摘要
针对工业生产中环形冲压工件图像在不同光照强度下匹配精度低和匹配速度慢等问题,提出了一种基于环形冲压工件形状特征的模板匹配算法。首先,对采集的图像进行预处理,获取模板图像的自适应参数、边缘轮廓和不同旋转角度步长下的块级强二值检测模式以及归一化梯度信息;再提取待匹配金字塔图像的轮廓信息、块级强二值检测模式和归一化梯度信息;最后,通过金字塔匹配策略进行三级匹配。实验结果表明,该算法在不同光照强度下良品工件百次平均匹配精度超过91.12%,非良品工件百次平均检出率超过82.11%,且模板建立时间低于1200 ms,匹配时间低于500 ms,检测精度和速度都达到预期效果,能够有效完成工件图像匹配任务。
Aiming at the problems of low matching accuracy and slow matching speed of annular stamping workpiece images under different light intensities in industrial production,a template matching algorithm based on the shape features of annular stamping workpieces is proposed.Firstly,the acquired images are preprocessed to obtain the adaptive parameters,edge contours,block-level strong binary detection patterns and normalized gradient information of the template image under different rotation angle steps;then the contour information,block-level strong binary detection patterns and normalized gradient information of the pyramid image to be matched are extracted;finally,three-level matching is carried out by the pyramid matching strategy.The experimental results show that the algorithm has an average matching accuracy of more than 91.12%for 100 times for good workpieces and an average detection rate of more than 82.11%for 100 times for non-good workpieces under different light intensities,and the template establishment time is lower than 1200 ms and the matching time is lower than 500 ms,so the detection accuracy and speed have reached the expected effect,and it can effectively complete the workpiece image matching task.
作者
卞跃超
孙付春
吴昊荣
高蕾娜
李晓晓
BIAN Yuechao;SUN Fuchun;WU Haorong;GAO Leina;LI Xiaoxiao(School of Mechanical Engineering,Chengdu University,Chengdu 610106,China;School of Electronic Information and Electrical Engineering,Chengdu University,Chengdu 610106,China)
出处
《组合机床与自动化加工技术》
北大核心
2025年第5期48-54,共7页
Modular Machine Tool & Automatic Manufacturing Technique
关键词
环形冲压工件
光照强度
模板匹配
轮廓信息
梯度信息
ring stamping workpiece
light intensity
template matching
contour information
gradient information
