基于改进PSO的医学图片颜色迁移及重构
|
摘要
通过对真彩图像与灰色MRI切片的亮度和纹理等特征的配准而进行灰色图像的彩色化,可以得到有效的而且符合真实人体器官组织的真彩仿真图像。着重研究将粒子群算法与点邻域亮度纹理分布和局部颜色结构分布相结合而应用于Welsh图像彩色化算法。通过粒子群寻优迭代方程的各参数的自适应调节和控制提升其性能,并将点邻域分布的颜色特征和彩色局部分布的颜色特征通过多种搜索算法如遗传算法和粒子群算法及其修正算法进行了适应度函数最优值搜索分析比较,其彩色化的MRI切片的三维重构模型能多层次清晰地反映器官组织的分布和构造。
Researches transferring a gray MRI slice into a colorful one that is valid and matches the human organizations by registering the features of brightness and textures of a true color image and a gray image.The Welsh color transferring algorithm is particularly probed based on the combination of particle swarm optimization and the feature of brightness within bit neighborhood and local color distribution.An improved particle swarm optimization(PSO) algorithm by self-adapting regulating and control of parameters of the optimal formula to promote the performance.A pixel's color neighbor distribution and local color structure distribution properties by exhaustion,random selection,genetic algorithm(GA),PSO and improved PSO and their integration for a fitness optimum searching and their analysis and comparison.Reconstructed 3D color model from the colored slices could clearly reveal multilevel structure and distribution.
Researches transferring a gray MRI slice into a colorful one that is valid and matches the human organizations by registering the features of brightness and textures of a true color image and a gray image.The Welsh color transferring algorithm is particularly probed based on the combination of particle swarm optimization and the feature of brightness within bit neighborhood and local color distribution.An improved particle swarm optimization(PSO) algorithm by self-adapting regulating and control of parameters of the optimal formula to promote the performance.A pixel's color neighbor distribution and local color structure distribution properties by exhaustion,random selection,genetic algorithm(GA),PSO and improved PSO and their integration for a fitness optimum searching and their analysis and comparison.Reconstructed 3D color model from the colored slices could clearly reveal multilevel structure and distribution.
引文
[1]LIU Yong,SHAO Changyong.Improvement of Welsh's grey imagecolorization arithmetics[J].Modern Electronics Technique,2009(24)141-143(in Chinese).[刘勇,邵常勇.对Welsh灰度图像彩色化算法的改进[J].现代电子技术,2009(24)141-143.]
[2]CHEN Qing.Improvement of Welsh's grey image colorization arith-metic[J].Journal of Wuhan University of Technology,2009,31(22)151-154(in Chinese).[陈倩.一种改进的Welsh灰度图像彩色化算法[J].武汉理工大学学报,2009,31(22)151-154.]
[3]LIU Xiaopei,WAN Liang,LIN S,et al.Intrinsic colorization[J].ACM Trans on Graphics,2008,27(5):1521-1529.
[4]XU Zhen,SUN Shaoyuan,DAI Zhonghua,et al.Texture librarybased color&texture transferring for LLL images[J].InfraredTechnology,2011,33(1):49-55(in Chinese).[许真,孙韶媛,代中华,等.基于纹理特征库的微光图像色彩纹理传递[J].红外技术,2011,33(1):49-55.]
[5]ZHENG Zhi,HUANG Jian.Analysis of the application mode of col-or transfer algorithm in images[J].Journal of Fujian Agricultureand Forestry University,2010,39(2):206-210(in Chinese).[郑之,黄键.图像色彩传递算法的应用模式分析[J].福建农林大学学报,2010,39(2):206-210.]
[6]FENG Jun,XUE Yuncan,JIANG Jinlong.Study on a new modifiedparticle swarm optimization algorithm[J].Journal of Hohai Univer-sity Changzhou,2006,20(1):10-13(in Chinese).[冯骏,薛云灿,江金龙.一种新的改进粒子群算法研究[J].河海大学常州分校学报,2006,20(1):10-13.]
[7]LIN Jinping,YU Jinxiang.Parameter estimation of multiple linearregression models based on t he improved particle swarm optimizationalgorithm[J].Computer Engineering&Science,2010,32(4):101-105(in Chinese).[刘锦萍,郁金祥.基于改进的粒子群算法的多元线性回归模型参数估计[J].计算机工程与科学,2010,32(4):101-105.]
[8]GUO Huiyong,WANG Lei,LI Zhengliang.Two-stage damage de-tection method based on improved particle swarm optimization algo-rithm[J].Journal of Southwest Jiaotong University,2011,46(6):926-932(in Chinese).[郭惠勇,王磊,李正良.基于改进PSO算法的两阶段损伤识别方法[J].西南交通大学学报,2011,46(6):926-932.]
[9]LIU Shicheng,ZHANG Jianhua,LIU Zongqi.Application of paral-lel adaptive particle swarm optimization algorithm in reactive poweroptimization of power system[J].Power System Technology,2012,36(1):108-112(in Chinese).[刘世成,张建华,刘宗岐.并行自适应粒子群算法在电力系统无功系统[J].电网技术,2012,36(1):108-112.]
[10]Suresh Chandra Satapathy,Gunanidhi Pradhan,Sabyasachi Patt-naik,et al.Performance comparisons of PSO based clustering[J].InterJRI Computer Science and Networking,2009,1(1):18-23.
[2]CHEN Qing.Improvement of Welsh's grey image colorization arith-metic[J].Journal of Wuhan University of Technology,2009,31(22)151-154(in Chinese).[陈倩.一种改进的Welsh灰度图像彩色化算法[J].武汉理工大学学报,2009,31(22)151-154.]
[3]LIU Xiaopei,WAN Liang,LIN S,et al.Intrinsic colorization[J].ACM Trans on Graphics,2008,27(5):1521-1529.
[4]XU Zhen,SUN Shaoyuan,DAI Zhonghua,et al.Texture librarybased color&texture transferring for LLL images[J].InfraredTechnology,2011,33(1):49-55(in Chinese).[许真,孙韶媛,代中华,等.基于纹理特征库的微光图像色彩纹理传递[J].红外技术,2011,33(1):49-55.]
[5]ZHENG Zhi,HUANG Jian.Analysis of the application mode of col-or transfer algorithm in images[J].Journal of Fujian Agricultureand Forestry University,2010,39(2):206-210(in Chinese).[郑之,黄键.图像色彩传递算法的应用模式分析[J].福建农林大学学报,2010,39(2):206-210.]
[6]FENG Jun,XUE Yuncan,JIANG Jinlong.Study on a new modifiedparticle swarm optimization algorithm[J].Journal of Hohai Univer-sity Changzhou,2006,20(1):10-13(in Chinese).[冯骏,薛云灿,江金龙.一种新的改进粒子群算法研究[J].河海大学常州分校学报,2006,20(1):10-13.]
[7]LIN Jinping,YU Jinxiang.Parameter estimation of multiple linearregression models based on t he improved particle swarm optimizationalgorithm[J].Computer Engineering&Science,2010,32(4):101-105(in Chinese).[刘锦萍,郁金祥.基于改进的粒子群算法的多元线性回归模型参数估计[J].计算机工程与科学,2010,32(4):101-105.]
[8]GUO Huiyong,WANG Lei,LI Zhengliang.Two-stage damage de-tection method based on improved particle swarm optimization algo-rithm[J].Journal of Southwest Jiaotong University,2011,46(6):926-932(in Chinese).[郭惠勇,王磊,李正良.基于改进PSO算法的两阶段损伤识别方法[J].西南交通大学学报,2011,46(6):926-932.]
[9]LIU Shicheng,ZHANG Jianhua,LIU Zongqi.Application of paral-lel adaptive particle swarm optimization algorithm in reactive poweroptimization of power system[J].Power System Technology,2012,36(1):108-112(in Chinese).[刘世成,张建华,刘宗岐.并行自适应粒子群算法在电力系统无功系统[J].电网技术,2012,36(1):108-112.]
[10]Suresh Chandra Satapathy,Gunanidhi Pradhan,Sabyasachi Patt-naik,et al.Performance comparisons of PSO based clustering[J].InterJRI Computer Science and Networking,2009,1(1):18-23.