摘要
针对带钢表面缺陷的特点,提出了一种基于图像零均值化的检测方法。首先,通过对测试图像进行零均值化,以消除光照对检测的影响;其次,利用维纳滤波对零均值化图像进行滤波除噪;在此基础上,采用Sobel进行锐化处理;最后,通过最大类间方差法进行图像分割,从而实现对带钢表面缺陷的检测。试验表明,本方法能够有效抑制图像背景干扰,有效地实现带钢缺陷的快速检测。
Through the analysis of surface defect characteristic about strip steel, a new method of based on zero- mean image processing was presented to strip steel defect detection. First of all, the test image was processed by zero-mean in order to eliminate the effect of light on the detection. Secondly, the zero-mean image was de-noised by using Wiener filtering. On this basis, the Sobel operator was used to share image target edge in order to detec- tion strip steel surface defect. Finally, the strip steel surface defect was rapidly detected by using oust image seg- mentation. The results show that the method can effectively suppress the image background interference, and more rapidly realize the detection of strip steel surface defect.
出处
《钢铁研究学报》
CAS
CSCD
北大核心
2013年第4期59-62,共4页
Journal of Iron and Steel Research
关键词
带钢缺陷
零均值化
维纳滤波
缺陷检测
strip steel defect
zero-mean
Wiener filtering
defect detection
