手背静脉图像骨架特征提取算法.pdf

第27 卷第1 期 计算机应用 VoI. 27 No. 1 2007 年1 月 Computer AppIications Jan. 2007 文章编号：1001 - 9081（2007 ）01 - 0152 - 03 手背静脉图像骨架特征提取的算法 张晋阳，孙懋珩 （同济大学 电子与信息工程学院，上海200092 ） ( markzhang2006@ 163. com) 摘 要：手背静脉近红外图像识别是一种新的非接触式生物认证技术。对采集的手背静脉图像 进行了增强处理。对二值化图像采用四邻域区域生长的方法，去除噪声斑块。对处理后的静脉图像 采用了一种快速细化的细化算法。分析和解决了细化后特征点———交叉点的提取。针对细化后骨架 中所引入的另一类噪声———毛刺和静脉图像细化后的特点，提出了一种毛刺修复算法。实验结果表 明，经过该算法处理后得到的骨架图像，能够较好的反映静脉纹理特性。 关键词：图像细化；区域生长；毛刺；交叉点 TP391. 4 1 文献标识码：A 中图分类号： Study on algorithm for skeleton features extraction of hand vein image ZHANG Jin-yang, SUN Mao-hang ( College of Electronic and Inf ormation Engineering, Tongj i Uniuersity, Shanghai 200092, China) Abstract: Biometric identification technoIogy based on hand vein subcutaneous network structure appears as a promising technigue for personaI recognition. The contrast of hand vein image is enhanced by using Wiener fiIter, normaIized mode, gray IeveI stretching, sharpening and so on. After binary image was obtained, four-domain region growing aIgorithm was presented to differentiate and get rid of the noise areas. The guick thinning aIgorithm was adopted, which had many advantages such as compIete thinning, fast speed, maintenance of minutiae, and so on. After hand vein image skeIeton was anaIyzed, the method for extracting the crossing points was proposed to soIve the probIems met in the feature extraction. The burr, other kind of noise, was introduced b