机构地区: 五邑大学信息学院
出 处: 《工业控制计算机》 2008年第7期24-24,26,共2页
摘 要: 首先介绍了基于机器视觉的印刷品在线检测系统的结构和工作原理,接着分析了实际系统中标准模板的建立等关键问题,提出了用邻域比较法来解决由于软性印刷品材料的拉伸变形给系统带来的误检问题,最后用MMX技术对程序进行优化加速。实际的工程应用中表明这些方法是有效的,满足系统设计的要求。 Firstly,this paper introduces structure and working principle of online printed matter defect detection system.Then the problem of how to establish standards image template are analyzed.The soft print material's stretch elongation brings to the detection by mistake,in order to solve this problem,the author propose neighborhood comparison algorithm.Finally,The procedure is optimized by MMX technology.Experiments show that the system is fast and effective.