Face recognition based on kirsch compass kernel operator

A face recognition system is used for identifying or validating an individual from a digital image or a video frame from a video source. Images and face database are used for comparing selected facial features. It is typically employed in security systems and can be compared to alternative biometrics like finger print or eye iris recognition systems. Recently, it has also becomes standard as a commercial identification and promoting tool. In this paper, we proposed a new methodology for clustered micro calcifications based on edge detection of kirsch edge operator and edge linking. First kirsch edge operator extracts initial edges and then the thin initial edges extracts since it is used as a derivative mask. It fills many gaps in the edge image so that recognition rate is improved comparatively LBP and its derivative methods.