Determining optimal Malsburg Gabor kernel for efficient non-rigid object recognition

Human face detection and recognition are still challenging questions in pattern recognition field. The facial features such as eyes, nose and mouth are detected in an image which contains a face and the rectangular area surrounding facial features is obtained. To achieve this, Gabor wavelet is the field of interest for many face/object recognition researchers. In this paper, we proposed the adjustable Malsburg Gabor kernel for mouth feature of face image. Mouth feature of face is under the prone effect of facial expression due to variation (noise) and largely affects face recognition system. We improve the Gabor wavelet Kernel for robust face recognition to be adaptable to the mouth. We enlarged the rate of the length to the edge of the kernel because the teeth become interference (noise) for the face recognition. We have designed, implemented and demonstrated our proposed approach addressing this issue. FERET face image dataset have been chosen for training and testing and we have achieved a very good success.