基于改进高斯混合模型的矿用输送带纵向撕裂检测方法  

Detection Method of Longitudinal Tear for Mine Conveyer Belt Based on Improved Gaussian Mixture Model

作　　者：郭健 乔铁柱 车剑 GUO Jian;QIAO Tiezhu;CHE Jian(Key Laboratory of Advanced Transducers and Intelligent Control System,Ministry of Education and Shanxi Province,Taiyuan University of Technology,Taiyuan 030024,China)

机构地区：[1]太原理工大学新型传感器与智能控制教育部重点实验室,山西大原030024

出　　处：《煤矿安全》2020年第12期167-170,共4页Safety in Coal Mines

基　　金：国家自然科学基金资助项目(201801D121180);NSFC-山西煤基低碳联合基金资助项目(U1810121)。

摘　　要：提出一种红外图像特征与改进高斯混合模型(Gaussian Mixture Model,GMM)相结合的矿用输送带纵向撕裂在线检测方法。设计了一种自适应混合中值滤波技术;针对高斯混合模型初始化易错的缺点,采用加权可选择模糊C-均值改善这一问题,把红外图像特征参数作为改进的GMM聚类的特征向量进行聚类分析,实现了矿用输送带纵向撕裂识别。试验结果表明:检测方法对输送带纵向撕裂检测的正确检测率可达99%。An online detection method for longitudinal tear of mine conveyor belt is proposed,and the method combines infrared image features with improved Gaussian Mixture Model(GMM).An adaptive hybrid median filtering technique is designed.In view of the error prone initialization of Gaussian Mixture Model,the problem is improved by using the weighted alternative fuzzy C-means,the infrared image feature parameters are used as eigenvectors of improved GMM clustering for cluster analysis to realize the longitudinal tear identification of mine conveyor belt.The experimental results show that the correct recognition rate for the longitudinal tear detection of conveyor belt can reach 99%with this method.

关 键 词：输送带纵向撕裂检测 自适应混合中值滤波 改进的高斯混合模型 加权可选择模糊C-均值算法 红外图像特征 

分 类 号：TD679[矿业工程—矿山机电]