Improved adaptive threshold algorithm for segmentation of fingerprint images
改进的指纹自适应阈值分割算法*

Based on the traditional adaptive threshold algorithm for segmentation of fingerprint images,this paper proposed an improved segmentation algorithm.According to either local gray-level-statistics of the junction regions between fingerprint and background,or regions with light or conglutinated ridges,the algorithm was capable of eliminating the blocking effect of the foreground image edges at a low computational cost.The experimental result demonstrates that the segmentation algorithm proposed is effective and robust to the pre-process of the fingerprint images.