摘要
通过对钢坯表面裂纹的形状特征的分析,运用图像增强、图像分割、边缘检测、特征提取及识别等方法,提出了一个更加精确、快速的钢坯表面裂纹识别算法。通过边缘检测及链码标识,可以得到更细致的裂纹特征,进行更精准和深入的分析。计算机通过阈值分割和边缘检测对图像的对象和背景进行分割,抽取图像的特征区域,对该区域的灰度值进行比较和分析,找出附近灰度值显著不同的面积区域。对于加强钢坯生产过程中的检测和控制,提高企业经济效益,可以有较大帮助。
By analysing the surface feature of the flaw in the steel billet,and by using the methods of image enhancing、image segmentation、edge detection、feature extraction and recognition,this paper introduced an algorithm to recognize the flaw on the surface of the steel billet which is faster and more accurate.By edge detection and chain code identifying,test can get more intensive surface feature,and do analysis deeper and more precisely.By threshold segmentation and edge detection,the computer can separate the object and the background,and by extracting the feature areas of the image and analyzing the gray values,the computer can find out the areas with distinctly different gray values.It＇s very helpful to reinforce the detection and control of the production process and to increase the economical benefits of the company.
出处
《机械设计与制造》
北大核心
2013年第1期84-85,88,共3页
Machinery Design & Manufacture
基金
湖北省教育厅重点项目(D20101128)
关键词
图像识别
表面裂纹检测
特征提取
识别算法
Image Recognition
Surface Flaw Detection
Feature Extraction
Recognition
Algorithm
