Face Detection Realizatlon By Useing Gabor Wavelet And BP Neural Network

In recent years, It’s well known that the number of traffic accidents which cause by driver fatigue is increasing heavily. Thus it is becoming a hot topic that discover a effective way to avoid the happening of fatigue driving. In this article, we put our focous on face detection which is a important part in fatigue driving detection system. A face detection algorithm which based on Gabor transform and BP neural network was proposed in this article. It runs in the platform of Matlab and use the transformation of Gabor wavelet convolve with the image of face.Then it’s finish the process of image feature extraction and we should make reduction of dimensional. In the algorithm,we construct a BP neural networks to train the samples extracted from the face database and finally do the face detection for the test sample using template matching theory. As this steps finished, we can achieve the face detection of test samples. This article contains the following modules:1、Introduce the basic concept of Gabor wavelet, and stress the advantage in the process of image feature extraction.2、Constrast the advantage of BP neural network in Artificial Neural Networks, and then determine BP neural network as the way to test sample. 3、For image feature extraction,it’s use PCA as the way to make the reduction of dimensional.4、Using the platform of Matlab, we realize the whole process of the face detection system.