deliverable of the research is that to provide complete knowledge about the biometric face recognition system. The various areas where it is useful and what are the risks and issues that are associated with it.

Work Breakdown Structure

Task Name / Duration / Start / Finish
Biometric: Face Recognition System / 51 days / Mon 29-08-16 / Mon 07-11-16
Starting Phase / 9 days / Mon 29-08-16 / Thu 08-09-16
Defining the problem / 2 days / Mon 29-08-16 / Tue 30-08-16
Defining the need / 2 days / Wed 31-08-16 / Thu 01-09-16
Identifying the technology / 3 days / Fri 02-09-16 / Tue 06-09-16
Understanding the technology / 2 days / Wed 07-09-16 / Thu 08-09-16
Requirements / 14 days / Fri 09-09-16 / Wed 28-09-16
Understanding the objectives / 3 days / Fri 09-09-16 / Tue 13-09-16
Identifying the various data collection techniques / 2 days / Wed 14-09-16 / Thu 15-09-16
Selecting the data collection technique / 1 day / Fri 16-09-16 / Fri 16-09-16
Collecting data / 3 days / Mon 19-09-16 / Wed 21-09-16
Identifying the resources / 2 days / Thu 22-09-16 / Fri 23-09-16
Finalising the resources / 1 day / Mon 26-09-16 / Mon 26-09-16
Analyse the data collected / 2 days / Tue 27-09-16 / Wed 28-09-16
Methodology / 5 days / Thu 29-09-16 / Wed 05-10-16
Identifying the methodology / 3 days / Thu 29-09-16 / Mon 03-10-16
Finalizing the methodology to be used / 2 days / Tue 04-10-16 / Wed 05-10-16
Implementation / 7 days / Thu 06-10-16 / Fri 14-10-16
Implementing the selected methodology / 7 days / Thu 06-10-16 / Fri 14-10-16
Testing / 9 days / Mon 17-10-16 / Thu 27-10-16
Comparing deliverables with objectives / 3 days / Mon 17-10-16 / Wed 19-10-16
Perform tests / 4 days / Thu 20-10-16 / Tue 25-10-16
Collect test results / 2 days / Wed 26-10-16 / Thu 27-10-16
Maintenance / 4 days / Fri 28-10-16 / Wed 02-11-16
Identifying new methods / 2 days / Fri 28-10-16 / Mon 31-10-16
Implementing new methods / 2 days / Tue 01-11-16 / Wed 02-11-16
Project ends / 3 days / Thu 03-11-16 / Mon 07-11-16
Complete Documentation / 3 days / Thu 03-11-16 / Mon 07-11-16

Risk analysis

Risk / Description / Level / Mitigation Plan
Budget / The estimated budget may exceed. / Medium / The budget should be flexible.
Deadline / The project may exceed the estimated time. / Medium / The timeline schedule should be flexible and extra resources should be allotted if a task is running more than the set time(Hsieh & Chen, 2011).
Quality / The facial recognition system may not provide desired outcomes. / Low / First a prototype is needed to be developed so that the system can be tested.

Duration

Total Time: 51 Days

Start Date: 29-08-2016

End Date: 07-11-2016

Gantt chart

REFERENCES

Achmad, B. & Firdausy, K. (2012). Neural Network-based Face Pose Tracking for Interactive Face Recognition System.International Journal On Advanced Science, Engineering And Information Technology,2(1), 105.

Agrawal, A. & Sharma, P. (2016). Pose Invarient Face Recognition System.International Journal Of Engineering And Computer Science.

Arca, S., Campadelli, P., & Lanzarotti, R. (2006).A face recognition system based on automatically determined facial fiducial points.Pattern Recognition,39(3), 432-443.

Chen, L., Liao, H., Lin, J., & Han, C. (2001). Why recognition in a statistics-based face recognition system should be based on the pure face portion: a probabilistic decision-based proof.Pattern Recognition,34(7), 1393-1403.

Firdausy, K. & Achmad, B. (2011).Automatic Frontal Face Pose Tracking for Face Recognition System.International Journal On Advanced Science, Engineering And Information Technology,1(4), 399.

Frick, K. (2009). Microcosting Quantity Data Collection Methods.Medical Care,47(Supplement), S76-S81.

Hock Koh, L., Ranganath, S., & Venkatesh, Y. (2002). An integrated automatic face detection and recognition system.Pattern Recognition,35(6), 1259-1273.

Hsieh, C. & Chen, W. (2011).A Face Recognition System Based on ASM Facial Components.AMM,58-60, 2314-2319.

Lin, S. (2000). An Introduction to Face Recognition Technology.Informing Science Special Issue On Multimedia Informing Technologies,3(1). Retrieved from

Moon, H., Seo, C., & Pan, S. (2016). A face recognition system based on convolution neural network using multiple distance face.Soft Comput.