An Effective Algorithm for Low Quality Fingerprint Segmentation
In this paper,we have proposed a novel algorithmfor fingerprint segmentation.Firstly,we originally usethe method of gradient projection to eliminate thosebackground regions which have smaller gray contrast,and then obtain the approximate foreground region ofthe fingerprint image.In addition,we adopt thegradient coherence criteria to exclude the regionswhich contain irregular stains and smudges.Finally,we creatively carry out morphological operations onthe edge image of fingerprint to get the exactforeground region of the fingerprint,through which wecould rule out blur parts in the fingerprint image andget a fingerprint foreground which has smooth contour.Experimental results show that our algorithm iseffective and robust for most fingerprint images eventhose of low quality.
Chengpu Yu Mei Xie Jin Qi
University of Electronic Science and Technology of China Chengdu,P.R.China Post Code:610054
国际会议
厦门
英文
1081-1085
2008-11-17（万方平台首次上网日期，不代表论文的发表时间）