低质量指纹图像分割算法研究  

Study on Low Quality Fingerprint Image Segmentation Algorithm

作　　者：周海徽[1,2] 李吉成[1] 

机构地区：[1]国防科学技术大学ATR实验室,湖南长沙410073 [2]武警上海总队第二支队,上海200030

出　　处：《计算机技术与发展》2011年第8期1-5,共5页Computer Technology and Development

基　　金：长沙市科技计划项目(K1001027-11)

摘　　要：针对低质量指纹图像的特点,提出了一种基于Canny算子和数学形态学的分割算法。首先研究了两种传统的指纹图像分割方法:基于D-S证据理论的指纹图像分割方法和结合遗传算法与方向图法的指纹图像分割方法。然后介绍了一种新的利用边缘提取和数学形态学相结合的指纹图像分割方法。该方法利用Canny算子进行粗分割,再用数学形态学的原理修正边界和去除噪声。实验结果表明,该方法对于低质量指纹图像的分割效果明显优于传统的分割方法。In allusion to the characteristics of the low quality fingerprint,a new way of fingerprint segmentation was put forward based on Canny operator and mathematical morphology.Two ways of the fingerprint image segmentation have been introduced firstly.One way is fingerprint image segmentation based on D-S evidence theory and another way is segmentation of fingerprint image using genetic algorithm and directional image method.The image was segmented by Canny operator.The noise of the image had been removed and the boundary had been modified by the morphological filter.The experimental results show that the proposed approach is superior to the traditional gray variance algorithm and gradual segmentation algorithm.

关 键 词：指纹 图像分割 CANNY算子 数学形态学 

分 类 号：TN911.73[电子电信—通信与信息系统]