一种电容层析成像局部方差的图像融合方法  

Image Fusion Method for Electrical Capacitance Tomography System

作　　者：于金平[1] 陈德运[1] 张云龙[1] 吉洪柱 王莉莉[1] 

机构地区：[1]哈尔滨理工大学测控技术与仪器黑龙江省高校重点实验室,黑龙江哈尔滨150080

出　　处：《哈尔滨理工大学学报》2015年第4期116-121,共6页Journal of Harbin University of Science and Technology

基　　金：国家自然科学基金(60572153;60972127);高等学校博士学科点专项科研基金(200802140001);黑龙江省自然科学基金(QC2012C059);哈尔滨市科技创新人才研究专项资金(2014RFXXJ022);黑龙江省教育厅科学技术研究项目(11541040;12511097)

摘　　要：针对电容层析成像ECT(electrical capacitance tomography)系统的"软场"效应和病态特性及欠定性的影响,依据电容层析成像原理和成像算法的原理,提出一种基于局部方差的电容层析成像系统图像融合方法.该方法以线性反投影算法、共轭梯度算法和信赖域算法为图像重建基础,利用各个图像的互补特性,将重建图像进行小波分解后的高频分量和低频分量分别采用局部方差和加权平均融合规则融合,最后通过小波逆变换得到新的重建图像.仿真实验结果表明:融合后的图像精确度得到提高,误差变小,图像更接近原型,为ECT图像重建的研究提供了一个新的方法.As for electrical capacitance tomography ECT （Electrical Capacitance Tomography） system is the ＂ soft field＂ effect and pathological characteristics and less qualitative effects, based on the principle of electrical ca- pacitance tomography imaging principle and imaging algorithm, proposes an image fusion method that based on the local variance for electrical capacitance tomography system. This method takes the linear back projection algorithm, conjugate gradient algorithm and the trust region algorithm all becoming foundation of image reconstruction. Be- cause of complementary characteristics of each image, the reconstructed images after wavelet decomposition of the high frequency component and low frequency component respectively using local variance and weighted average fu- sion rule fusion. Finally, by using the inverse wavelet transforms to obtain a new image reconstruction. The simula- tion experiment results show that the accuracy of image fusion after improved, the error becomes small, the image is closer to the prototype, provides a new method for the research of ECT image reconstruction.

关 键 词：电容层析成像系统 图像重建算法 图像融合方法 局部标准方差 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]