采用旋转校正的指静脉图像感兴趣区域提取方法  

A region of interest extraction method using rotation rectified finger vein images

作　　者：马慧 王科俊[2] 

机构地区：[1]黑龙江大学电子工程学院,黑龙江哈尔滨150080 [2]哈尔滨工程大学自动化学院,黑龙江哈尔滨150001

出　　处：《智能系统学报》2012年第3期230-234,共5页CAAI Transactions on Intelligent Systems

基　　金：国家自然科学基金资助项目(60975022);国家"863"计划资助项目(2006AA04Z248);黑龙江大学青年基金资助项目(QL201111)

摘　　要：针对静脉图像采样过程中存在的旋转、平移等非线性因素造成手指静脉图像定位困难的问题,考虑图像非接触式采集特点,提出一种采用旋转校正的手指静脉图像感兴趣区域提取方法.首先对读入的手指静脉图像采用Kapur熵阈值法分割出手指区域,再依据图像的质心对图像进行旋转校正,最后根据图像中每列像素竖直方向上的投影值和手指区域的边缘轮廓,确定出感兴趣区域的位置.实验结果表明,该方法能够准确地提取出静脉图像的感兴趣区域,有效地提高识别系统的性能.In order to reduce the influence of nonlinear translation and rotation on the positioning of finger vein images in the process of vein image sampling,a region of interest extraction method that utilizes a rotation rectified finger vein image was proposed.The method took account of the non-contact collecting characteristics.First,the finger regions of finger vein images were extracted using the Kapur entropy threshold method.These images were then rotated along their centroids;finally,the regions of interest were extracted according to the vertical projection value of every column pixel and the outline of the finger regions.Experimental results show that this algorithm can not only accurately extract the regions of interest of finger vein images,but also effectively improve the performance of the vein recognition system.

关 键 词：静脉图像 手指静脉 感兴趣区域提取 旋转校正 图像分割 静脉识别 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]