一种基于改进的遗传算法的脑MR图像去偏移场模型

计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (15): 29-32.

一种基于改进的遗传算法的脑MR图像去偏移场模型

陈允杰²,张建伟¹,王利²,王平安³,夏德深²

1.南京信息工程大学数理学院，南京 210044
2.南京理工大学计算机学院，南京 210094
3.香港中文大学计算机科学与工程学系，香港

收稿日期:2008-01-08 修回日期:2008-03-03 出版日期:2008-05-21 发布日期:2008-05-21
通讯作者: 陈允杰

Brain MR images de-bias model based on adapted genetics algorithm model

CHEN Yun-jie¹,ZHANG Jian-wei¹,WANG Li²,WANG Ping-an³,XIA De-shen²

1.School of Math and Physics，Nanjin University of Information Science and Technology，Nanjin 210044，China
2.School of Computer Science & Technology，Nanjin University of Science and Technology，Nanjin 210094，China
3.Department of Computer Science & Engineering，Hong Kong Chinese University，Satian Hong Kong，China

Received:2008-01-08 Revised:2008-03-03 Online:2008-05-21 Published:2008-05-21
Contact: CHEN Yun-jie

摘要/Abstract

摘要： 由于磁共振图像（Magnetic Resonance Images，MRI）常含有偏移场，影响后继图像分割。采用Legendre多项式基函数来拟合偏移场，以去除偏移场对图像分割的影响。当使得恢复图像的信息熵达到最小时，求得的偏移场最优。求偏移场的过程中需要求解基函数的参数，由于传统的梯度下降法易陷入局部最优，将遗传算法引入到参数求解过程中，然而传统的遗传算法时间复杂度高，易陷入局部最优，对遗传算法进行了改进，更容易得到全局最优解且时间复杂度较低。实验证明该算法可以得到精确的偏移场，得到准确的分割结果。

关键词: MRI, 偏移场, 信息熵, 梯度下降法, 遗传算法, 局部最优, 全局最优, ,

Abstract: Intrascan intensity in homogeneities is a common source of difficulty for MRI segmentation.The authors estimate the bias field by Legendre polynomials to find the parameters with minimum entropy，conventional ways such as gradient-descent method often find local best，to find global best，the authors present genetics algorithm to find best parameters to estimate the bias field，but it can not always find global best neither.Then the authors make some modification of genetic algorithm to make it easier to find global best.Experiments on the segmentation of brain magnetic resonance images show the modification in this paper can find more accurate bias field and have better effect in image segmentation.

Key words: Magnetic Resonance Imaging（MRI）, bias field, entropy, gradient-descent, genetic algorithm, local best, global best

陈允杰²,张建伟¹,王利²,王平安³,夏德深². 一种基于改进的遗传算法的脑MR图像去偏移场模型[J]. 计算机工程与应用, 2008, 44(15): 29-32.

CHEN Yun-jie¹,ZHANG Jian-wei¹,WANG Li²,WANG Ping-an³,XIA De-shen². Brain MR images de-bias model based on adapted genetics algorithm model[J]. Computer Engineering and Applications, 2008, 44(15): 29-32.