基于Gabor函数的小波域指纹图像增强算法.pdfVIP

第26 卷第3 期 计算机应用 Voi. 26 No. 3 2006 年3 月 Computer Appiications Mar. 2006 文章编号：1001 - 9081（2006 ）03 - 0589 - 03 基于Gabor 函数的小波域指纹图像增强算法 温苗利，梁 彦，潘 泉，张洪才 （西北工业大学 自动化学院，陕西 西安7 10072 ） （wenm178@ 163. com ） 摘 要：针对指纹大规模采集库中存在的指纹图像局部区过干或过湿的问题，提出了一种基于 Gabor 函数的小波域指纹增强算法。该算法在小波域利用低频系数图估计指纹方向，从而抑制了指 纹局部过干或过湿的影响，进而分别实现基于Gabor 函数的小波域各子图增强，最终将各增强子图利 用小波逆变换实现重构。通过对FVC2004 的DB1 指纹库中的部分低质量图像的增强结果比较，该算 法对低质量指纹图像的增强效果明显，且处理速度明显快于现存的Gabor 增强方法。 关键词：指纹增强；小波变换；Gabor 函数；纹线方向 中图分类号：TP391. 4 1 文献标识码：A Gabor function-based fingerprint enhancement algorithm in wavelet domain WEN Miao-ii, LIANG Yan, PAN Ouan, ZHANG Hong-cai （College of Automation, Northwestern Polytechnical Uniuersity, Xifan Shaanxi 7 10072, China ） Abstract: In iarge-scaie fingerprint image databases, there unavoidabiy exist some over/ under inking images. A Gabor- based fingerprint image enhanced aigorithm in waveiet domain was proposed based on the idea that the dry and wet can be iooked as high-freguency disturbance. After normaiized preprocessing and waveiet decomposition, oniy approximation coefficient image was used to determine the ridge orientation, so that the effect of the dry and wet couid be restrained. Then using above estimated ridge orientation, each direction coefficient images were enhanced based on Gabor function and further reconstruction. The simuiation resuits on some typicai images in FVC2004 DB1 database show that our proposed aigorithm has better enhanced performance and iower computation burden than curr