Face Recognition System with Attendance, Papers of Artificial Intelligence

Download Face Recognition System with Attendance and more Papers Artificial Intelligence in PDF only on Docsity!

REAL TIME FACE ATTENDANCE SYSTEM USING

DEEP LEARNING

REAL TIME FACE ATTENDANCE SYSTEM

A project report submitted in partial fulfilment of the requirements for the award of Bachelor of Engineering ________ Engineering Faculty of Engineering _________________

APPROVAL FOR SUBMISSION

I certify that this project report entitled “REAL TIME FACE ATTENDANCE SYSTEM” was prepared by __________ has met the required standard for submission in partial fulfillment of the requirements for the award of Bachelor of Engineering _______________________________ Approved by, Signature : Supervisor : Date :

ACKNOWLEDGEMENTS

I would like to thank everyone who has contributed to the successful completion of this project. First, I would like to express my utmost gratitude to my research supervisor, _______________ who despite being extraordinarily busy with her/his duties, took time to give invaluable advice and guidance throughout the development of the research. In addition, I would also like to express my deepest appreciation to my loving parents and family members for their constant support and encouragement. Last but not least, I am grateful for the unselfish cooperation and assistance that my friends had given me to complete this task.

TABLE OF CONTENTS

DECLARATION ii APPROVAL FOR SUBMISSION iii ACKNOWLEDGEMENTS v ABSTRACT vi TABLE OF CONTENTS vii LIST OF TABLES x LIST OF FIGURES xii LIST OF SYMBOLS / ABBREVIATIONS xv LIST OF APPENDICES xvii CHAPTER 1 INTRODUCTION 1 1.1 Background 1 1.2 Problem Statement 3 1.3 Aims and Objectives 4 1.4 Thesis Organization 4 2 LITERATURE REVIEW 5 2.1 Student Attendance System 5 2.2 Face Detection 6 2.2.1 Viola-Jones Algorithm 10 2.3 Pre-Processing 12 2.4 Feature Extraction 16 2.4.1 Types of Feature Extraction 20 2.5 Feature Classification And Face Recognition 21 2.6 Evaluation 22 3 METHODOLOGY 24 3.1 Methodology Flow 24 3.2 Input Images 27 3.2.1 Limitations of the Images 28 3.3 Face Detection 29

3.3.1.4 Contrast Limited Adaptive Histogram

 - 3.3.1 Pre-Processing - 3.3.1.1 Scaling of Image - 3.3.1.2 Median Filtering - 3.3.1.3 Conversion to Grayscale Image - Equalization 

• 3.4 Feature Extraction
  • 3.4.1 Working Principle of Original LBP
  • 3.4.2 Working Principle of Proposed LBP
  • 3.4.3 Working Principle of PCA
  • 3.4.4 Feature Classification
  • Recognition 3.4.5 Subjective Selection Algorithm and Face
• 4 RESULT AND DISCUSSION
  • 4.1 Result
  • 4.2 Discussion
  • 4.3 Comparison of LBP and PCA
  • 4.4 Comparison with Previous Research
  • 4.5 Comparison with Luxand Face Recognition Application
  • 4.6 Weakness of the Algorithm
  • 4.7 Problems Faced and Solutions Taken
• 5 CONCLUSION AND RECOMMENDATION
  • 5.1 Conclusion
  • 5.2 Recommendation
• REFERENCES

4.8 Overall Performance of PCA 4.9 Overall Performance of the Proposed Approach 4.10 Performance of Proposed Algorithm in Different Intensity Range 4.11 Summary of Comparison with Previous Research 4.12 Comparison of Proposed Algorithm and Luxand Face Recognition

LIST OF FIGURES

FIGU

RE

TITLE

1.1 Block Diagram of the General Framework 2.1 Haar Feature (Docs.opencv.org, 2018) 2.2 Integral of Image (Srushti Girhe et al., 2015) 2.3 False Face Detection (Kihwan Kim, 2011) 2.4 Images show Checkerboard Effect significantly increasing from left to right (Gonzalez, R. C., & Woods, 2008) 2.5 Facial images were converted to grayscale, histogram equalization was applied and images were resized to 100x100 (Shireesha Chintalapati and M.V. Raghunadh, 2013) 2.6 PCA Dimension Reduction (Liton Chandra Paul and Abdulla Al Sumam, 2012) 17 2.7 Class Separation in LDA (Suman Kumar Bhattacharyya and Kumar Rahul, 2013) 18 2.8 LBP Operator (Md. Abdur Rahim et.al, 2013) 2.9 Artificial Neural Network (ANN) (Manisha M. Kasar et al., 2016) 2.10 Deepface Architecture by Facebook (Yaniv Taigman et al, 2014) 3.1 Flow of the Proposed Approach (Training Part) 25 3.2 Flow of the Proposed Approach (Recognition Part) 3.3 Sample Images in Yale Face Database (Cvc.cs.yale.edu, 1997)

4.12 Images with Different Intensity 4.13 Performance of Proposed Algorithm in Different Intensity Range 4.14 Images of Students With or Without Wearing Glasses 4.15 Training Image VS Testing Image

LIST OF SYMBOLS / ABBREVIATIONS

χ^2 Chi-square statistic 𝑑 distance 𝑥 input feature points 𝑦 trained feature points 𝑚𝑥 mean of x 𝑆𝑥 covariance matrix of x 𝑋𝑐 x coordinate of center pixel 𝑌𝑐 y coordinate of center pixel 𝑋𝑝 x coordinate of neighbour pixel 𝑌𝑝 y coordinate of neighbour pixel 𝑅 radius 𝜃 angle 𝑃 total sampling points 𝑁 total number of images 𝑀 length and height of images 𝛤𝑖 column vector 𝜑 mean face Φ𝑖 mean face subtracted vector 𝐴 matrix with mean face removed 𝐴𝑇^ transpose of 𝐴 𝐶 covariance matrix 𝑢𝑖 eigenvector of 𝐴𝐴𝑇 𝑣𝑖 eigenvector of 𝐴𝑇𝐴 λ eigenvalue

LIST OF APPENDICES

APPENDIX TITLE

A Create Database by Enhanced LBP B Enhanced LBP Encoding Process with Different Radius Sizes C Enhanced LBP Separate into Blocks and Its Histogram D Enhanced LBP with Distance Classifier Chi Square Statistic E Create Database by PCA F PCA Test Image Feature Extraction G PCA with Distance Classifier Euclidean Distance H MATLAB GUI

CHAPTER 1

INTRODUCTION

The primary goal of this project is to create an automated student attendance system based on facial recognition. The test pictures and training images of this suggested technique are restricted to frontal and upright facial images that only contain a single face to improve performance. To ensure no quality variation, the test photographs and training images must be taken using the same equipment. To be identified, the pupils must also register in the database. The user-friendly interface allows for immediate enrollment. 1.1 Background To recognize relatives, friends, or other people we are familiar with, face recognition is essential in daily life. We might not realize that several steps have been taken to recognize human faces. Our ability to acquire information and analyze it throughout the recognition process is a result of human intellect. The image that is transmitted to our eyes, and more particularly, the retina, provides us with information in the form of light. Electromagnetic waves take the shape of light, which is projected to human eyesight after being transmitted from a source onto an object. According to Robinson-Riegler, G., & Robinson-Riegler, B. (2008), when the human visual system has processed the image, we really identify the object's shape, size, contour, and texture in order to analyze the data. The analysed information will be compared to other representations of objects or face that exist in our memory to recognize.

ratios between the located features and the shared reference points. Goldstein, Harmon, and Lesk expanded on the experiments in 1970 by automating the detection of other traits including hair color and lip thickness. Principle component analysis (PCA) was initially presented as a solution to the face identification problem by Kirby and Sirovich in 1988. Then, and up until the present day, several experiments on facial recognition have been undertaken (Ashley DuVal, 2012).

1.2 Problem Statement The traditional method of recording student attendance frequently has a number of issues. By doing away with traditional student attendance marking methods like calling out student names or verifying individual identity cards, the facial recognition student attendance system highlights its simplicity. Not only do they obstruct the teaching process, but they also divert students' attention at test times. During lecture sessions, an attendance list is distributed around the classroom in addition to calling names. It may be challenging to circulate the attendance sheet around a lecture class, particularly one with a big number of pupils. The manual signing of students' presence, which is cumbersome and causes them to become distracted, is thus recommended to be replaced with a facial recognition student attendance system. Additionally, the automatic facial recognition-based student attendance system can overcome the issue of fraudulent approaches, and lecturers no longer need to count the number of students repeatedly to confirm their attendance. The challenges of face recognition are mentioned in the study put forth by Zhao, W. et al. Identifying recognized from unknown photos is one of the challenges in face recognition. However, the training procedure for the face recognition student attendance system is sluggish and time-consuming, according to a paper by Pooja G.R. et al. (2010). Also, it was noted in the research by Priyanka Wagh et al. (2015) that variations in illumination and head postures are frequently the issues that might impair the effectiveness of a facial recognition-based student attendance system. As a result, a real-time functioning student attendance system must be created, which implies the identification procedure must be completed within specific time limits to avoid omission. The extracted face traits that serve as a representation of the students' identities must remain constant while the background, lighting, stance, and expression vary. High precision and quick computation times will be used to gauge performance.