机构地区: 广东工业大学信息工程学院
出 处: 《光学与光电技术》 2006年第5期101-103,共3页
摘 要: 为了更有效地从二值图像中提取更高质量的曲线轮廓,提出一种基于邻域统计编码的轮廓提取算法,它能在快速地提取出待测零件轮廓的同时,滤除毛刺和噪声。在检测精度不变的情况下,具有速度高和鲁棒性较好的特点。 Getting the boundary from a black-white image is very important in CCD 2D multi-measurement testing. In order to get higher quantity, this paper presents neighborhood statistic coding algorithm to get the boundary of the parts rapidly, filter the burrs and noise at the same time. The experiments show that the algorithm has quite high speed and robust at the same testing accuracy.