REAL-TIME WIRELESS FACE RECOGNITION SYSTEM (RT-WiFARES) UMMI NAZIRAH BINTI JAMIL KHAIR This Report Is Submitted In Partial Fulfillment Of The Requirements For The Bachelor Degree In Electronic Engineering (Wireless Communication) Faculty Of Computer Engineering And Electronic Engineering. Universiti Teknikal Malaysia Melaka Jun 2013
24
Embed
REAL-TIME WIRELESS FACE RECOGNITION SYSTEM (RT …eprints.utem.edu.my/12843/1/Real-Time_Wireless_Face_Recognition_System...GStreamer dan OpenCV. Sistem pengecaman wajah ini akan mengesan
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
REAL-TIME WIRELESS FACE RECOGNITION SYSTEM (RT-WiFARES)
UMMI NAZIRAH BINTI JAMIL KHAIR
This Report Is Submitted In Partial Fulfillment Of The Requirements For The
Bachelor Degree In Electronic Engineering (Wireless Communication)
Faculty Of Computer Engineering And Electronic Engineering.
Universiti Teknikal Malaysia Melaka
Jun 2013
ii
iii
DECLARATION
“I hereby admit that this report is my own work except that every such summaries
and excerpts only have me explain the source.”
Signature
Name
Date
iv
SUPERVISOR DECLARATION
"I acknowledge that I have read this work in my / our this work is sufficient in scope
and quality for the award of a Bachelor of Electronic Engineering (Wireless
Communication).”
Signature
Name
Date
v
For mom, dad and family
vi
ACKNOWLEDGEMENT
Alhamdulillah, thanks, to the Almighty Allah Most Gracious, Most Merciful for the
blessing and grace, I have been able to implement this undergraduate project. I
would like to take this opportunity to express appreciation to all those who have
helped me along the implement of this PSM project especially my supervisor, Mr.
Engr. Khairul Muzzammil Bin Saipullah on giving me a lot of advices, ideas and
support me on designing this software. I express my gratitude and hope God will
give back to good guys.
vii
ABSTRACT
This project is aimed to design a real – time wireless face recognition system
that is using two kinds of libraries in the C++ environment that are GStreamer and
OpenCV libraries. This system is going to detect a human face on a live recording
video session that are using IP camera. It then will match the captured image with the
database. This process happens in real-time concept which is the output response to
the input is synchronous and no time delay is expected. A sample image will be used
as reference in the database. It will have the detail of the reference image for
example their names. Thus when an image is captured by the video recording, it then
will match the captured image with the sample image that is already stored in the
database. The objective of this project is to enhanced current CCTV limitation such
as reduce the human intervention in the monitoring process and produce a wireless
connection between the IP camera, the router and the controller (PC). Other than
automatically detect a human face, this project will also enable an update to the
database by creating a GUI interface. This interface will allow the inserting of new
image into database if required.
viii
ABSTRAK
Projek ini bertujuan untuk mencipta satu sistem masa nyata pengecaman
wajah tanpa wayar yang berasaskan dua jenis fungsi di dalam sistem C++ iaitu
GStreamer dan OpenCV. Sistem pengecaman wajah ini akan mengesan wajah
manusia melalui rakaman video yang menggunakan kamera IP. Ianya kemudian akan
memadankan wajah yang ditangkap dengan sistem pangkalan data. Proses ini berlaku
dalam konsep masa nyata iaitu tindak balas sistem adalah berkadar terus dengan
input yang dikesan dan melibatkan penggunaan masa yang pantas. Proses
pengecaman wajah dilakukan dan dikawal oleh satu pangkalan data. Pangkalan data
ini mengandungi sampel imej termasuklah pengenalan mengenai imej tersebut
seperti nama dan sebagainya. Apabila satu imej telah dikesan oleh system, imej
tersebut akan melalui proses pengecaman wajah oleh pangkalan data. Objektif bagi
projek ini adalah untuk memperbaiki kekurangan yang terdapat pada system CCTV
masa kini seperti mengurangkan penglibatan manusia di dalam proses pemantauan
dan juga menggunakan sambungan tanpa wayar di antara alatan yang digunakan iaitu
kamera IP, router dan komputer pengawalan. Selain daripada boleh mengesan wajah
dengan pantas, projek ini juga akan membolehkan proses mengemaskini data
dilakukan dengan mewujudkan satu paparan antaramuka pengguna. Paparan ini akan
membenarkan imej baru dimasukkan ke dalam pangkalan data sekiranya diperlukan.
ix
CONTENTS
CHAPTERS CIRCUMSTANCES PAGE
PROJECT TITLE i
DECLARATION
SUPERVISOR DECLARATION
DEDICATION
ACKNOWLEDGEMENT
ABSTRACT
iii
iv
v
vi
vii
ABSTRAK viii
CONTENTS ix
LIST OF FIGURES
LIST OF TABLES
xii
xiv
LIST OF ABBEREVIATIONS
LIST OF APPENDIX
xv
xvi
I INTRODUCTION
1.1 Introduction 1
1.2 Project objective 4
1.3 Problem statement 4
1.4 Project scope 5
1.5 Project importance
1.6 Impact of commercialization and research
advancement
1.7 Progression flowchart for PSM I
1.8 Progression flowchart for PSM II
6
6
7
8
x
1.9 Thesis outline 9
II LITERATURE REVIEW
2.1 Introduction 10
2.2 WLAN connection 11
2.2.1 IP Camera
2.2.1.1 FOSCAM IP Camera
12
13
2.2.2 Wireless Router 16
2.3 C++ Visual Programming 17
2.3.1 GStreamer Library 17
2.3.2 OpenCV
2.3.3 Win32 Console Applications
2.3.4 MFC
18
21
22
2.4 Image Processing 23
2.5 Face Detection and Recognition 24
2.5.1 Importance of Face Detection 25
2.5.2 Face Recognition 26
2.5.3 SURF 27
2.6 Movement Detection 28
III METHODOLOGY
3.1 Project Methodology 29
3.2 Project Architecture 30
3.2.1 Project Flowchart
3.2.2 Gantt Chart
3.3 Coding Developments
3.4 GUI Creating For Database Updating
3.5 Software Testing
30
35
37
39
41
xi
IV RESULT AND DISCUSSION
4.1 Expected Result 41
4.1.1 Video live recording
4.1.2 WLAN Connection
4.2 Discussion
41
47
48
V CONCLUSION AND SUGGESTION
5.1 Conclusion
5.2 Suggestion
50
51
REFERENCES
APPENDICES A: SOURCE CODES
52
53
xii
LISTS OF FIGURES
No Title Page
1 WLAN devices 12
2 IP camera and analog camera 13
3 FOSCAM FI8910W used in this project 14
4 Wireless router 16
5 GStreamer pipeline 18
6 OpenCV structure 20
7 The interface of Win32 Console App 21
8 The interface of MFC 23
9 Image enhancement in term of contrast 24
10 Example of face detection 25
11 Example of face recognition 26
12 Example of SURF Keypoint on A Palm 27
13 Screen shots of object recognition using SURF 28
14 Project architecture 30
15 System flowchart 31
16 The function of image conversion from RGB to gray 33
17 Function of Haar Cascade 33
18 FLANN extractor 34
19 The recorder used 35
20 Gantt chart 36
21 SURF extractor 37
22 Example FLANN coding 37
23 Flowchart of the face detection and recognition
process
38
24 Image processing towards the frame 24
xiii
25 The captured image and its respective name 40
26 The database interface 40
27 The interface of the GUI 41
28 Adding first the person in the database 44
29 The person is successfully being detected 45
30 Adding an image into a database 46
31 Two image recognized 46
32 One subject recognized as ‘unknown’ 47
33 WLAN connections between devices 47
34 Window appears when fail to capture the image 48