手指背关节皮纹识别方法

Method of finger-back articular skin texture recognition

利用自主设计的采集装置获得手背图像,由Canny算子分割并定位手指背关节皮纹,根据卷积定理,对要测试的两背关节皮纹进行快速配准,以相关分类器作为判据,检验两者是否为同一模式.在小型数据库上的试验表明:手指背关节皮纹识别具有很好的稳定性和可靠性,在等错误拒绝率和等错误接受率情况下,识别准确率初步可高达95.58%.

The hand back image is captured by the device made by ourselves. The articular skin texture is segmented and located by utilizing Canny operator. According to convolution theorem, the fast match algorithm of two articular skin textures is given. The correlation classifier is used to judge whether the two articular skin textures are the same pattern. The experiment results on a small database show that the pattern has good reliability and stability and the identify rate is 95.58 percent as FAR equals to FRR.