基于形态学空间特征的高光谱遥感图像分类方法

吕俊伟; 樊利恒; 石晓航

光学技术, 2016, 42 (5): 385, 网络出版: 2016-10-19

基于形态学空间特征的高光谱遥感图像分类方法

Classification of hyperspectral remotely sensed images based on the extraction of spatial feature

论文大纲

吕俊伟 ^1,*樊利恒 ²石晓航 ¹

作者单位

¹ 海军航空工程学院控制工程系, 　山东　烟台　264001

² 海军航空装备计量监修中心, 　上海　200436

高光谱遥感图像空间特征形态学 k-NN分类算法 hyperspectral remotely sensed image spatial feature morphology k-NN classification

摘要

传统的高光谱图像分类主要是基于像素的光谱特征, 在一定程度上忽略了高光谱遥感图像中像素之间的空间相关性。为了充分利用高光谱图像中的空间信息, 提出了一种基于加权多结构元素无偏差形态学的空间特征提取方法, 并基于形态学的多尺度特征和结构保持性提出了基于邻域的多尺度空间特征提取方法, 得到了高光谱遥感图像的空间特征。对k-NN分类算法进行改进, 提出了基于变精度粗糙集和重构误差的k-NN分类算法, 实现了基于空间特征的高光谱遥感图像分类。在两个不同的高光谱遥感图像的实验验证了基于空间特征和改进k-NN分类算法的性能。

Abstract

Traditional classification methods for hyperspectral remotely sensed images are mostly based on the character of spectral feature. The spatial features of hyperspectral images are somehow neglected. To make a better use of the spatial information of hyperspectral images, a spatial feature extraction method is proposed based on weighted multi-structure elements mathematical morphological processing without bias. Another spatial feature extraction method based on neighborhood based on the multiscale character of morphological structure is proposed. Then the spatial feature of hyperspectral is formed. An improved k-NN classification method is proposed based on variable precision rough sets and reconstruction error and the method is used to classify hyperspectral image. Experiment results show that the proposed method can classify hyperspectral images successfully.

参考文献

[1] YEU Y. Feature extraction from hyperspectral imagery for object recognition［D］. Columbus: The Ohio State University, 2011.

[2] QUESADA-BARRIUSO P, ARGUELLO F, HERAS D B. Spectral-spatial classification of hyperspectral images using wavelets and extended morphological profiles［J］. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014, 7(4): 1177-1185.

[3] 孔祥兵,等.结合空间和光谱信息的高光谱影像端元光谱自动提取［J］.光谱学与光谱分析, 2013, 33(6): 1647-1652.

KONG Xiangbing, et al.. Integration of spatial-spectral information based endmember extraction for hyperspectral image［J］. Spectroscopy and Spectral Analysis, 2013, 33(6): 1647-1652.

[4] RAJADELL O, GARCIA-SEVILLA P, PLA F. Spectral–spatial pixel characterization using gabor filters for hyperspectral image classification［J］. IEEE Geoscience and Remote Sensing Letters, 2013,10(4): 860-864.

[5] FAUVEL M, et al. A spatial–spectral kernel-based approach for the classification of remote-sensing images［J］. Pattern Recognition, 2012,45: 381-392.

[6] 刘建军,吴泽彬, 韦志辉. 基于空间相关性约束稀疏表示的高光谱图像分类［J］.电子与信息学报, 2012, 34(11): 2666-2671.

LIU Jianjun, WU Zebin, WEI Zhihui, et al.. Spatial correlation constrained sparse representation［J］. Journal of Electronics & Information Technology, 2012, 34(11): 2666-2671.

[7] 岳江,等. 一种基于空间一致性降元的高光谱图像非监督分类［J］. 光谱学与光谱分析, 2012, 32(7): 1860-1864.

YUE Jiang, et al.. An unsupervised classification of hyperspectral images based on pixels reduction with spatial coherence property［J］. Spectroscopy and Spectral Analysis, 2012, 32(7): 1860-1864.

[8] 赵辽英,等.基于数学形态学的高光谱图像组合核目标检测［J］.光学学报, 2011, 31(12): 269-274.

ZHAO Liaoying, et al.. Composite kernel target detection based on mathematical morphology for hyperspectral imagery［J］. Acta Optica Sincia, 2011, 31(12): 269-274.

[9] KETTIG R L, et al. Classification of multispectral image data byextraction and classification of homogeneous objects［J］. IEEE Trans. Geosci. Electro., 1976, GE-14(1): 19-26.

[10] BREEN E J, JONES R. Attribute openings, thinnings, and granulometries［J］. Comput. Vis. Image Underst., 1996, 64(3): 377-389.

[11] BENEDIKTSSON J A, et al. Classification of hyperspectral data from urban areas based on extended morphological profiles［J］. IEEE Trans. Geosci. Remote Sens., 2005, 43(3): 480-491.

[12] MURA M D, et al. Extended profiles with morphological attribute filters for the analysis of hyperspectral data［J］. Int. J. Remote Sens, 2010, 31(22): 5975-5991.

[13] 高恒振.高光谱遥感图像分类技术研究［D］.长沙: 国防科学技术大学,2011: 44-56.

GAO Hengzhen. Research on classification technique for hyperspectral remote sensing imagery［D］. Changsha: National University of Defense Technology, 2011: 44-56.

[14] SALEMBIER P, SERRA J. Flat zones filtering, connected operators, and filters by reconstruction［J］. IEEE Trans. Image Process, 1995,(4): 1153-1160.

[15] BOSWORTH J H, ACTON S T. Morphological scale-space in image processing［J］. Digital Signal Processing, 2003, 13: 338-367.

[16] 代六玲,等.中文文本分类中特征抽取方法的比较研究［J］.中文信息学报,2004,18(1): 26-32.

DAI Liuling, et al. A comparative study on feature selection in chinese text categorization［J］. Journal of Chinese Information Processing, 2004,18(1): 26-32.

[17] 余鹰,等. 基于变精度粗糙集的KNN 分类改进算法［J］. 模式识别与人工智能,2012,25(4): 617-613.

YU Ying, et al.. An improved KNN algorithm based on variable precision rough sets［J］. PR & AI, 2012,25(4): 617-613.

[18] FU Y, YAN S, HUANG T S. Classification and feature extraction by simplexization［J］. IEEE Transactions on Information Forensics and Security ,2008 ,3(1): 91-100.

[19] ZIARKO W. Variable precision rough sets model［J］. Journal of Computer and System Science, 1993, 46(1): 39-59.

吕俊伟, 樊利恒, 石晓航. 基于形态学空间特征的高光谱遥感图像分类方法[J]. 光学技术, 2016, 42(5): 385. LV Junwei, FAN Liheng, SHI Xiaohang. Classification of hyperspectral remotely sensed images based on the extraction of spatial feature[J]. Optical Technique, 2016, 42(5): 385.

基于形态学空间特征的高光谱遥感图像分类方法

关于本站 Cookie 的使用提示

全站搜索

基于形态学空间特征的高光谱遥感图像分类方法

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索