一种基于视觉的表面质量检测方法

结合边缘检测技术和数学形态操作,提出了一种基于视觉的铝带表面检测方法。应用中值算法滤除缺陷图像噪声后,用边缘算子提取缺陷边缘,经形态学处理后得到完整缺陷目标,然后提取缺陷的形态特征,进行缺陷分类。实验结果表明:这种方法不仅能有效地识别缺陷,还能准确地判别缺陷类型和缺陷位置。

Approach on Surface on Quality Inspection Based on Vision

A visual surface inspection approach using edge detection and morphological processing algorithm is presented.Median filter,edge detection segmentation and morphological operation algorithms are used to reduce noise,extract edges,and extract morphological features respectively.Finally,a classifier is designed to judge the type of defects.Experimental results show that the proposed method can not only identify detects effectively,but also obtain the type and location of defects accurately.