Wavelet and Cellular Automata Based Image De-noising Method for Wire Bonding

According to the characteristics of wire bonding image and to meet the requirement for strong noise rejection, a de-noising algorithm based on wavelet and cellular automata is presented in this paper. Evolution rules are given by using the direction information and edge orderliness of the pixels. The accurate noise information can be detected by automatic evolution of cellular automata and then the de-noised image is obtained by reconstruction from the processed coefficients. The algorithm can effectively eliminate the image noise and keep edge information without blurring image edge. The algorithm especially suits for the wire bonding image which need high edge detection accuracy. The simulation results show the effectiveness of the proposed algorithm. The algorithm improves the visual quality of the image and presents much higher peak signal to noise ratio compared with traditional method.