Research on preprocessing of palmprint image based on adaptive threshold and Euclidian distance

The approach of combining the characteristics of palmprint image database with an improved adaptive binary segmentation which is based on gray-scale distribution and structure of the palmprint image is presented in this article. We use median filter to remove the cross and obscure parts of the breakpoint, take advantage of an improved global threshold algorithm to get the binarization image and carry out corrosion, swelling, edge extraction and boundary tracking. Then we use the Euclidian distance to get the region of interest(ROI) automatically. The simulation experiment results show that the method can extract the ROI of palmprint image effectively and eliminate the noise in palmprint image.