[1]李岩,杜永斌,宋海丰,等.ECT系统轮换对称svM图像重建改进算法[J].哈尔滨理工大学学报,2015,(03):40-044.
 LI Yan .DU Yong-bin .SONG Hai-fence.MAN Zhi-qian}.REN Xiang-h ua.Improved Method of Electrical Capacitance Tomography Based onSVM Algorithm of Cyclic Symmetrical Partition[J].哈尔滨理工大学学报,2015,(03):40-044.
点击复制

ECT系统轮换对称svM图像重建改进算法
()
分享到:

《哈尔滨理工大学学报》[ISSN:1007-2683/CN:23-1404/N]

卷:
期数:
2015年03期
页码:
40-044
栏目:
计算机与控制工程
出版日期:
2015-06-25

文章信息/Info

Title:
Improved Method of Electrical Capacitance Tomography Based on
SVM Algorithm of Cyclic Symmetrical Partition
文章编号:
1007一2683(2015)03一0040一05
作者:
李岩杜永斌宋海丰满志强任相花
(哈尔滨理工大学计算机科学与技术学院,黑龙江哈尔滨150080)
Author(s):
LI Yan .DU Yong-bin .SONG Hai-fence.MAN Zhi-qian}.REN Xiang-h ua
(School of Computer Science and Technology,Harbin University of Science and Technology, Harbin 150080,China)
关键词:
电容层析成像支持向量机轮换对称性选择分块图像重建
Keywords:
electrical capacitance tomographysupport vector machinecyclic symmetrychoice and segmentationimage reconstruction
分类号:
TP391 .4
文献标志码:
A
摘要:
针对在处理大规模样本集的ECT系统数据时,SVM算法存在的图像重建的成像精度不高及速度慢的问题,采用了轮换对称分块支持向量机CSPSVM算法.算法利用ECT系统模型的轮换对称性,将大样本矩阵按照成像单元某一层按轮换对称性进行简化,并选择性分块,形成多个小样本矩阵;然后分别采用SVM算法进行训练,用得出的决策函数进行样本预侧;最后将各成像单元组合成像.图像重建实验结果表明使用CSPSVM改进算法要比单独使用SVM算法重建图像具有更高的分类精度和更短的成像时间.
Abstract:
According to support vector machine(SVM)has low accuracy and low training speed to deal witharge scale sample matrix in ECT system,a new algorithm that combined SVM with the cyclic symmetrical partition(CSPSVM)is presented. By the cyclic symmetry of ECT system model,a large sample matrix is simplified according to a layer of the imaging unit,and segments block selectively into multiple smaller sample matrixes. Then theyare trained by SVM respectively,and the decision function obtained can be used to classify the prediction sample.Finally,all prediction units are combined for imaging. Experimental results show that the image reconstruction using CSPSVM algorithm has higher classification accuracy and shorter imaging time than using SVM alone.

相似文献/References:

[1]陈德运,李乐天,胡海涛,等.基于迭代Tikhonov正则化的电容层析成像图像重建[J].哈尔滨理工大学学报,2009,(02):1.
 CHEN De-yun,LI Le-tian,HU Hai-tao.Image Reconstruction Algorithm Based on Iterated Tikhonov Regularization for Electrical Capacitance Tomography[J].哈尔滨理工大学学报,2009,(03):1.
[2]郭虹,李岩,张礼勇,等.12电极相干电容层析成像检测系统[J].哈尔滨理工大学学报,2009,(05):31.
 GUO Hong,LI Yan,ZHANG Li-yong.12-electrodes Electrical Capacitance Tomography Detection System Based on Relevant Technology[J].哈尔滨理工大学学报,2009,(03):31.
[3]陈宇,孙帆,张健,等.三项共轭梯度的电容层析成像图像重建算法[J].哈尔滨理工大学学报,2009,(06):42.
 CHEN Yu,SUN Fan,ZHANG Jian.A Novel Three Conjugate Gradient Image Reconstruction Algorithm for Electrical Capacitance Tomography System[J].哈尔滨理工大学学报,2009,(03):42.
[4]陈宇,孙帆,张健,等.基于自适应权重粒子群的电容层析成像边界灰度补偿算法[J].哈尔滨理工大学学报,2010,(03):44.
 CHEN Yu,SUN Fan,ZHANG Jian.A Novel Gray Boundary Compensation Algorithm for Electrical Capacitance Tomography System Based on Adaptive Particle Swarm Weight[J].哈尔滨理工大学学报,2010,(03):44.
[5]张新闻,周春燕,李学生,等.优化核参数的SVM在电能质量扰动分类中的应用[J].哈尔滨理工大学学报,2011,(03):50.
 ZHANG Xin-wen,ZHOU Chun-yan,LI Xue-sheng.Recognition of Power Quality Disturbances Based on Support Vector Machine with Optimal Kernel-parameter[J].哈尔滨理工大学学报,2011,(03):50.
[6]李岩,朱艳丹,袁小花,等.基于ANSYS电容层析成像结构参数分析与优化[J].哈尔滨理工大学学报,2012,(01):54.
 LI Yan,ZHU Yan-dan,YUAN Xiao-hua,et al.Analysis and Optimization of Structural Parameters Based on ANSYS in ECT[J].哈尔滨理工大学学报,2012,(03):54.
[7]孙永倩,王培东.基于支持向量机的并行CT图像分割方法[J].哈尔滨理工大学学报,2013,(03):42.
 SUN Yong-qian,WANG Pei-dong.[J].哈尔滨理工大学学报,2013,(03):42.
[8]黄英来,孙晓芳,刘镇波,等.微博转发预测算法评测系统的建立及性能比较[J].哈尔滨理工大学学报,2013,(04):52.
[9]陈宇,许莉薇,江露,等.成像流型辨识算法[J].哈尔滨理工大学学报,2014,(04):111.
 CHEN Yu,XU Li-wei,JIANG Lu,et al.[J].哈尔滨理工大学学报,2014,(03):111.
[10]宋蕾,陈德运,姚玉梅,等.Elman神经网络在ECT系统流型辨识中的应用[J].哈尔滨理工大学学报,2014,(05):103.
 SONG Lei CHEN De-yun YAO Yu-mei LlJia-nan WANG Li-Li.Application of Elman Neural Network in Pattern Identification for Electrical Capacitance Tomography[J].哈尔滨理工大学学报,2014,(03):103.

备注/Memo

备注/Memo:
黑龙江省教育厅科学技术研究项目(}2s2}}oo};黑龙江省自然科学基金(F2o}so3s};哈尔滨市优秀学科带头人基金(2013 RFXXJ034 ).
更新日期/Last Update: 2015-08-24