摘要
为满足钢板表面缺陷在线检测系统宽幅面、高速、高分辨率的检测要求,讨论了基于线阵CCD的钢板表面缺陷视觉检测系统实现的关键技术;优化设计了视觉检测系统的光学照明部分,以检测不同类型的缺陷。通过软件系统的特殊设计,以保证实时在线检测。针对缺陷图像低对比度、高噪声的特点,提出了基于灰度统计特性的图像边缘检测方法,并实现了对缺陷图像的自适应阈值分割。依据图像的缺陷统计特性,定义了缺陷的灰度、几何等特征量,用于缺陷分类。本系统样机已在实验室环境下运行。
In order to satisfy the requirements in broad width, high speed, high resolution for steel strip inspection product line, the key technology on computer vision for steel strip based on linear CCD is discussed. An optical part of surface defect visual inspection system is optimally designed to detect different type defects, and software system is especially designed for real time on-line detection. The edge detection based on gray-statistical characteristic is put forward in allusion to the characteristic of low-contrast and high-noise images, and the method of adaptive threshold to split image is realized. Gray scale and geometric characteristic variables of defects are defined for defect detach according as the gray-statistical characteristic. The model machine has run in the lab.
出处
《计量学报》
CSCD
北大核心
2007年第3期216-219,共4页
Acta Metrologica Sinica
基金
国家科技部科研院所专项基金(SY01院所专项-01)
"跨世纪优秀人才支持计划"