基于端元字典稀疏解混的高光谱图像亚像元定位

doi:10.16180/j.cnki.issn1007-7820.2019.04.008

电子科技 ›› 2019, Vol. 32 ›› Issue (4): 33-39.doi: 10.16180/j.cnki.issn1007-7820.2019.04.008

基于端元字典稀疏解混的高光谱图像亚像元定位

顾正之^1,^2,³,王素玉^1,^2,³

^1. 北京未来网络科技创新中心,北京 100124
^2. 北京物联网软件与系统工程研究中心,北京 100124
^3. 北京工业大学信息学部,北京 100124

收稿日期:2018-03-18 出版日期:2019-04-15 发布日期:2019-03-27
作者简介:顾正之(1991-),男,硕士研究生。研究方向:数字图像处理。
基金资助:
国家自然科学基金(61201361);北京市教委科学基金(KM201710005011);北京市人才培养计划基金(2013D005015000008)

Hyperspectral Image Sub-Pixel Mapping Based on Sparse Unmixing of Endmember Dictionary

GU Zhengzhi^1,^2,³,WANG Suyu^1,^2,³

^1. Beijing Advanced Innovation Center for Future Internet Technology, Beijing 100124,China
^2. Beijing Engineering Research Center for IoT Software and Systems, Beijing 100124,China
^3. Faculty of Information Technology, Beijing University of Technology, Beijing 100124,China

Received:2018-03-18 Online:2019-04-15 Published:2019-03-27
Supported by:
National Natural Science Foundation of China(61201361);Science Foundation of the Beijing Education Commission(KM201710005011);Training Program Foundation for the Talents in Beijing City(2013D005015000008)

摘要/Abstract

摘要：

针对高光谱图像中普遍存在的混合像元中各端元空间分布定位困难的问题,文中提出一种基于K-SVD的光谱解混算法,利用其解混结果进行亚像元定位。算法首先通过KNN分类来区分待处理图像中的混合像元和纯像元,然后借鉴基于冗余字典的稀疏分解相关理论,以标准光谱库为基础,通过基于K-SVD的字典训练算法训练产生最具代表性的地物光谱曲线,构建端元冗余字典,通过基于K-SVD的稀疏分解算法实现各端元丰度的求解。最后利用求得的丰度系数在两种空间性相关性约束下进行亚像元定位。实验结果表明,采用该算法进行模拟数据和真实数据的亚像元的定位可以取得不错的定位结果。

关键词: 高光谱图像, 光谱解混, 亚像元定位, K-SVD, 稀疏表示, 冗余字典

Abstract:

In general, it is difficult to locate the spatial distribution of each endmember in the hyperspectral image. To solve the problem, this study proposed a K-SVD-based spectral unmixing algorithm whose result of the demixing was further used for performing the subpixel location. First, the mixed pixel and pure pixel were distinguished by KNN classification, and then the sparse decomposition correlation theory based on redundant dictionary was used for reference. Based on the standard spectral library, the K-SVD based dictionary training algorithm was used to train the most representative material spectral curves, and the endmember redundancy dictionary was subsequently constructed. The abundances were solved by the sparse decomposition algorithm based on K-SVD. Finally, the obtained abundance coefficient was used to locate the sub-pixel under the two spatial correlation constraints. Experimental results showed that the proposed algorithm had reliable performance for the sub-pixel mapping effects of simulated data and measured data.

Key words: hyperspectral image, spectral unmixing, sub-pixel mapping, K-SVD, sparse representation, redundant dictionary

中图分类号:

TP391

顾正之,王素玉. 基于端元字典稀疏解混的高光谱图像亚像元定位[J]. 电子科技, 2019, 32(4): 33-39.

GU Zhengzhi,WANG Suyu. Hyperspectral Image Sub-Pixel Mapping Based on Sparse Unmixing of Endmember Dictionary[J]. Electronic Science and Technology, 2019, 32(4): 33-39.

图/表 12

图1

图2

图3

图4

表1

表2

图5

图6

图7

图8

表3

图9

参考文献 18

[1]	Duponchel L . Exploring hyperspectral imaging data sets with topological data analysis[J]. Analytica Chimica Acta, 2018,1000(7):123-131. doi: 10.1016/j.aca.2017.11.029 pmid: 29289301
[2]	宋义刚, 吴泽彬, 韦志辉 , 等. 稀疏性高光谱解混方法研究[J]. 南京理工大学学报, 2013,37(4):486-492. doi: 10.3969/j.issn.1005-9830.2013.04.005
	Song Yigang, Wu Zebin, Wei Zhihui , et al. Research on sparse hyperspectral unmixing methods[J]. Journal of Nanjing University of Technology, 2013,37(4):486-492. doi: 10.3969/j.issn.1005-9830.2013.04.005
[3]	Somers B, Asner G P, Tits L , et al. Endmember variability in spectral mixture analysis: a review[J]. Remote Sensing of Environment, 2011,115(7):1603-1616. doi: 10.1016/j.rse.2011.03.003
[4]	Atkinson P M, Culter M E J, Lewis H . Mapping sub-pixel proportional land cover with AVHRR imagery[J]. International Journal of Remote Sensing, 1997,18(4):917-935. doi: 10.1080/014311697218836
[5]	Chen Y, Ge Y, Chen Y , et al. Subpixel land cover mapping using multiscale spatial dependence[J]. IEEE Transactions on Geoscience & Remote Sensing, 2018(99):1-10. doi: 10.1109/TGRS.2018.2808410
[6]	Wu K, Du Q, Wang Y , et al. Supervised sub-pixel mapping for change detection from remotely sensed images with different resolutions[J]. Remote Sensing, 2017,9(3):284. doi: 10.3390/rs9030284
[7]	吴健康 . 基于稀疏表示的高光谱图像解混研究[D]. 西安: 西安电子科技大学, 2014.
	Wu Jiankang . Hyperspectral image unmixing based on sparse representation[D]. Xi’an:Xidian University, 2014.
[8]	张亚坤, 张洪艳, 沈焕锋 , 等. 一种基于稀疏表达的遥感影像时空融合方法[J]. 电子科技, 2017,30(11):56-59. doi: 10.16180/j.cnki.issn1007-7820.2017.11.016
	Zhang Yakun, Zhang Hongyan, Shen Huanfeng , et al. A spatiotemporal fusion algorithm based on sparse representation for remote sensing imagery[J]. Electronic Science and Technology, 2017,30(11):56-59. doi: 10.16180/j.cnki.issn1007-7820.2017.11.016
[9]	Iordache M D, Plaza A, Dias J B . On the use of spectral libraries to perform sparse unmixing of hyperspectral data[C]. Reykjavik:IEEE GRSS Workshop on Hyperspectral Image and Signal Processing:Evolution in Remote Sensing (WHISPERS), 2010.
[10]	Iordache M D, Bioucas-Dias J M, Plaza A . Collaborative sparse regression for hyperspectral unmixing[J]. IEEE Transactions on Geoscience & Remote Sensing, 2014,52(1):341-354. doi: 10.1109/TGRS.2013.2240001
[11]	沈丽, 韩彦芳 . 基于稀疏表示的超分辨率图像重建[J]. 电子科技, 2015,28(9):144-147. doi: 10.16180/j.cnki.issn1007-7820.2015.09.037
	Shen Li, Han Yanfang . Image super-resolution reconstruction based on sparse representation[J]. Electronic Science and Technology, 2015,28(9):144-147. doi: 10.16180/j.cnki.issn1007-7820.2015.09.037
[12]	刘万军, 杨秀红, 曲海成 , 等. 基于光谱信息散度与光谱角匹配的高光谱解混算法[J]. 计算机应用, 2015,35(3):844-848. doi: 10.11772/j.issn.1001-9081.2015.03.844
	Liu Wanjun, Yang Xiuhong, Qu Haicheng, et al. Hyperspectral unmixing algorithm based on spectral information divergence and spectral angle matching[J]. Computer Application, 2015,35(3):844-848. doi: 10.11772/j.issn.1001-9081.2015.03.844
13	凌峰, 吴胜军, 肖飞 . 遥感影像亚像元定位研究综述[J]. 中国图象图形学报, 2011,16(8):1335-1345. doi: 10.13485/j.cnki.11-2089.2014.0047
	Ling Feng, Wu Shengjun, Xiao Fei . Survey of subpixel localization in remote sensing images[J]. China Journal of Image and Graphics, 2011,16(8):1335-1345. doi: 10.13485/j.cnki.11-2089.2014.0047
[14]	Atkinson P M . Mapping sub-pixel boundaries from remotely sensed images[C]. London:Innovations in GIS 4, 1997.
[15]	Krafft C, Diderhoshan M A, Recknagel P , et al. Crisp and soft multivariate methods visualize individual cell nuclei in Raman images of liver tissue sections[J]. Vibrational Spectroscopy, 2011,55(1):90-100. doi: 10.1016/j.vibspec.2010.09.003
[16]	Dias J B. Alternating direction algorithms for constrained sparse regression: application to hyperspectral unmixing [C].Grenoble:First IEEE GRSS Workshop on Hyperspectral Image and Signal Processing, 2009.
[17]	Iordache M D, Bioucas-Diasv J M, Plaza A . Total variation regulatization in sparse hyperspectral unmixing [C]. Lisbon:Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, 2011.
[18]	范明阳 . 基于光谱解混和目标优化的高光谱图像亚像元定位研究[D]. 杭州:杭州电子科技大学, 2016.
	Fan Mingyang . Research on sub-pixel localization of hyperspectral imagery based on optimization of spectral mixture and target optimization[D]. Hangzhou: Hangzhou Dianzi University, 2016.

	NCLS ^[9]	SUnSAL ^[12]	CLSUnSAL ^[21]	GSUnSAL ^[13]	本文算法
20DB	1.051 4	0.384 7	0.273 5	0.272 0	0.237 7
30DB	0.486 5	0.264 6	0.230 6	0.229 2	0.197 5
40DB	0.281 7	0.148 1	0.118 0	0.101 9	0.097 9

	NCLS	SUnSAL	CLSUnSAL	GSUnSAL	Proposed
20DB	-5.864 9	2.867 4	5.830 9	5.826 4	6.143 2
30DB	0.828 5	6.118 7	7.312 7	7.388 9	7.511 2
40DB	5.573 0	11.153 6	13.130 6	14.409 2	14.382 9

	对比算法^[18]	提出的算法
块状数据	0.015 8	0.010 6
小目标数据	0.026 2	0.024 8

基于端元字典稀疏解混的高光谱图像亚像元定位

Hyperspectral Image Sub-Pixel Mapping Based on Sparse Unmixing of Endmember Dictionary

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献 18

相关文章 7

Metrics

本文评价

推荐阅读 0

[1]	骆冲,邬春学. 一种改进的字典学习方法在医疗疾病分析中的应用[J]. 电子科技, 2019, 32(2): 47-50.
[2]	陈益, 李文钧. 基于正交匹配追踪算法的语音信号重构研究[J]. , 2018, 31(3): 44-.
[3]	郑昊辰, 姜维. 基于压缩感知的阅卷系统手写汉字识别算法[J]. , 2018, 31(3): 75-.
[4]	沈丽,韩彦芳. 基于稀疏表示的超分辨率图像重建[J]. , 2015, 28(9): 144-.
[5]	田香玲,席志红. 压缩感知观测矩阵的优化算法[J]. , 2015, 28(8): 102-.
[6]	曹蒙蒙,李新叶,范月坤. 基于判别低秩矩阵恢复和稀疏表示的车标识别[J]. , 2015, 28(4): 57-.
[7]	刘甜甜. 基于稀疏和低秩表示的显著性目标检测[J]. , 2015, 28(2): 112-.