| 一种新的浮选泡沫图像识别方法 |
| |
| 引用本文： | 郝元宏,韩静,齐春.一种新的浮选泡沫图像识别方法[J].西安交通大学学报,2011,45(4):104-108. |
| |
| 作者姓名： | 郝元宏 韩静 齐春 |
| |
| 作者单位： | 1. 西安交通大学电子与信息工程学院,710049,西安
2. 山西大学商务学院理学系,030031,太原 |
| |
| 摘 要： | 针对灰度共生矩阵法提取的浮选泡沫图像纹理特征相互混叠,不利于聚类和识别的问题,提出一种基于正交保局投影和支持向量机的浮选泡沫图像识别新方法.该方法利用正交保局投影法对原始纹理特征参数进行变换处理,有效改变了不同类别特征参数的聚集程度,并利用支持向量机进行分类.实验结果表明,所提方法的正确识别率能够达到93.5%,与基于最近邻分类器的主元分析法相比,其性能更好.
|
| 关 键 词： | 浮选 泡沫图像 机器视觉 正交保局投影 支持向量机 |
A New Recognition Method for Flotation Froth Images |
| |
| HAO Yuanhong,HAN Jing,QI Chun.A New Recognition Method for Flotation Froth Images[J].Journal of Xi'an Jiaotong University,2011,45(4):104-108. |
| |
| Authors: | HAO Yuanhong HAN Jing QI Chun |
| |
| Institution: | 1(1.School of Electronics and Information Engineering,Xi′an Jiaotong University,Xi′an 710049,China; 2.Faculty of science,Business College of Shanxi University,Taiyuan 030031,China) |
| |
| Abstract: | A novel approach to classify the flotation froth images based on OLPP and support vector machine is presented to deal with the problem that the texture feature vectors of digitized images of the froth obtained from the use of gray-level dependence matrix method can not provide a compact representation of froth properties.The texture feature vectors from OLPP is transformed,and then,classification is investigated by using a multi-class SVM.Results obtained encourage to apply the approach to develop a classif... |
| |
| Keywords: | flotation froth image machine vision orthogonal locality preserving projection support vector machine |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
