الفهرس 
titelOnly 14 pages are availabe for public view
 |  | from | 32 |  |  |
| | | from | 32 | | |
Abstract
has received much attention in recent years due to its many applications such as human computer interface, video surveillance and face image database management. The purpose of this research is to study various techniques for face recognition and compare their performance by implementing them, then investigate the possible improvements on these techniques increase their efficiency or to find new technique that has more advantages such as reduce the execution time and its flexibility. In this way, the objective of this thesis is obtained by proposed new algorithm in face detection, and another algorithm of face recognition.
The thesis presents general concepts about human face recognition and face recognition system methods and applications. We propose enhanced face detection technique based on color correction approach and SMQT Features because face detection is an important part of face recognition as the first step of face recognition system. The proposed method is applied on color images gathered from various sources such as Internet, and Georgia Database. Then the thesis presents face recognition system using Artificial Neural Network with Pattern Net. The proposed method was tested on Olivetti and ORL face database. The result derived from proposed method is compared with the other techniques such as K-means, and fuzzy ant with fuzzy C-means the recognition rate are obvious exaltation than other methods. Therefore, the proposed recognition system is good for facial recognition. Thesis consists mainly of five chapters as follows:
Chapter One presents the motivation behind this thesis and defined the fundamental problem considered with real applications. It also summarized the contributions made toward this problem.
Chapter Two: contains general concepts about human face recognition, methods and applications.
Chapter Three: introduces a proposed new technique that can detect human faces from an image based on Color Correction Approach (CCA) and Successive Mean Quantization Transform (SMQT) Features. Numerical results underline the robustness of the presented approach for tested images of faces which have a specific complexion varying under certain range are shown It.