Abstract：Hyperspectral images classification is one of the important methods to identify image information, which has great significance for feature identification, dynamic monitoring and thematic information extraction, etc. Unsupervised classification without prior knowledge is widely used in hyperspectral image classification. This article proposes a new hyperspectral images unsupervised classification algorithm based on harmonic analysis(HA), which is called the harmonic analysis classifer (HAC). First, the HAC algorithm counts the first harmonic component and draws the histogram, so it can determine the initial feature categories and the pixel of cluster centers according to the number and location of the peak. Then, the algorithm is to map the waveform information of pixels to be classified spectrum into the feature space made up of harmonic decomposition times, amplitude and phase, and the similar features can be gotten together in the feature space, these pixels will be classified according to the principle of minimum distance. Finally, the algorithm computes the Euclidean distance of these pixels between cluster center, and merges the initial classification by setting the distance threshold. so the HAC can achieve the purpose of hyperspectral images classification. The paper collects spectral curves of two feature categories, and obtains harmonic decomposition times, amplitude and phase after harmonic analysis, the distribution of HA components in the feature space verified the correctness of the HAC. While the HAC algorithm is applied to EO-1 satellite Hyperion hyperspectral image and obtains the results of classification. Comparing with the hyperspectral image classifying results of K-MEANS, ISODATA and HAC classifiers, the HAC, as a unsupervised classification method, is confirmed to have better application on hyperspectral image classification.
杨可明，魏华锋，史钢强，孙阳阳，刘 飞 . 光谱谐波分析的新型HAC非监督分类器 [J]. 光谱学与光谱分析, 2015, 35(07): 2001-2006.
YANG Ke-ming, WEI Hua-feng, SHI Gang-qiang, SUN Yang-yang, LIU Fei . A New HAC Unsupervised Classifier Based on Spectral Harmonic Analysis . SPECTROSCOPY AND SPECTRAL ANALYSIS, 2015, 35(07): 2001-2006.
[1] TONG Qing-xi, ZHANG Bing, ZHENG Fen-lan(童庆禧，张 兵，郑芬兰). Hyperspectral Remote Sensing: Principles, Technology and Applications(高光谱遥感：原理、技术与应用). Beijing: Higher Education Press(北京：高等教育出版社), 2006. 169. [2] CHEN Ping-sheng(陈平生). Journal of Jiangxi University of Science and Technology(江西理工大学学报), 2012, 33(1): 79. [3] YANG Guo-peng, YU Xu-chu, CHEN Wei, et al(杨国鹏，余旭初，陈 伟，等). Journal of Remote Sensing(遥感学报), 2008, 12(4): 580. [4] JIA Kun, LI Qiang-zi, TIAN Yi-chen, et al(贾 坤，李强子，田亦陈，等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2011, 31(10): 2619. [5] George J Wakileh. Power Systems Harmonics Fundamentals, Analysis and Filter Design(电力系统谐波—基本原理、分析方法和滤波器设计). Translated by XU Zheng(徐 政，译). Beijing: China Machine Press(北京：机械工业出版社), 2003. 4. [6] McGranaghan M. IEEE Power Engineering Review, 2001, 21(10): 3. [7] WEN He, TENG Zhao-sheng, WANG Yong, et al(温 和，滕召胜，王 永，等). Transactions of China Electrotechnical Society(电工技术学报), 2012, 27(12): 270. [8] NIU Sheng-suo, LIANG Zhi-rui, ZHANG Jian-hua, et al(牛胜锁，梁志瑞，张建华，等). Proceedings of the CSEE(中国电机工程学报), 2012, 32(16): 130. [9] Jakubauskas M E, Legates D R, Kastens J H. Photogrammetric Engineering & Remote Sensing, 2001, 67(4): 461. [10] Jakubauskas M E, Legates D R, Kastens J H. Computers and Electronics in Agriculture, 2002, 37(1-3): 127. [11] LIANG Shou-zhen, XING Qian-guo, SHI Ping, et al(梁守真，邢前国，施 平，等). Chinese Journal of Ecology(生态学杂志), 2011, 30(1): 59. [12] Toshihiros, Masayuki Y, Hitoshi T, et al. Remote Sensing of Environment. 2005, 96(3/4): 366. [13] YANG Ke-ming, ZHANG Tao, WANG Li-bo, et al(杨可明，张 涛，王立博，等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2013, 33(9): 2497. [14] YANG Ke-ming, XUE Chao-hui, JIA Tao-tao, et al(杨可明，薛朝辉，贾涛涛，等). Acta Geodaetica et Cartographica Sinica(测绘学报), 2013, 42(1): 36. [15] GAN Zheng-ru, ZENG Xian-gui(干正如，曾宪珪). Journal of Jiangxi University of Science and Technology(江西理工大学学报), 2007, 28(4): 39.
