A Novel Fingerprint Image Binarization Algorithm Based on Orientation Information
一种结合方向信息的指纹图像二值化算法

Fingerprint image binarization, as one key step of fingerprint image preprocessing, is the precondition of fingerprint image thinning. In order to utilize the texture character of the fingerprint image effectively, the paper introduces the orientation information in fingerprint image binarization. After considering the fingerprint image orientation information and self-adapt local threshold synthetically, this paper brings forward a kind of self-adapt local threshold binary algorithm using the orientation information. The contrastive experimental results indicate that the presented algorithm can achieve more exact binarization results than the self-adapt binarization algorithm. Moreover the algorithm can repair the disconnected line in the fingerprint image in a certain extent. The algorithm is important for improving the minutia extraction precision.