Hyperspectral remote sensing image parallel processing based on cluster and GPU

Wang Maozhi; Guo Ke; Xu Wenxi

The parallel algorithms design and implementation of covariance matrix, related to PCA and MNF, and SCM used in hyperspectral remote sensing image data process was discussed in this paper. The covariance matrix parallel algorithm was designed and implemented under cluster circumstance based on MPI. On the other hand, parallel algorithm of SCM was designed and implemented based on GPU. Both of these two parallel algorithms were verified during the application on hyperspectral remote sensing data processing. The experiment results prove that high performance computing is effective during the data process of hyperspectral remote sensing image, and it should be an important technique for generalization of hyperspectral remote sensing to engineering quick application. The results also prove the correctness of the parallel algorithms proposed in this paper.

1. Geomathematics Key Lab of Sichuan Province,Chengdu University of Technology,Chengdu 610059,China

Received Date: 2013-03-19

Rev Recd Date: 2013-04-24

Publish Date: 2013-11-25

Key words:

Abstract: The parallel algorithms design and implementation of covariance matrix, related to PCA and MNF, and SCM used in hyperspectral remote sensing image data process was discussed in this paper. The covariance matrix parallel algorithm was designed and implemented under cluster circumstance based on MPI. On the other hand, parallel algorithm of SCM was designed and implemented based on GPU. Both of these two parallel algorithms were verified during the application on hyperspectral remote sensing data processing. The experiment results prove that high performance computing is effective during the data process of hyperspectral remote sensing image, and it should be an important technique for generalization of hyperspectral remote sensing to engineering quick application. The results also prove the correctness of the parallel algorithms proposed in this paper.

HTML

Reference (25)

[1]
[2]	Antonio Plaza, Jon Atli Benediktsson, Boardman Joseph W, et al. recent advances in techniques for hyperspectral image processing[J]. Remote Sensing of Environment, 2009, 113: S110-S122.
[3]
[4]	Antonio Plaza, David Valencia, Javier Plaza. An experimental comparison of parallel algorithms for hyperspectral analysis using heterogeneous and homogeneous networks of workstations[J]. Parallel Computing, 2008, 34: 92-114.
[5]
[6]	Wang Wei, Zhao Huijie, Dong Chao. Parallel algorithm of anomalies detection in hyperspectral image with projection pursuit[J]. Journal of Beijing University of Aeronautics and Astronautics, 2009, 35(3): 342-346. (in Chinese)王维, 赵慧洁, 董超. 基于投影寻踪的高光谱图像异常检测并行算法[J]. 北京航空航天大学学报, 2009, 35(3): 342-346.
[7]
[8]	Shen Zhanfeng, Luo Jiancheng, Chen Qiuxiao, et al. High-efficiency remotely sensed image parallel processing method study based on MPI[J]. Journal of Image and Graphics, 2007, 12(12): 2132-2136. (in Chinese)沈占锋, 骆剑承, 陈秋晓, 等. 基于MPI的遥感影像高效能并行处理方法研究[J]. 中国图象图形学报, 2007, 12(12): 2132-2136.
[9]
[10]	Lv Jie, Zhang Tianxu, Zhang Biyin. Applications of MPI parallel computing on image processing[J]. Infrared and Laser Engineering, 2004, 33(5): 496-499. (in Chinese)吕捷, 张天序, 张必银. MPI并行计算在图像处理方面的应用[J]. 红外与激光工程. 2004, 33(5): 496-499.
[11]
[12]	Yang Renzhong, Chen Minhao, Shi Lu. Interferential hyper-spectral real time spectrum reconstruction technology based on CUDA[J]. Remote Sensing Technology and Application, 2011, 26(4): 420-425. (in Chinese)杨仁忠, 陈敏浩, 石璐. 基于CUDA 的干涉型高光谱实时光谱复原处理技术[J]. 遥感技术与应用, 2011, 26(4): 420-425.
[13]	He Guojing, Liu Delian, Zhang Jianqi. High speed spectral matching approach for hyperspectral image based on CUDA[J]. Aero Weaponry, 2011, 4: 3-6. (in Chinese)何国经, 刘德连, 张建奇. CUDA 架构下高光谱图像光谱匹配的快速实现[J]. 航空兵器, 2011, 4: 3-6.
[14]
[15]	Shi Kun, Hao Yingming, Wang Mingming, et al. Real-time simulation method of infrared sea background[J]. Infrared and Laser Engineering, 2012, 41(1): 25-29. (in Chinese)石坤, 郝颖明, 王明明, 等. 海面背景红外实时仿真[J]. 红外与激光工程, 2012, 41(1): 25-29.
[16]
[17]
[18]	Jason Sanders, Edward Kandrot. CUDA by Example: An Introduction to General-Purpose GPU Programming[M]. Boston: Addison-Wesley Professional, 2011: 126-150.
[19]	Wei Feng, He Mingyi, Mei Shaohui. Hyperspectral data feature extraction using spatial coherence based neighborhood preserving embedding[J]. Infrared and Laser Engineering, 2012, 41(5): 1249-1254. (in Chinese)魏峰, 何明一, 梅少辉. 空间一致性邻域保留嵌入的高光谱数据特征提取[J]. 红外与激光工程, 2012, 41(5): 1249-1254.
[20]
[21]	Tiranee Achalakula, Stephen Taylor. A distributed spectral-screening PCT algorithm[J]. J Parallel Distrib Comput, 2003, 63: 373-384.
[22]
[23]	Kleanthis Psarris. Program analysis techniques for transforming programs for parallel execution[J]. Parallel Computing, 2002, 28: 455-469.
[24]
[25]	Shane Ryoo, Christopher I Rodrigues, Sam S Stone, et al. Program optimization carving for GPU computing[J]. J Parallel Distrib Comput, 2008, 68: 1389-1401.

Hyperspectral remote sensing image parallel processing based on cluster and GPU

Abstract

References

Proportional views

通讯作者: 陈斌, bchen63@163.com

Article Metrics

Related

Proportional views