一种改进的MCSM/H-PSO全极化SAR影像分类方法

doi:10.13474/j.cnki.11-2246.2018.0349

测绘通报 ›› 2018, Vol. 0 ›› Issue (11): 53-57.doi: 10.13474/j.cnki.11-2246.2018.0349

一种改进的MCSM/H-PSO全极化SAR影像分类方法

余莎莎¹, 余洁^1,2,3,4, 朱腾⁵, 王彦兵^1,2,3

1. 首都师范大学资源环境与旅游学院, 北京 100048;
2. 首都师范大学城市环境过程与数字模拟国家重点实验室培育基地, 北京 100048;
3. 首都师范大学三维信息获取与应用教育部重点实验室, 北京 100048;
4. 首都师范大学资源环境与地理信息系统北京市重点实验室, 北京 100048;
5. 广东工业大学机电工程学院, 广东广州 510006

收稿日期:2018-01-28 修回日期:2018-04-23 出版日期:2018-11-25 发布日期:2018-11-29
通讯作者: 余洁。E-mail:yuj2011@whu.edu.cn E-mail:yuj2011@whu.edu.cn
作者简介:余莎莎(1994-),女,硕士,主要研究方向为极化SAR影像分类。E-mail:Vivian_world@163.com
基金资助:
国家自然科学基金（41671417）；科技创新服务能力建设-基本科研业务费（科研类）（025185305000/191）

A Modified Classification Algorithm of MCSM/H-PSO of Fully Polarimetric SAR Image

YU Shasha¹, YU Jie^1,2,3,4, ZHU Teng⁵, WANG Yanbing^1,2,3

1. College of Resource Environment and Tourism, Capital Normal University, Beijing 100048, China;
2. State Key Laboratory Incubation Base of Urban Environmental Processes and Digital Simulation, Capital Normal University, Beijing 100048, China;
3. Key Laboratory of 3-Dimensional Information Acquisition and Application, Ministry of Education, Capital Normal University, Beijing 100048, China;
4. Beijing Key Laboratory of Resources Environment and Geographic Information System, Capital Normal University, Beijing 100048, China;
5. School of Electromechanical Engineering, Guangdong University of Technology, Guangzhou 510006, China

Received:2018-01-28 Revised:2018-04-23 Online:2018-11-25 Published:2018-11-29

摘要/Abstract

摘要：

粒子群算法由于其优秀的随机全局寻优能力，在遥感图像分类领域应用广泛。为进一步提高粒子优劣判别能力，使得最终聚类中心更具合理性，本文在应用PSO算法进行全极化SAR影像分类时，考虑影像相邻像素间具有空间相关性，提出了加权PSO算法，以提高分类精度。同时，在进行全极化SAR影像分类时，为了更充分地利用全极化SAR影像极化特征，采用多成分散射模型分解（MCSM）方法结合散射熵提取影像6种极化特征。改进的MCSM/H-PSO全极化SAR影像分类方法，首先通过MCSM分解和散射熵对全极化SAR影像进行基于散射机理的初分类，再将分类结果作为加权PSO算法的初始类别划分，并通过迭代实现地物分类。采用北京市Radarsat 2全极化SAR数据和美国旧金山AIRSAR全极化SAR数据分别进行试验，本文方法分类总体精度分别可达90.57%和93.25%。

关键词: 全极化SAR影像分类, MCSM分解, PSO算法, 散射熵

Abstract:

Particle swarm optimization (PSO) is widely used in image classification because of the ability of random global optimization.The quality of particle has a major influence on the identification of cluster center.Because of the spatial correlation of adjacent pixels in fully polarimetric SAR image,we proposed to use weighted PSO algorithm in SAR image classification to make the cluster center more reasonable for the improvement of classification accuracy.Meanwhile,we used the multiple-component scattering model (MCSM) method combined with the scatter entropy to extract the 6 polarization features of the images,making full use of the polarimetric characteristics of the polarimetric SAR images.In the proposed method,firstly,the SAR image is classified as preliminary classificat based on scattering mechanism by MCSM decomposition and scatter entropy. Secondly,the result of classification is used as the initial classification of the weighted PSO algorithm to achieve the classification of objects by iteration.The result of using the full polarimetric SAR data of Radarsat2 in Beijing and the full polarimetric SAR data of AIRSAR in San Francisco AIRSAR show that the total accuracy of the proposed method was 90.57% and 93.25%,respectively.

Key words: SAR image classification, MCSM decomposition, the PSO algorithm, scattering entropy

中图分类号:

P237

余莎莎, 余洁, 朱腾, 王彦兵. 一种改进的MCSM/H-PSO全极化SAR影像分类方法[J]. 测绘通报, 2018, 0(11): 53-57.

YU Shasha, YU Jie, ZHU Teng, WANG Yanbing. A Modified Classification Algorithm of MCSM/H-PSO of Fully Polarimetric SAR Image[J]. 测绘通报, 2018, 0(11): 53-57.

参考文献

[1] 杨杰,赵伶俐,史磊,等.基于最优极化相干系数的倾斜建筑物解译研究[J].测绘学报,2012,41(4):577-583.
[2] 吴永辉,计科峰,郁文贤.基于H-α和改进C-均值的全极化SAR图像非监督分类[J].电子与信息学报,2007,29(1):30-34.
[3] 赵一博.基于SVM和RBF神经网络的极化SAR图像分类方法研究[D].长沙:国防科学技术大学,2013.
[4] 刘利敏,余洁,燕琴,等.引入纹理特征的SSVM算法在极化SAR影像分类中的应用[J].测绘通报,2012(8):7-10.
[5] 李林宜,李德仁.基于粒子群优化算法的航空影像纹理分类[J].计算机工程,2008,34(6):196-198.
[6] KENNEDY J,EBERHART R.Particle Swarm Optimization[C]//Proceedings of IEEE International Conference on Neural Networks,1995.[S.l.]:IEEE,2002:1942-1948.
[7] LEE J S,GRUNES M R,AINSWORTH T L,et al.Unsupervised Classification Using Polarimetric Deco-mposition and the Complex Wishart Classifier[J].IEEE Transactions on Geoscience and Remote Sensing,1999,37(5):2249-2258.
[8] 王超.全极化合成孔径雷达图像处理[M].北京:科学出版社,2008.
[9] CLOUDE S R,POTTIER E.An Entropy Based Classification Scheme for Land Applications of Polarimetric SAR[J].IEEE Transactions on Geoscience & Remote Sensing,1997,35(1):68-78.
[10] FREEMAN A,DURDEN S L.Three-component Scattering Model to Describe Polarimetric SAR Data[J].Proc. Spie,1993,52(3):1843-1855.
[11] ZHANG L,LU D,TANG W.A GS-based Built-up Area Detection Method Using Polarimetric SAR Images[C]//2012 IEEE International Conference on Geoscience and Remote Sensing Symposium (IGARSS).[S.l.]:IEEE,2012:5911-5914.
[12] LEE J S,GRUNES M R,DE GRANDI G.Polarimetric SAR Speckle Filtering and Its Implication for Classification[J].IEEE Transactions on Geoscience and Remote Sensing,2002,37(5):2363-2372.
[13] ZHANG L,ZOU B,CAI H,et al.Multiple-component Scattering Model for Polarimetric SAR Image Decomposition[J].IEEE Geoscience & Remote Sensing Letters,2008,5(4):603-607.
[14] SARIF B,ABD-EL-BARR M.Functional Synthesis Using Discrete Particle Swarm Optimization[C]//2008 IEEE Swarm Intelligence Symposium. St.Louis, MO:IEEE,2008.

一种改进的MCSM/H-PSO全极化SAR影像分类方法

A Modified Classification Algorithm of MCSM/H-PSO of Fully Polarimetric SAR Image

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价