基于局部灰度梯度特征点的快速人耳身份鉴别

针对旋转失真的人耳图像难以识别的问题,提出一种新的人耳身份鉴别算法。首先将一幅人耳图像等分为若干个子图像,将每个子图像等分为若干个子区域;然后利用每个子图像中灰度梯度最大的子区域位置信息构筑人耳图像特征矩阵,将特征矩阵在空间上直接对准进行人耳匹配。在北京科技大学(USTB)的人耳图像库遍历的实验结果表明,通过与分块快速傅立叶变换(FFT)和分块均值标准差的人耳特征提取方法比较,提出的算法对具有±30°旋转的人耳库达到98.89%的正确识别率(CRR),说明算法对旋转失真的人耳图像具有很强的适应性,而且系统响应时间仅为21.33ms,是一种具有高实用性的有效身份鉴别方法。

A fast ear identification algorithm based on local gray gradient feature points

A new algorithm of ear identification is proposed in this paper aiming at the rotating distortion problem.First,the ear gray-scale image is divided into several sub-images,and then these sub-images are equally divided into several sub-regions.The location information in sub-region which contains the maximum gray-scale sum of all points in this region,will build the feature matrix of the ear image.At last,it will realize recognition by matching the feature matrix in space directly.We test the algorithm with the databases of USTB.The experimental results indicate that the recognition accuracy of the proposed algorithm is up to 98.89% for ±30°rotation images compared with the feature extraction methods of block Fourier transformation and block mean and standard variance,so the algorithm has more strong robustness for the rotating-distortion ear images.Furthermore,the system response time is only 21.33 ms,so it is an effective ear identification algorithm with high practicability.