基于高斯平滑的自适应角点检测.PDFVIP

第 卷 第 期 计算机辅助设计与图形学学报 ， l5 ll Voi . l5 No . ll 年 月 ， 2003 ll JOURNAL OF COMPUTER-AIDED DESIGN COMPUTER GRAPHICS Nov . 2003 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 基于高斯平滑的自适应角点检测 肖 茜 鲁宏伟 （华中科技大学计算机科学与技术学院 湖北 430074 ） 摘 要 文中算法首先在曲线尺度空间中通过高斯平滑以消除噪声；然后再基于自适应弯曲度计算和角点筛选准 则确定角点. 该算法不需要预先输入参数，具有较好的抗干扰性，实现简单有效. 关键词 高斯平滑；尺度空间；自适应；角点检测 中图法分类号 TP39l Adaptive Corner Detection Based on Gaussian Smoothing Xiao Oian Lu Hongwei （ ， ， ） College of Computer Science and Technology Huazhong Uniuersity of Science and Technology Wuhan 430074 Abstract In image processing，corner point means the dominant point of maximum curvature aiong the bounding edge of pianar object . A method of detecting and iocaiizing corners of pianar curvature based on Gaus- sian scaie space is presented . An adaptive curvature estimate method measures the significance of each point on ； the boundary preprocessed by Gaussian smoothing then a criterion is estabiished to choose the desired corner ， points . The method reguires no input parameter and experiments were performed to show that the scaie space corner detector based on Gaussian smoothing is reiiabie for objects with muitipie-size features and noisy bound- aries . ； ； ； Key words Gaussian smoothing scaie space adaptive corner detection 点求取算法，这些算法的优点是实现简单且运算速 1 引 言 度快，但都需要预先确定相应的