针对已有静脉识别方法存在的识别结果易受图像质量影响、计算量大的问题,提出一种使用改进的方向滤波与修正的Hausdorff距离技术的指静脉识别方法.根据指静脉图像具有的方向特性设计邻域方向模板及方向滤波器模板对图像进行滤波增强处理后,利用修正的Hausdorff距离模板匹配的思想,在距离变换空间内实现指静脉图像的匹配识别.实验结果表明,该方法对指静脉图像的匹配识别具有较好的准确率.
In order to reduce the computational cost and the influence of input image quality in traditional finger vein recognition systems, a novel finger vein recognition algorithm with improved oriented filter and modified Hausdorff distance metric is presented. The proposed algorithm is based on an improved neighborhood direction template and oriented filter template which facilitates the enhancement of the finger vein image taking into full account the image orientation. This method applies modified Hausdorff distance metric to achieve the recognition of finger vein image in the distance transform space. Experimental results show that this new algorithm is fast with accurate identification.