一种有效的MSTAR SAR图像分割方法

林达; 徐新; 董浩; 谢文涛

doi:10.13203/j.whugis20130572

一种有效的MSTAR SAR图像分割方法

武汉大学电子信息学院, 湖北武汉, 430079

基金项目: 国家重点基础研究发展计划(973计划)资助项目(2013CB733404)。

详细信息

作者简介:
林达,博士生,研究方向为图像处理、计算机视觉。E-mail:linda_giggle@whu.edu.cn

中图分类号: P237.3
计量
- 文章访问数: 1775
- HTML全文浏览量: 103
- PDF下载量: 508
出版历程
- 收稿日期: 2013-12-23
- 发布日期: 2015-10-04

An Effective Segmentation Algorithm for MSTAR SAR Target Chips

Electronic Information School, Wuhan University, Wuhan 430079, China

Funds: The National Basic Research Program(973 Program) of China, No. 2013CB733404.

摘要

摘要: 提出了一种有效的MSTAR SAR图像分割方法。该方法首先对待处理图像进行过分割操作,得到过分割图像区域,然后对过分割后的图像进行图像区域级和像素级的特征提取,得到用于表示图像的特征向量,接着对MSTAR SAR图像使用空间隐含狄利克雷分配模型(sLDA)和马尔科夫随机场(MRF)建立本文所提出的模型,得到能量泛函,最后运用Graph-Cut算法和Branch-and-Bound算法对能量泛函进行优化,得到最终的分割结果。通过使用MSTAR SAR图像进行分割实验比较,仿真结果表明了方法的有效性。
- MSTAR SAR图像分割 /
- 空间隐含狄利克雷分配模型 /
- 马尔科夫随机场 /
- 能量优化
Abstract: We present an effective segmentation algorithm for MSTAR SAR target chips. First, the image over-segmentation is implemented to acquire image regions. Then the region-level and pixel-level features are generated to represent SAR images of MSTAR SAR chips. Finally, the Graph-Cut and Branch-and-Bound algorithms are applied to the energy function obtained by sLDA and MRF to achieve the final segmentation results. Through a comparison of distinct SAR image segmentation experiments, our simulation results demonstrate the superior performance of our proposed method in terms of effectiveness.
- MSTAR SAR image segmentation /
- spatial latent Dirichlet allocation (sLDA) /
- Markov random field (MRF) /
- energy optimization

HTML全文

参考文献(12)

[1]	Liao Mingsheng, Lin Hui.Synthetic Aperture Radar Interferometry—Principle and Signal Processing[M].Beijing:Surveying and Mapping Press,2003(廖明生,林珲.雷达干涉测量学:原理与信号处理基础[M].北京:测绘出版社,2003)
[2]	Blei D M, Ng A Y, Jordan M I. Latent Dirichlet Allocation[J]. Journal of Machine Learning Research, 2003,3(1): 993-1 022
[3]	Cao L L, Li F F. Spatially Coherent Latent Topic Model for Concurrent Segmentation and Classification of Objects and Scenes[C]. IEEE 11th International Conference on Computer Vision, Riode Janeiro,Brazil,2007
[4]	Verbeek J, Triggs B. Region Classification with Markov Field Aspect Models[C]. IEEE Conference on Computer Vision and Pattern Recognition, Minneapolis, Minnesota, USA,2007
[5]	Liu Mengling, He Chu, Su Xin, et al.A pLSA Based on Topo-MRF Model Method for SAR Images Classification[J].Geomatics and Information Science of Wuhan University,2011,36(1):122-125(刘梦玲, 何楚, 苏鑫, 等. 基于 pLSA 和 Topo-MRF 模型的 SAR 图像分类算法研究[J]. 武汉大学学报·信息科学版, 2011, 36(1): 122-125)
[6]	Lempitsky V, Blake A, Rother C. Branch-and-mincut: Global Optimization for Image Segmentation with High-level Priors[J]. Journal of Mathematical Imaging and Vision, 2012, 44(3): 315-329
[7]	Lawler E L, Wood D E. Branch-and-Bound Methods: A Survey[J]. Operations Research, 1966, 14(4): 699-719
[8]	Hofmann T. Probabilistic Latent Semantic Analysis[C]//Proceedings of the 15th Conference on Uncertainty in Artificial Intelligence. Stockholm,Sweden:Morgan Kaufmann Publishers Inc, 1999: 289-296
[9]	Griffiths T L, Steyvers M. Finding Scientific Topics[J]. The National Academy of Sciences of the United States of America, 2004, 101(Suppl 1): 5 228-5 235
[10]	Minka T, Lafferty J. Expectation-Propagation for the Generative Aspect Model[C]//Proceedings of the 18th Conference on Uncertainty in Artificial Intelligence. Alberta,Canada:Morgan Kaufmann Publishers Inc, 2002: 352-359
[11]	Bleau A, Leon L J. Watershed-based Segmentation and Region Merging[J]. Computer Vision and Image Understanding, 2000, 77(3): 317-370
[12]	Wu F Y. The Potts Model[J]. Reviews of Modern Physics, 1982, 54(1): 235-268