摘要
钢板的表面缺陷是影响钢板质量的主要因素,通过改进轧制工艺可以减少缺陷发生外,及时检测出钢板的表面缺陷也非常重要.对于钢板表面缺陷的检测,需要获取图像,然后对图像进行初步处理,重要的步骤就是对缺陷进行分割.基于图像灰度信息的不同,本文采用了两种图像分割模型(C-V模型和H-T-B模型),当图像的灰度信息均匀时,采用C-V模型对图像进行分割;当图像的灰度信息不均匀时,则采用H-T-B模型对图像进行分割.通过两种模型的组合应用可以对钢板的各类表面缺陷进行识别,获取缺陷区域,有利于提高钢板生产质量.
The surface defect of steel strip is an important factor affecting the quality of steel sheet. In addition to the improvement of rolling process to reduce the occurrence of defects, it is impor- tant to detect the surface defects of the steel plate. For the detection of surface defects of the steel strip, we need to obtain the image. Then the image is processed, and the most important step is to segment the defects. In this paper, two image segmentation models （C-V model and H-T-B model） are used in this paper based on the different gray information. When the gray information of the image is uniform, the C-V model is used to segment the image. When the gray information of the image is not uniform, the H-T-B model is used to segment the image. Through the combi- nation of the two models, various types of surface defects of the steel plate can be identified, and the defect area can be obtained. It is conducive to improve the quality of steel plate production.
出处
《北京交通大学学报》
CAS
CSCD
北大核心
2016年第2期47-52,共6页
JOURNAL OF BEIJING JIAOTONG UNIVERSITY
基金
江西省青年科学基金资助项目(20132BAB216028)
江西省教育厅科技项目资助(GJJ150530)
