基于小区域模板匹配的发动机缸体缺陷检测  

Defect detection of engine cylinder based on small region template matching

作　　者：殷苏民[1] 鲍红力[1] 吉彬斌[1] 刘金亮[1] 张建刚[1] 

机构地区：[1]江苏大学机械工程学院,江苏镇江212013

出　　处：《传感器与微系统》2012年第6期143-145,共3页Transducer and Microsystem Technologies

基　　金：国家自然科学基金资助项目(51179076)

摘　　要：针对发动机缸体人工检测效率低不可靠的问题,提出一种基于小区域模板匹配的检测算法来实现缸体的在线自动检测。将发动机缸体的特征图像制作成小区域模板库,在发动机缸体特定小区域使用粗精结合搜索策略的改进序贯相似度算法确定待匹配图像的准确位置和相似度,从而实现发动机缸体缺陷检测。实验结果表明:该算法能够满足发动机缸体缺陷检测要求,基本消除了人工检测的误判,显著降低了人工检测时间。Aiming at the problem of lower efficiency by manual detection,a detection of engine cylinder algorithm for the on-line automatic based on small region template matching is presented.The engine cylinder image is made into a small region image template library,and coarse and fine searching strategy combined method with improved sequence similarity detection algorithm is used to determine the exact location and similarity at the particular small region.The purpose of defect detection is achieved.Experimental results show that the algorithm is able to meet the real-time requirments for engine cylinder detection and eliminate the misjudgement of manual detection.

关 键 词：发动机缸体 灰度匹配 小区域模板 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置] U464[自动化与计算机技术—控制科学与工程]