【作者】 王凤随； 李素文； 冯桂；

【Author】 Wang Feng sui~1 Li Su wen~1 Feng Gui~2 (1 Department of Physics and Electronics Information Huaibei Coal Industry Teachers’ College, Huaibei Anhui,235000,P.R.China;2College of Information Science and Engineering,Huaqiao University,Quanzhou Fujian 362021,P.R.China)

【机构】 淮北煤炭师范学院物理与电子信息学院； 华侨大学信息科学与工程学院；

【摘要】 虹膜纹理上的差异是不同虹膜相互区别的最主要原因和体现。因此,要想获得很高的虹膜识别率,就要有效地提取虹膜的纹理信息。通过对虹膜纹理单元的分析,提出一种基于像素方向相关性和局部沃尔什变换提取虹膜纹理特征的新方法。在求取归一化虹膜图像中每个像素的局部沃尔什变换系数的基础上,在纹理单元内估计变换系数的各阶矩能量测度作为虹膜的特征向量。实验结果表明,算法能够有效地提高虹膜识别率,可达99%以上,且算法简单、快速。更多还原

【Abstract】 Irises can be distinguished mainly due to textural difference among them.In order to acquire a high recognition rate,the textural information of iris images must be extraction effectively. A novel iris texture feature extraction algorithm based on pixel’s directional relativity and local walsh transform was presented according to analyze iris texture units.For each pixel in normalized iris images,the local walsh transform coefficients were computed,and their order moments were estimated as the iris’s texture features in texture unit.According to experiment,the algorithm can enhance iris recognition rate,achieving above 99 percent.Furthermore,algorithm is simple and fast.更多还原