Abstract: An ear detection algorithm is proposed in order to make image matching accurate.The algorithm gives weighted inside and outside contour and uses Hausdorff distance to detect ear image.If image has noises or outside ear contour is not continuous，the result of detection is not good by using classical Hausdorff distance.For making the position of detection close to the outside ear contour，the outside ear contour needs to be emphasized.It is realized by giving larger weight for outside ear contour than inside ear contour.And then weighted contour combining with weighted Hausdorff distance is to match ear image.Experiments show that the algorithm proposed in this paper is effective.Compared with classical Hausdorff distance and average Hausdorff distance，the result of weighted contour algorithm is more precise.

Key words: Hausdorff distance, human ear detection, image matching, weighted contour

摘要： 为了提高人耳检测中图像匹配的精确性，提出对内外耳轮廓加权，并利用Hausdorff距离进行人耳检测的算法。在传统的Hausdorff距离匹配中，图像如果受噪声干扰或边缘不连续等情况，检测结果不理想。因此为使检测位置更加接近外耳轮廓，需要强调外耳轮廓的作用，这通过对外耳加大权值、对内耳加小权值实现，然后再结合加权Hausdorff距离进行图像匹配计算。仿真实验表明，提出的算法是有效的。相比传统Hausdorff距离和平均Hausdorff距离的匹配，人耳轮廓加权的算法更加精确。

关键词: Hausdorff距离, 人耳检测, 图像匹配, 轮廓加权