GLCM和NGLDM特征提取方法在锑浮选泡沫图像上的比较  

Comparison of Feature Extraction Methods Between GLCM and NGLDM on Froth Image of Antimony Floatation

作　　者：彭霞 张向华[1,2] 胡鹤宇 陈意军[1] 郝伟涛[3] PENG Xia;ZHANG Xiang hua;HU He -yu;CHEN Yi -jun;HAO Wei tao(College of Electrical and Information Engineering, Hunan Institute of Engineering, Xiangtan 411104, China;The Key Laboratory of Hunan Province on Wind Turbine Generator Set and Contro Hunan Institute of Engineering, Xiangtan 411104, China;College of Electrical ＆ Information Engineering, Changsha University of Science ＆ Technology, Changsha 410114, China)

机构地区：[1]湖南工程学院电气信息学院,湘潭411104 [2]湖南工程学院风力发电机组及控制湖南省重点实验室,湘潭411104 [3]长沙理工大学电气与信息工程学院,长沙410114

出　　处：《湖南工程学院学报（自然科学版）》2018年第3期17-22,共6页Journal of Hunan Institute of Engineering(Natural Science Edition)

基　　金：湖南省自然科学基金青年项目(2017JJ3050);湖南工程学院青年科研重点项目(XJ1502)

摘　　要：泡沫图像特征与工况紧密相关,对于指导生产操作,调节生产状态发挥着重要作用.纹理特征是泡沫图像中的关键特征,它能表征泡沫表面纹理粗细、形态、灰度等.分别采用GLCM和NGLDM提取锑浮选泡沫图像纹理特征,采用最近邻分类器对工况进行离线识别.结果表明,GLCM具有相对较高的分类正确率,而NGLDM由于具有远优于前者的处理速度更适用于在线工况识别.The feature of the froth image is closely related to the working condition, which plays an impor tant role in gui key feature of f ding roth the production operation and regulating the production state. The texture feature is the image, which can characterize the roughness, shape and gray level of the froth surface In this paper, GLCM and NGLDM are used to extract the texture features of the froth image of antimon floatation, and the off line recognition of the working condition is carried out by the nearest neighborhoo classifier. The results show that, though GLCM has a relatively higher classification accuracy, NGLDM is more suitable for on line working condition recognition for the high speed in processing .

关 键 词：GLCM NGLDM 特征提取 锑浮选 泡沫图像 

分 类 号：TP274.3[自动化与计算机技术—检测技术与自动化装置]