自动找点的掌纹图像定位分割方法许爽.pdfVIP

计算机工程与设计 Computer Engineering and Design 3102 2011, Vol.32, No.9 计算机工程与设计 Computer Engineering and Design 自动找点的掌纹图像定位分割方法 1,2 1 2 许 爽 ， 索继东 ， 赵继印 (1. 大连海事大学信息科学技术学院，辽宁 大连 116026 ； 2. 大连民族学院信息与通信工程学院，辽宁 大连 116600) 摘 要：重点研究掌纹图像在产生旋转和平移情况下的定位分割方法。对采集的掌纹图像做中值滤波后得到二值图像，跟 踪二值图像边界，转换边界点重新建立只显示边界新图像，最后分析手掌形态，分离手指与手掌，利用去孤立点的自动找点 方法提取食指与中指、无名指与小拇指两处定位点，并以此对图像做旋转校正，建立直角坐标系，确定图像的感兴趣区域。 实验结果表明，该方法的定位准确率高，有利于进一步对掌纹图像进行特征提取和特征匹配。 关键词：掌纹图像; 定位; 分割; 二值图像; 感兴趣区域 中图法分类号：TP391.4 1 文献标识码：A 文章编号：1000-7024 (2011) 09-3102-04 Palmprint image positioning and segmentation methods based on auto finding points 1,2 1 2 XU Shuang , SUO Ji-dong , ZHAO Ji-yin (1. College of Information Science and Technology, Dalian Maritime University, Dalian 116026, China; 2. College of Information and Communication Engineering, Dalian Nationalities University, Dalian 116600, China) Abstract ：The positioning and segmentation of palmprint image is discussed in detail, which involves rotation and transition. Firstly, palmprint image used median filter is converted binary image. Then the boundary of binary image is tracked and a new image is recreated, which shows only the border. Finally, the fingers and palms are separated and the two positioning points of the index finger and middle finger, ring finger and little finger are found. Thus, the palmprint image is rotated and corrected based on the two positioning points ( ) and coordinate system is created according to positioning points,