首页 | 本学科首页   官方微博 | 高级检索  
     

矿坝输送带实时图像探伤方法
引用本文:liukai,牛昱光. 矿坝输送带实时图像探伤方法[J]. 电视技术, 2014, 38(7): 43-45,37
作者姓名:liukai  牛昱光
作者单位:太原理工大学,太原理工大学
基金项目:山西省科技攻关项目(20110321036)
摘    要:长期工作在艰苦的环境下,钢丝绳芯输送带易发生断裂故障,对运输生产造成影响。提出一种基于模糊聚类算法的图像识别方法对输送带的损伤与接头进行检测,据背景差分法与阈值分割,分离出输送带前景、背景视频图像得到目标。利用Sobel算子检测得到损伤与接头处边缘关系,消除冗余信息,最后用模糊聚类算法对图像分类,从而对输送带是否有损伤作出判断。实验结果用以验证算法的有效性。

关 键 词:Sobel边缘检测  输送带损伤检测  模糊聚类  模板匹配
收稿时间:2013-03-27
修稿时间:2013-05-10

Mine conveyor belt inspection method of real-time image
Affiliation:Taiyuan University of Technology,Taiyuan University of Technology
Abstract:Steel cord conveyor belt due to the long-term work in harsh conditions, the wire rope is prone to fracture, affecting transport safety. In this paper, based on fuzzy clustering algorithm with template matching method to detect damage to the conveyor belt.. In order to get the moving target segmentation, isolated the conveyor belt prospects from background video image based on difference method with threshold. Sobel operator edge detection breaks edge relationship, remove redundant information with fuzzy clustering algorithm for pixel classification. Finally, the use of template matching to determine the conveyor belt fracture failure. Proposed algorithm based on the OpenCV library functions, and the experimental results verify the effectiveness of the proposed algorithm.
Keywords:Sobel edge detection   conveyor belt damage detection   fuzzy clustering   template matching
本文献已被 CNKI 等数据库收录!
点击此处可从《电视技术》浏览原始摘要信息
点击此处可从《电视技术》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号