Includes CD

Includes references (76-82 p.)

'ABSTRACT Automatic recognition of people has received much attention during the recent years due to its many applications in different fields such as law enforcement, security applications or video indexing. Face recognition is an important and very challenging technique to automatically recognize people's face. Face recognition systems play a vital role in many applications including surveillance, biometrics and security as the rapid development of computer science and pattern recognition, visual technology is widely applied in industry and daily life. The goal of this research is to implement face recognition on a real life platform for security system. The software should be able to take an image of a person through the camera connected to the USB of a computer, recognize the individual based on previously stored samples and grant access or deny access through a device connected to another port of the same computer. The method that will be considered is principal component analysis (PCA) method for face recognition. A database of trained images is stored on the computer for recognizing a test image. The image from the camera is resized and cropped with MATLAB. This image is further processed by comparing it with images in the database. If it matches any of the images in the database, a signal is sent to a device connected to another USB port (Embedded system) to grant access for the person that the image is taken from. If the image does not match any of the images in the database then, the accessis denied for the person. Key words: Face recognition, PCA, Embedded system'

1 CHAPTER 1

1 INTRODUCTION

2 BIOMETRIC SYSTEM

3 Applications

3 Biometric error

4 INTRODUCTION TO FACE RECOGNITION

5 Problem that may occur during face recognition

5 Scale invariance

5 Shift invariance

6 Illumınatıon invariance

6 Emotional expression and detail invariance

6 Noise invariance

7 GOAL

7 OVERVIEW

8 CHAPTER 2

8 LITERATURE SURVEY

8 INTRODUCTION

8 FACE DETECTION

9 Skin tone face detection

11 Viola-jones face detection algorithm

11 Local SMQT based face detection

14 FACE RCOGNITION

14 Background and related work

16 Outline of a typical face recognition system

16 The Image acquisition module

17 The pre-processing module

19 The feature extraction module

19 The classification module

19 Tranning set

20 Face Library or Face database

20 Coventional techniques in face recognition

20 Correlation

21 Linear Discriminant Analysis

23 Independent Component Analysis

24 Other conventional methods for face recognition

25 Local features analysis

26 Nonnegative matirx factorization

27 Increments nonnegative matrix factorization

29 Histogram based method for face recognition

29 Local binary pattern for face recognition

30 PDF based face recognition system

32 CHAPTER 3

32 FACE RECOGNITION USING PCA

32 INTRODUCTION

37 EIGENFACES CALCULATION

42 USING EIGENFACES TO CALSSIFY A FACE IMAGE

43 REBUILDING A FACE IMAGE WITH EIGENFACES

46 CHAPTER 4

46 EMBEDDED SYSTEM

46 INTRODUCTION

46 EMBEDDED SYSTEM

48 EMBEDDED SYSTEM USED IN THIS RESEARCH

48 De-bounced switch

49 The switching circuit (LCD)

50 Electric Lock

50 Liquid Crystal Display (LCD)

52 Microcontroller (UC) PIC18F4550

53 Overview of the USB peripheral

54 USB Status and Control

55 IMPLEMENTATION OF THE ES

56 Communication with the ES

57 CHAPTER 5

57 TESTING AND RESULT

57 DATA SETS

57 AT & T database of faces(formerly known as ORL database of faces )

58 Face Recognition data, university of ESSEX, UK

58 Personal data set

58 SYSTEM HARDWARE FOR TESTING

59 SYSTEM SOFTWARE FOR TESTING

60 TEST RESULT FOR AT&T DATABASE OF FACES

62 TEST RESULT FOR ESSEX DATA SET

64 TEST WITH PERSONAL DATA SET WITH NO FACE DETECTION

69 TEST WITH PERSONAL DATA SET WITH FACE DETECTION

71 OVERALL TEST OF THE SYSTEM

73 CHAPTER 6

73 CONCLUSION AND FURTHER RESEARCH

73 CONCLUSION

74 FURTHER RESEARCH

76 REFERENCES