Face Detection Using Artificial Neural Network and Wavelet Neural Network

Recognition and detection of face are the most important aspects in different fields like video surveillance, interface of computer with human and database management of images. This paper discusses about the feature extraction of an image using Gabor transform and then the extracted output is used as the input to the neural network. The training of the neural network is done with input image data. The Gabor wavelet gives emphasis to the vital parts of the face containing cheeks, mouth, nose and eye. Neural network used as a classifier technique for doing the recognition and unrecognizing of face. Also in this paper a hybrid technique using wavelet transform and neural network called as wavelet neural network (WNN) is used for the face detection. The multi scale analysis quality of wavelet transforms and the classification ability of artificial neural network is combined in the WNN. The accuracy of face detection is more with this hybrid WNN approach. Three main issues of face recognition are pre-processing of image, feature extraction of the image and classification of the input image. The results of both the proposed images are disused to show its accuracy.