Abstract: Because the wavelet transform has different focus for different scale layer, the choice of the threshold in Mallat’s multiscale edges extraction algorithm based on wavelet is improved. According to characters of the micro images one can set different threshold values for cell edge and background noise by using the self adaptive time-frequency feature of the wavelet. Then the cell edge can be effectively determined by separating the modulo angle. Experiment results of choosing different threshold value show that it is effective.

Key words: medical image processing, edge detection, wavelet transforms