由于数据量大, 目前大多数端元提取算法均需较长的计算时间, 限制了这些算法的有效应用。本文提出了以光谱梯度特征为搜索条件的快速端元提取方法, 其核心包括基于光谱梯度特征的候选端元快速筛选和基于光谱解混误差的端元识别两部分。由于能够从影像中快速筛选出少量的像元光谱作为候选端元, 故具有较好的计算性能;同时由于避免了非端元光谱参与端元识别, 使得识别的结果具有更高的精度。试验表明, 相比经典的IEA算法和ECHO算法, 该算法不仅能大幅度地提高端元提取速度, 而且具有更准确的端元识别能力。同时, 基于该算法原理, 也可对现有各种算法进行改进, 提升现有的各种端元提取算法的运算速度。
Due to the large amount of image data, most algorithms for endmember extraction cost huge time, which limits the wide application of them. A fast endmember extraction algorithm is proposed by using Spectrum Gradient Features as the searching rule. The core idea is composed of two parts, namely, rapid screening of candidate endmembers based on Spectral Gradient Features and endmember identification based on spectrum unmixing residual. Being able to quickly screen out a small amount of pixels from the image as candidate endmembers, the algorithm has excellent computational performance. This algorithm can also avoid non-endmember spectrum participating in endmember identification and can obtain a result of higher accuracy. The experimental result shows that this new algorithm can greatly improve the endmember extraction speed and recognize endmembers more accurately compared with IEA and ECHO. What's more, existing algorithms for endmember extraction can be applied better based on the principle of this algorithm, and the extraction speed can be improved remarkably.
[1] MIAO L D, QI H R. Endmember Extraction from Highly Mixed Data Using Minimum Volume Constrained Nonnegative Matrix Factorization [J]. IEEE Transactions on Geoscience and Remote Sensing, 2007, 45(3): 765-777.
[2] IFARRAGUERRI A, CHANG C I. Multispectral and Hyperspectral Image Analysis with Convex Cones [J]. IEEE Transactions on Geoscience and Remote Sensing,1999, 37(2):756-770.
[3] BOWLES J H , PALMADESSO P J, ANTONIADES J A, et al. Use of Filter Vectors in Hyperspectral Data Analysis[C]//Proceedings of SPIE.New York:[s.n.], 1995, 2553: 148-157.
[4] BROWN M, LEWIS H G, GUNN S R. Linear Spectral Mixture Models and Support Vector Machines for Remote Sensing [J]. IEEE Transactions on Geoscience and Remote Sensing, 2000, 38(5):2346-2360.
[5] NEVILLE R A, STAENZ K, SZEREDI T, et al. Automatic Endmember Extraction from Hyperspectral Data for Mineral Exploration[C]//Proceedings of the Fourth International Air-borne Remote Sensing Conference and Exhibition/21st Canadian Symposium on Remote Sensing. Ottawa:[s.n.], 1999: 21-24.
[6] PENN B S. Using Simulated Annealing to Obtain Optimal Linear Endmember Mixtures of Hyperspectral Data [J]. Computers and Geosciences, 2002, 28(7):809-817.
[7] XUE Qi, KUANG Gangyao, LI Zhiyong. Endmember Extraction Algorithms from Hyperspectral Image Based on the Linear Mixing Model: An Overview [J]. Remote Sensing Technology and Application,2004,19(3):197-201. (薛绮,匡纲要,李智勇.基于线性混合模型的高光谱图像端元提取[J]. 遥感技术与应用,2004,19(3):197-201.)
[8] KUMAR S MIN H A. Some Issues Related with Subpixel Classification Using Hyperion Data[C]//ISPRS XXI VII. Beijing:[s.n.], 2008:249-254.
[9] ROBILA S A, MACIAK L G A. Parallel Unmixing Algorithm for Hyperspectral Images[C]//Intelligent Robots and Computer Vision XXIV.[S.l.]: SPIE Press,2006.
[10] LUO Wenfei, GAO Lianru. Two-level Parallel Independent Component Analysis Endmember Extraction Algorithms [J]. Journal of Remote Sensing, 2011, 15(6): 1202-1214. (罗文斐, 高连如. 二级并行独立成分分析端元提取算法[J].遥感学报, 2011,15(6):1202-1214.)
[11] PLAZA A, VALENCIA D, PLAZA J, et al. Parallel Implementation of Endmember Extraction Algorithms from Hyperspectral Data [J]. IEEE Geoscience and Remote Sensing Letters, 3(3): 334-338.
[12] PLAZA A, VALENCIA D, PLAZA J, et al. Commodity Cluster-based Parallel Processing of Hyperspectral Imagery [J]. Journal of Parallel and Distributed Computing, 2005, 66(3): 345-358.
[13] WANG Jie, YANG Liao, SHEN Jinxiang, et al. Two Endmember Extraction Algorithms with Combined Spatial and Spectral Domain TM Image[J]. Spectroscopy and Spectral Analysis, 2011, 31(5):1286-1290. (王杰,杨辽,沈金祥,等. 两种基于空间与光谱相结合的TM影像端元提取算法[J]. 光谱学与光谱分析, 2011, 31(5): 1286-1290.)
[14] GAO Xiaohui, XIANG Libin, WEI Ruyi, et al. Research on Endmember Extraction Algorithm Based on Spectral Classification[J]. Spectroscopy and Spectral Analysis, 2011,31(7):1995-1998.(高晓惠,相里斌,魏儒义,等. 基于光谱分类的端元提取算法研究[J]. 光谱学与光谱分析, 2011, 31(7):1995-1998.)
[15] ZORTEA M, PLAZA A. Spatial Preprocessing for Endmember Extraction[J]. IEEE Transactions on Geoscience and Remote Sensing, 2009,47(8): 2679-2693.
[16] ZHU Shulong, QI Jiancheng, ZHU Baoshan, et al. Fast Extraction of Endmembers from Convex Simplex's Boundary[J]. Journal of Remote Sensing, 2010,14(3): 482-492.(朱述龙, 齐建成, 朱宝山, 等.以凸面单体边界为搜索空间的端元快速提取算法[J].遥感学报,2010, 12(3):482-492.)
[17] ROGGE D M, RIVARD B, ZHANG J K, et al. Iterative Spectral Unmixing for Optimizing Per-pixel Endmember Sets[J]. IEEE Transactions on Geoscience and Remote Sensing, 2006, 44(12): 3725-3736.
[18] GENG Xiurui, ZHAO Yongchao, ZHOU Guanhua. An Automatic Endmember Extraction Algorithm Using Single form Volume from Hyperspectral Image[J].Progressing Natural Science, 2006,16(9):1196-1200.(耿修瑞,赵永超,周冠华.一种利用单形体体积自动提取高光谱图像端元的算法[J].自然科学进展,2006,16(9):1196-1200.)
[19] WU Bo, XIONG Zhuguo. Unmixing of Hyperspectral Mixture Pixels Based on Spectral Multiscale Segmented Features[J]. Acta Geodaetica et Cartographica Sinica, 2012, 41(2): 205-212.(吴波,熊助国.基于光谱最佳尺度分割特征的高光谱混合像元分解[J].测绘学报,2012,41(2):205-212.)
[20] RESMINI R G, KAPPUS M E, ALDRICH W S, et al. Mineral Mapping with Hyperspectral Digital Imagery Collection Experiment (HYDICE) Sensor-data at Cuprite, Nevada, USA[J]. International Journal of Remote Sensing,1997, 18(7): 1553-1570.
[21] WU Bo, ZHANG Liangpei, LI Pingxiang. Automatic Extraction of Endmember from Hyperspectral Imagery by Iterative Unmixing[J]. Journal of Remote Sensing,2005, 9(3):286-293.(吴波, 张良培, 李平湘. 高光谱端元自动提取的迭代分解方法[J].遥感学报,2005,9(3):286-293.)
