Please wait a minute...
img

官方微信

遥感技术与应用  2019, Vol. 34 Issue (6): 1305-1314    DOI: 10.11873/j.issn.1004-0323.2019.6.1305
数据与图像处理     
由粗到细的高光谱图像多端元光谱混合分析
左成欢1(),赵辽英1,陆海强2,厉小润3()
1.杭州电子科技大学计算机应用技术研究所,浙江 杭州 310018
2.嘉兴市恒创电力设备有限公司,浙江 嘉兴 314033
3.浙江大学电气工程学院,浙江 杭州 310027
A Corse-to-Fine Scheme for Multiple Endmember Spectral Mixture Analysis of Hyperspectral Images
Chenhuan Zuo1(),Liaoying Zhao1,Haiqiang Lu2,Xiaorun Li3()
1.Institute of Computer Application Technology,Hangzhou Dianzi University,Hangzhou 310018,China
2.Jiaxing Hengchuang Electric Power Equipment Co. LTD,Jiaxing 314033,China
3.College of Electrical Engineering, Zhejiang University,Hangzhou 310027,China
 全文: PDF(5243 KB)   HTML
摘要:

光谱可变性是影响高光谱图像光谱混合分析精度的重要因素,多端元光谱混合分析是解决该问题的有效手段。为了降低光谱混合分析时间复杂度的同时提高其精度,提出了一种由粗到细的多端元光谱混合分析算法,该算法首先基于扩展的端元集对每个像元进行全约束光谱混合粗分析,确定含所有地物的初始端元集,在此基础上进一步进行精细光谱混合分析,迭代光谱混合分析构建端元子集,最终根据重构误差变化量确定各个像元的最优端元集。实验结果表明:相比迭代光谱混合分析法和分层多端元光谱混合分析法,所提出的由粗到细的高光谱图像多端元光谱混合分析能有效降低算法反演丰度误差并改善计算效率。

关键词: 高光谱图像多端元光谱混合分析重构误差变化量    
Abstract:

Spectral variability is an important factor which influences the accuracy of spectral analysis in hyperspectral images. Multiple endmembers spectral mixture analysis is an effective method to solve this problem. In order to reduce the time complexity of spectral mixing analysis and improve the accuracy in the same time, a multiple endmember spectral mixture analysis algorithm based on corse-to-fine scheme is proposed. Based on the extended endmember set for each pixel, the proposed algorithm firstly make fully-constrained spectral mixing coarse analysis to determine the initial set of end-members containing all land cover material. On this basis, the algorithm further conducts fine spectral mixture analysis, iterative spectral mixture analysis to build end-member subsets and the optimal end-member set is finally determined according to the variation of reconstruction error. The experimental results show that compared with the iterative spectral mixture analysis method and the hierarchical multi-endmember spectral mixture analysis algorithm, the proposed algorithm reduces the error of inversion abundance and improves computational efficiency greatly.

Key words: Hyperspectral images    Multiple endmembers    Spectral mixture analysis    The variation of reconstruction error
收稿日期: 2018-07-06 出版日期: 2020-03-23
ZTFLH:  TP751.1  
基金资助: 国家自然科学基金项目(61671408);教育部联合基金项目(6141A02022314)
通讯作者: 厉小润     E-mail: 2685119427@qq.com;lxr@zju.edu.cn
作者简介: 左成欢(1992-),女,安徽安庆人,硕士研究生,主要从事高光谱图像光谱混合分析方面的研究。E?mail: 2685119427@qq.com
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
左成欢
赵辽英
陆海强
厉小润

引用本文:

左成欢,赵辽英,陆海强,厉小润. 由粗到细的高光谱图像多端元光谱混合分析[J]. 遥感技术与应用, 2019, 34(6): 1305-1314.

Chenhuan Zuo,Liaoying Zhao,Haiqiang Lu,Xiaorun Li. A Corse-to-Fine Scheme for Multiple Endmember Spectral Mixture Analysis of Hyperspectral Images. Remote Sensing Technology and Application, 2019, 34(6): 1305-1314.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2019.6.1305        http://www.rsta.ac.cn/CN/Y2019/V34/I6/1305

图1  CFSMA算法流程图
图2  CFSMA算法解混示意图
图3  4类端元光谱曲线
图4  合成数据(第3波段)
图5  4类端元的丰度图
算法明矾石橄榄石绿泥石白云母平均
sSMA0.120.060.140.100.10
CFSMA0.030.030.030.030.03
ISMA0.030.040.040.030.04
HMESMA0.040.050.040.030.04
表1  解混丰度误差对比
图6  丰度误差直方图
图7  4类端元丰度图对比至上向下分别表示明矾石、橄榄石、绿泥石和白云母
图8  不同信噪比下RMSE值
端元组合数算法明矾石橄榄石绿泥石白云母均值
6sSMA0.120.060.140.100.10
CFSMA0.030.030.030.030.03
ISMA0.030.040.040.030.04
HMESMA0.040.050.040.030.04
9sSMA0.120.070.120.100.10
CFSMA0.020.030.030.020.03
ISMA0.020.040.040.020.03
HMESMA0.030.040.040.020.03
12sSMA0.100.110.100.110.11
CFSMA0.010.020.030.020.02
ISMA0.020.030.030.020.03
HMESMA0.020.040.040.020.03
15sSMA0.110.100.130.100.11
CFSMA0.010.030.020.010.02
ISMA0.020.020.030.010.02
HMESMA0.020.030.030.020.03
表2  4种解混算法与端元组合数目的丰度误差比较
图9  平均丰度误差曲线图
端元组合数目691215
sSMA0.340.360.350.35
CFSMA4.724.734.734.73
ISMA27.3631.1336.1842.25
HMESMA3.023.123.123.13
表3  4种解混算法与端元组合数目的时间比较(单位:s)
图10  Cuprite矿物数据(第20波段)
图11  Cuprite 8种端元丰度图(左边为CFSMA解混丰度图,中间为ISMA解混丰度图,右边为HMESMA)
图12  3种算法重构误差直方图
解混算法CFSMAISMAHMESMA
时间/s148.165100.35112.61
表4  3种解混算法消耗时间对比
1 Tong Qingxi, Zhang Bing, Zheng Lanfen. Hyperspectral Remote Sensing[M]. Beijing: Higher Education Press, 2006.
1 童庆禧,张兵,郑兰芬.高光谱遥感-原理、技术与应用[M]. 北京:高等教育出版社, 2006.
2 Liu Ailin, Guo Baoping, Li Yanshan. Piecewise Convex Mulutiple-model Hypersupectral Imagery END-member-extraction based on Discrete Particle Swarm Optimization[J]. Remote Sensing Technology and Application, 2018, 33( 2): 227- 232.
2 刘爱林, 郭宝平, 李岩山. 基于离散粒子群算法的凸多模态高光谱图像端元提取研究[J]. 遥感技术与应用, 2018, 33( 2): 227- 232.
3 Halimi A, Honeine P, Bioucas-Dias J M. Hyperspectral Unmixing in Presence of Endmember Variability, Nonlinearity, or Mismodeling Effects[J]. IEEE Transactions on Image Processing, 2016, 25( 10): 4565- 4579.
4 Li Hengkai, Bin Ou, Liu Yuting, et al. Citrus Recognition Methods of Hyperspectral Remote Sensing Image based on Spectral Unmixing Model[J]. Remote Sensing Technology and Application, 2017, 32( 4): 743- 750.
4 李恒凯, 欧彬, 刘雨婷, 等. 基于混合像元分解的高光谱影像柑橘识别方法[J]. 遥感技术与应用, 2017, 32( 4): 743- 750.
5 Chang C I, Zhao X L, Althouse M L G, et al. Least Squares Subspace Projection Approach to Mixed Pixel Classification for Hyperspectral Images[J]. IIEEE Transactions on Geoscience & Remote Sensing, 1998, 36( 3): 898- 912.
6 Heinz D C, Chang C. Fully Constrained Least Squares Linear Spectral Mixture Analysis Method for Material Quantification in Hyperspectral Imagery[J]. IEEE Transactions on Geoscience & Remote Sensing, 2002, 39( 3): 529- 545.
7 Roberts D A, Gardner M, Church R, et al. Mapping Chaparral in the Santa Monica Mountains Using Multiple Endmember Spectral Mixture Models.[J]. Remote Sensing of Environment, 1998, 65( 3): 267- 279.
8 Rogge D M, Rivard B, Zhang J, et al. Iterative Spectral Unmixing for Optimizing Per-Pixel Endmember Sets[J]. IEEE Transactions on Geoscience and Remote Sensing, 2006, 44( 12): 3725- 3736.
9 Zhao Chunhui, Cui Shiling, Liu Wu. Multi-endmember Hierarchical Mixture Analysis for Spectra[J]. Journal of OptoelectronicsLaser, 2014( 9): 1830- 1836.
9 赵春晖, 崔士玲, 刘务. 基于分层的多端元光谱解混算法[J]. 光电子·激光, 2014( 9): 1830- 1836.
10 Tits L, Heylen R, Somers B, et al. A Geometric Unmixing Concept for the Selection of Optimal Binary Endmember Combinations[J]. IEEE Geoscience and Remote Sensing Letters, 2015, 12( 1): 82- 86.
11 Ghaffari O, Zoej M J V, Mokhtarzade M. Reducing the Effect of the Endmembers' Spectral Variability by Selecting the Optimal Spectral Bands[J]. Remote Sensing, 2017, 9( 9): 884. doi: 10.3390/rs9090884.
doi: 10.3390/rs9090884
12 Zhuang Lina. Hyperspectral Mixture Analysis Incorporating Endmember Variability[D]. Beijing: Chinese Academy of Sciences, 2015.
12 庄丽娜. 基于端元变异性分析的高光谱混合像元分解研究[D]. 北京:中国科学院大学, 2015,
13 Zou S, Zare A. Hyperspectral Unmixing with Endmember Variability Using Partial Membership Latent Dirichlet Allocation[C]// IEEE International Conference on Acoustics, Speech and Signal Processing, 2017: 6200- 6204.
14 Zhou Y, Rangarajan A, Gader P D. A Gaussian Mixture Model Representation of Endmember Variability in Hyperspectral Unmixing.[J]. IEEE Transactions on Image Processing, 2017, (99).
15 Thouvenin P A G, Dobigeon N, Tourneret J Y. A Hierarchical Bayesian Model Accounting for Endmember Variability and Abrupt Spectral Changes to Unmix Multitemporal Hyperspectral Images[J]. IEEE Transactions on Computational Imaging, 2016, 99: 1- 1.
16 Duan Jinliang, Wang Jie, Wen Xingyue. Comparison Analysis between Two Spectral Mixture Analysis Methods of Incorporating Endmember Variab[J]. Resource Development&Market2017, 33( 6): 651- 655.
16 段金亮, 王杰, 文星跃. 基于端元变化的两种混合像元分解算法比较研究[J]. 资源开发与市场, 2017, 33( 6): 651- 655.
17 Tromp M, Epema G F. Spectral Mixture Analysis for Mapping Land Degradation in Semi-arid Areas[J]. Geologie En Mijnbouw, 1998, 77( 2): 153- 160.
18 Meiping So, Zhang Yongrong, An Jubai, et al. Effective Endmembers based Bilinear Unmixing Model[J]. Spectroscopy and Spectral Analysis, 2014, 34( 1): 196- 200.
18 宋梅萍, 张甬荣, 安居白, 等. 基于有效端元集的双线性解混模型[J]. 光谱学与光谱分析, 2014, 34( 1): 196- 200.
19 Ding Haiyong, Shi Wenzhong. Fast N-FINDR Algorithm for Endmember Extraction based on Chi-square Distribution[J]. Journal of Remote Sensing, 2013, 17( 1): 122- 137.
19 丁海勇, 史文中. 利用卡方分布利用卡方分布改进N-FINDR端元提取算法[J]. 遥感学报, 2013, 17( 1): 122- 137.
20 Wu Bo, Zhou Xiaocheng, Zhao Yindi. Study on the Relationships between Endmember Variance and Decomposition Accuracy of Mixture Pixel[J]. Remote Sensing Information2007( 3): 3- 7.
20 [ 吴波, 周小成, 赵银娣. 端元光谱变化与混合像元分解精度的关系研究[J]. 遥感信息, 2007( 3): 3- 7.
21 Nascimento J M P, Dias J M B. Vertex Component Analysis: A Fast Algorithm to Unmix Hyperspectral Data[J]. IEEE Transactions on Geoscience and Remote Sensing, 2005, 43( 4): 898- 910.
[1] 李哲, 宫兆宁, 刘先林, 关晖, 王颖. 基于面向对象多端元混解模型的植被覆盖度反演及其时空分布研究[J]. 遥感技术与应用, 2018, 33(6): 1149-1158.
[2] 刘爱林,郭宝平,李岩山 . 基于离散粒子群算法的凸多模态高光谱图像端元提取研究[J]. 遥感技术与应用, 2018, 33(2): 227-232.
[3] 王光镇,王静璞,邹学勇,韩柳,宗敏. 遥感技术估算非光合植被覆盖度研究综述[J]. 遥感技术与应用, 2018, 33(1): 1-9.
[4] 张霞,孙伟超,帅通,孙艳丽. 基于小波变换的图像条带噪声去除方法[J]. 遥感技术与应用, 2015, 30(6): 1168-1175.
[5] 张文婷,金可懿,宋开山,杭艳红. 基于多端元混合光谱模型与Landsat影像的北京不透水层动态研究[J]. 遥感技术与应用, 2015, 30(2): 321-330.
[6] 袁宗泽,孙浩,计科峰,邹焕新. 基于Fisher字典学习稀疏表示的高光谱图像分类[J]. 遥感技术与应用, 2014, 29(4): 646-652.
[7] 李 桢,谭永滨,李 霖,于忠海,兰洪浩. 基于线性光谱混合分析的武汉市地表组分研究[J]. 遥感技术与应用, 2013, 28(5): 780-784.
[8] 高利鹏,赵华亮,刘明翔,张华. 基于遥感影像不透水层估算的震后城区损坏面积评估[J]. 遥感技术与应用, 2013, 28(4): 582-587.
[9] 刘正春,曾永年,何丽丽,吴孔江,靳文凭. 基于光谱归一化的变组分光谱混合分析(NMESMA)方法及其应用[J]. 遥感技术与应用, 2012, 27(2): 159-167.
[10] 李晓雪,安如,吴红,杨仁敏,曲春梅,陆玲,龚天宇. 基于一种改进的多端元混合像元分解方法的三江源中东部地区植被盖度信息的提取[J]. 遥感技术与应用, 2011, 26(3): 383-391.
[11] 潘竟虎,刘春雨,李晓雪. 基于混合光谱分解的兰州城市热岛与下垫面空间关系分析[J]. 遥感技术与应用, 2009, 24(4): 462-468.
[12] 谢虹,杨太保. 基于Landsat 的城市热特征研究——以兰州市为例[J]. 遥感技术与应用, 2008, 23(6): 677-681.
[13] 李俊杰,何隆华,戴锦芳,李杏朝,傅俏燕. 基于不透水地表比例的城市扩展研究[J]. 遥感技术与应用, 2008, 23(4): 424-427.
[14] 张文杰,张钧萍,张晔. 基于区域活动轮廓模型的高光谱图像分割方法[J]. 遥感技术与应用, 2008, 23(3): 351-355.
[15] 况军, 罗建书, 向露. 基于自适应预测的高光谱遥感图像无损压缩算法[J]. 遥感技术与应用, 2007, 22(6): 739-742.