计算机工程与应用 ›› 2013, Vol. 49 ›› Issue (13): 110-113.

• 数据库、数据挖掘、机器学习 • 上一篇    下一篇

选择分块SVM电容层析成像改进方法

李  岩,袁小花,刘精松,柳培新,郑洁琼,张  迪   

  1. 哈尔滨理工大学 计算机学院,哈尔滨 150080
  • 出版日期:2013-07-01 发布日期:2013-06-28

Improved method of electrical capacitance tomography based on SVM algorithm of choice and segmentation

LI Yan, YUAN Xiaohua, LIU Jingsong, LIU Peixin, ZHENG Jieqiong, ZHANG Di   

  1. College of Computer Science and Technology, Harbin University of Science and Technology, Harbin 150080, China
  • Online:2013-07-01 Published:2013-06-28

摘要: 针对SVM在处理具有样本集规模大的ECT系统数据时,存在ECT图像重建的成像精度不高和速度慢的问题,采用了选择分块支持向量机CSSVM算法。将ECT系统样本数据构成列数固定的样本矩阵,每个样本作为样本矩阵的行,66个电容值和66个敏感度值作为矩阵的列。该算法将大样本矩阵按照某一成像单元进行选择性分块,并形成多个小样本矩阵,再分别采用SVM算法进行训练和预测,将各个成像单元组合成像。数值实验证明,使用CSSVM新算法比单独使用SVM算法重建图像具有更高的分类准确率和更短的成像时间。

关键词: 支持向量机, 选择分块, 电容层析成像, 数据预处理, 图像重建

Abstract: According to Support Vector Machine(SVM) has low training speed and low accuracy to deal with large scale data in Electrical Capacitance Tomography(ECT) system, a new algorithm that combined SVM with the Choice and Segmentation(CS) is presented and it comes into being a new classifier. Data in ECT system composes a data matrix which is fixed matrix column componented of sixty-six capacitance values and sixty-six sensitivity, the samples as its rows.It divides block selectively from large scale samples for one imaging unit. The numerical experiments show that the mixed algorithm can not only improve the accuracy compared to sole SVM, but also shorten time in imaging.

Key words: Support Vector Machine(SVM), Choice and Segmentation(CS), Electrical Capacitance Tomography(ECT), data preprocessing, image reconstruction