Face Detection by Fine Tuning the Gabor Filter Parameter

Download Now Date Added: Dec 2011
Format: PDF

Face detection technique recognition is developed by using and fine tuning the Gabor Wavelet parameters. An extensive study of these parameters have been made and checked on a large acquired data set of face images, for extracting the features. The Facial data consists of 320 frontal face 180 non-face images. (Total=500 images) Artificial Neural Network is then used on the extracted features for training. An effect of pixels size of the face on the image was studied. A strong correlation between the pixels size of the Face and the Scaling parameter of Gabor filter was observed.