基于光谱库优化学习的光谱超分与分类精度提升

韩晓琳; 张欢; 孙卫东

doi:10.11834/jrs.20210591

模型与方法 | 浏览量 : 0 下载量: 217 CSCD: 0 更多指标

R-PDF
PDF
导出
分享
收藏
专辑

基于光谱库优化学习的光谱超分与分类精度提升
Spectral super-resolution using optimized dictionary learning via spectral library and its effects on classification
2023年27卷第11期页码：2530-2540
纸质出版日期： 2023-11-07 ，
DOI： 10.11834/jrs.20210591

扫描看全文

韩晓琳，张欢，孙卫东.2023.基于光谱库优化学习的光谱超分与分类精度提升.遥感学报，27（11）： 2530-2540

Han X L，Zhang H and Sun W D. 2023. Spectral super-resolution using optimized dictionary learning via spectral library and its effects on classification. National Remote Sensing Bulletin， 27（11）：2530-2540
韩晓琳，张欢，孙卫东.2023.基于光谱库优化学习的光谱超分与分类精度提升.遥感学报，27（11）： 2530-2540 DOI： 10.11834/jrs.20210591.

Han X L，Zhang H and Sun W D. 2023. Spectral super-resolution using optimized dictionary learning via spectral library and its effects on classification. National Remote Sensing Bulletin， 27（11）：2530-2540 DOI： 10.11834/jrs.20210591.

摘要

光谱库优化学习是将光谱库中的光谱数据作为训练样本，在严格理论推导下构建字典优化学习过程。基于光谱库优化学习，本研究提出了一种光谱超分辨率重建方法，该方法在稀疏表示框架下，通过波段匹配，将光谱库映射为与待重建高光谱图像波段相对应的特定光谱库；并利用映射后的特定光谱库与高分多光谱图像，从理论上推导、并构建基于ADMM算法的光谱字典与稀疏系数优化学习过程。多种数据集上的对比分析表明，即使仅使用一幅高分多光谱图像，本研究方法仍能恢复重建出高质量的高分高光谱图像，同时光谱超分辨率重建后的高分高光谱图像可显著提升地物分类精度。结果表明，本研究实现了仅由一幅高分多光谱图像到高分高光谱图像的高质量光谱超分辨率重建。

Abstract

The spectral library can contain the spectral information on the whole types of ground surface objects in the observation area of hyperspectral images. Thus

the optimized dictionary learning via spectral library refers to the process of constructing optimized spectral dictionary under strict theoretical derivation

in which the spectra in the spectral library are used as training samples. The abovementioned process enables the spectra in the hyperspectral image to be sparsely represented under the learned spectral dictionary. To this end

a new spectral super-resolution method using optimized dictionary learning via spectral library is proposed in this study. This method uses only one high spatial multispectral image to reconstruct high spatial hyperspectral image. The aforementioned problem is formulated in the framework of sparse representation

as an estimation of the band matching matrix

the optimized spectral dictionary

and the corresponding sparse coefficients. Specifically

a band matching method is proposed to map the common spectral library to a specific spectral library corresponding to the reconstructed high spatial hyperspectral image. Then

an optimization of spectral dictionary and its corresponding sparse coefficients is derived theoretically using the alternating direction method of multipliers (ADMM) algorithm and by utilizing the abovementioned specific spectral library and the high spatial multispectral image. Comparison results with the relative methods demonstrate that our method not only can achieve a high-quality reconstruction of the high spatial hyperspectral image but also can significantly improve the classification accuracy of multispectral images by even only using one high spatial multispectral image.

We aim to reconstruct high spatial hyperspectral image only from one high spatial multispectral image with high quality.

Three steps of our proposed method are discussed in detail. First

the band matching matrix is estimated using the band wavelength information. Second

the matched spectral dictionary is optimized using the matched spectral library and the high spatial multispectral image. Third

the equivalent sparse coefficient matrix with respect to the matched spectral dictionary is derived theoretically and estimated iteratively.

Extensive experiments and comparative analyses of the proposed method are conducted on various datasets to demonstrate the performance and practical application value of our proposed method. The improvement in classification accuracy on the reconstructed high spatial hyperspectral images is also evaluated using some typical classification methods.

A spectral super-resolution method is proposed

and it uses only one high spatial multispectral image to reconstruct high spatial hyperspectral image. A band matching matrix

which is used to map the common spectral library to a specific spectral library

is obtained by solving the minimum distance problem. A spectral dictionary and its corresponding sparse coefficient matrix are optimized from the matched spectral library and the high spatial hyperspectral image by minimizing augmented Lagrangian function using ADMM iteratively. Experiments on simulated and real datasets demonstrate that our proposed method can produce comparable results for the spectral super-resolution to the other relative state-of-the-art reconstruction or fusion-based methods using additional low spatial hyperspectral image. It can also provide higher reconstruction quality than the HIRSL method without optimization. Our proposed SODL method that uses only one multispectral image may help develop new light and small high spatial hyperspectral imaging equipment.

关键词

光谱超分辨率光谱库稀疏表示优化学习地物分类

Keywords

spectral super-resolutionspectral librarysparse representationoptimized dictionary learninglandcover classification

references

Aiazzi B, Baronti S and Selva M. 2007. Improving component substitution pansharpening through multivariate regression of MS + Pan data. IEEE Transactions on Geoscience and Remote Sensing, 45(10): 3230-3239 [DOI: 10.1109/TGRS.2007.901007http://dx.doi.org/10.1109/TGRS.2007.901007]

Akhtar N, Shafait F and Mian A. 2014. Sparse spatio-spectral representation for hyperspectral image super-resolution\\13th European Conference on Computer Vision. Cham: Springer: 63-78 [DOI: 10.1007/978-3-319-10584-0_5http://dx.doi.org/10.1007/978-3-319-10584-0_5]

Boyd S, Parikh N, Chu E, Peleato B and Eckstein J. 2011. Distributed optimization and statistical learning via the alternating direction method of multipliers. Foundations and Trends® in Machine Learning, 3(1): 1-122 [DOI: 10.1561/2200000016http://dx.doi.org/10.1561/2200000016]

Carper W J, Lillesand T M and Kiefer P W. 1990. The use of intensity-hue-saturation transformations for merging SPOT panchromatic and multispectral image data. Photogrammetric Engineering and Remote Sensing, 56(4): 457-467.

Dong C, Loy C C, He K M and Tang X O. 2014. Learning a deep convolutional network for image super-resolution//13th European Conference on Computer Vision. Zurich: Springer: 184-199 [DOI: 10.1007/978-3-319-10593-2_13http://dx.doi.org/10.1007/978-3-319-10593-2_13]

Dong W S, Fu F Z, Shi G M, Cao X, Wu J J, Li G Y and Li X. 2016. Hyperspectral image super-resolution via non-negative structured sparse representation. IEEE Transactions on Image Processing, 25(5): 2337-2352 [DOI: 10.1109/TIP.2016.2542360http://dx.doi.org/10.1109/TIP.2016.2542360]

Ertürk A, Güllü M K, Çeşmeci D, Gerçek D and Ertürk S. 2014. Spatial resolution enhancement of hyperspectral images using unmixing and binary particle swarm optimization. IEEE Geoscience and Remote Sensing Letters, 11(12): 2100-2104 [DOI: 10.1109/LGRS.2014.2320135http://dx.doi.org/10.1109/LGRS.2014.2320135]

Ertürk A, Iordache M D and Plaza A. 2016. Sparse unmixing-based change detection for multitemporal hyperspectral images. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 9(2): 708-719 [DOI: 10.1109/JSTARS.2015.2477431http://dx.doi.org/10.1109/JSTARS.2015.2477431]

Ghasrodashti E K, Karami A, Heylen R and Scheunders P. 2017. Spatial resolution enhancement of hyperspectral images using spectral unmixing and bayesian sparse representation. Remote Sensing, 9(6): 541 [DOI: 10.3390/rs9060541http://dx.doi.org/10.3390/rs9060541]

Grohnfeldt C, Zhu X X and Bamler R. 2013. Jointly sparse fusion of hyperspectral and multispectral imagery//2013 IEEE International Geoscience and Remote Sensing Symposium. Melbourne: IEEE: 4090-4093 [DOI: 10.1109/IGARSS.2013.6723732http://dx.doi.org/10.1109/IGARSS.2013.6723732]

Han X L, Yu J, Luo J Q and Sun W D. 2019. Reconstruction from multispectral to hyperspectral image using spectral library-based dictionary learning. IEEE Transactions on Geoscience and Remote Sensing, 57(3): 1325-1335 [DOI: 10.1109/TGRS.2018.2866054http://dx.doi.org/10.1109/TGRS.2018.2866054]

Hu J, Li Y S and Xie W Y. 2017. Hyperspectral image super-resolution by spectral difference learning and spatial error correction. IEEE Geoscience and Remote Sensing Letters, 14(10): 1825-1829 [DOI: 10.1109/LGRS.2017.2737637http://dx.doi.org/10.1109/LGRS.2017.2737637]

Huang B, Song H H, Cui H B, Peng J G and Xu Z B. 2014. Spatial and spectral image fusion using sparse matrix factorization. IEEE Transactions on Geoscience and Remote Sensing, 52(3): 1693-1704 [DOI: 10.1109/TGRS.2013.2253612http://dx.doi.org/10.1109/TGRS.2013.2253612]

Iordache M D, Bioucas-Dias J M and Plaza A. 2014. Collaborative sparse regression for hyperspectral unmixing. IEEE Transactions on Geoscience and Remote Sensing, 52(1): 341-354 [DOI: 10.1109/TGRS.2013.2240001http://dx.doi.org/10.1109/TGRS.2013.2240001]

Kawakami R, Matsushita Y, Wright J, Ben-Ezra M, Tai Y W and Ikeuchi K. 2011. High-resolution hyperspectral imaging via matrix factorization//CVPR 2011. Colorado Springs: IEEE: 2329-2336 [DOI: 10.1109/CVPR.2011.5995457http://dx.doi.org/10.1109/CVPR.2011.5995457]

Keshava N and Mustard J F. 2002. Spectral unmixing. IEEE Signal Processing Magazine, 19(1): 44-57 [DOI: 10.1109/79.974727http://dx.doi.org/10.1109/79.974727]

Lan J H, Zou J L, Hao Y S, Zeng Y L, Zhang Y Z and Dong M W. 2018. Research progress on unmixing of hyperspectral remote sensing imagery. Journal of Remote Sensing, 22(1): 13-27

蓝金辉, 邹金霖, 郝彦爽, 曾溢良, 张玉珍, 董铭巍. 2018. 高光谱遥感影像混合像元分解研究进展. 遥感学报, 22(1): 13-27 [DOI: 10.11834/jrs.20186502http://dx.doi.org/10.11834/jrs.20186502]

Li S T, Li C Y and Kang X D. 2021. Development status and future prospects of multi-source remote sensing image fusion. National Remote Sensing Bulletin, 25(1): 148-166

李树涛, 李聪妤, 康旭东. 2021. 多源遥感图像融合发展现状与未来展望. 遥感学报, 25(1): 148-166 [DOI: 10.11834/jrs.20210259http://dx.doi.org/10.11834/jrs.20210259]

Loncan L, De Almeida L B, Bioucas-Dias J M, Briottet X, Chanussot J, Dobigeon N, Fabre S, Liao W Z, Licciardi G A, Simoes M, Tourneret J Y, Veganzones M A, Vivone G, Wei Q and Yokoya N. 2015. Hyperspectral pansharpening: a review. IEEE Geoscience and Remote Sensing Magazine, 3(3): 27-46 [DOI: 10.1109/MGRS.2015.2440094http://dx.doi.org/10.1109/MGRS.2015.2440094]

Lu X Q, Yuan Y and Yan P K. 2014. Alternatively constrained dictionary learning for image superresolution. IEEE Transactions on Cybernetics, 44(3): 366-377 [DOI: 10.1109/TCYB.2013.2256347http://dx.doi.org/10.1109/TCYB.2013.2256347]

Mei S H, Yuan X, Ji J Y, Wan S, Hou J H and Du Q. 2017. Hyperspectral image super-resolution via convolutional neural network//2017 IEEE International Conference on Image Processing (ICIP). Beijing: IEEE: 4297-4301 [DOI: 10.1109/ICIP.2017.8297093http://dx.doi.org/10.1109/ICIP.2017.8297093]

Nezhad Z H, Karami A, Heylen R and Scheunders P. 2016. Fusion of hyperspectral and multispectral images using spectral unmixing and sparse coding. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 9(6): 2377-2389 [DOI: 10.1109/JSTARS.2016.2528339http://dx.doi.org/10.1109/JSTARS.2016.2528339]

Ni D and Ma H B. 2015. Spectral-spatial classification of hyperspectral images based on neighborhood collaboration. Acta Automatica Sinica, 41(2): 273-284

倪鼎, 马洪兵. 基于近邻协同的高光谱图像谱—空联合分类. 自动化学报, 2015, 41(2): 273-284 [DOI: 10.16383/j.aas.2015.c140043http://dx.doi.org/10.16383/j.aas.2015.c140043]

Pan Z X, Yu J, Xiao C B and Sun W D. 2014. Spectral similarity-based super resolution for hyperspectral images. Acta Automatica Sinica, 40(12): 2797-2807

潘宗序, 禹晶, 肖创柏, 孙卫东. 基于光谱相似性的高光谱图像超分辨率算法. 自动化学报, 40(12): 2797-2807 [DOI: 10.3724/SP.J.1004.2014.02797http://dx.doi.org/10.3724/SP.J.1004.2014.02797]

Simões M, Bioucas-Dias J, Almeida L B and Chanussot J. 2015. A convex formulation for hyperspectral image superresolution via subspace-based regularization. IEEE Transactions on Geoscience and Remote Sensing, 53(6): 3373-3388 [DOI: 10.1109/TGRS.2014.2375320http://dx.doi.org/10.1109/TGRS.2014.2375320]

Veganzones M A, Simoes M, Licciardi G, Yokoya N, Bioucas-Dias J M and Chanussot J. 2016. Hyperspectral super-resolution of locally low rank images from complementary multisource data. IEEE Transactions on Image Processing, 25(1): 274-288 [DOI: 10.1109/TIP.2015.2496263http://dx.doi.org/10.1109/TIP.2015.2496263]

Villa A, Chanussot J, Benediktsson J A, Ulfarsson M and Jutten C. 2010. Super-resolution: an efficient method to improve spatial resolution of hyperspectral images//2010 IEEE International Geoscience and Remote Sensing Symposium. Honolulu: IEEE, 2003-2006 [DOI: 10.1109/IGARSS.2010.5654208http://dx.doi.org/10.1109/IGARSS.2010.5654208]

Wang J, Peng J Y, Jiang X Y, Feng X and Zhou J H. 2017. Remote-sensing image fusion using sparse representation with sub-dictionaries. International Journal of Remote Sensing, 38(12): 3564-3585 [DOI: 10.1080/01431161.2017.1302106http://dx.doi.org/10.1080/01431161.2017.1302106]

Wang Z, Bovik A C, Sheikh H R and Simoncelli E P. 2004. Image quality assessment: from error visibility to structural similarity. IEEE Transactions on Image Processing, 13(4): 600-612 [DOI: 10.1109/TIP.2003.819861http://dx.doi.org/10.1109/TIP.2003.819861]

Wei Q, Bioucas-Dias J, Dobigeon N and Tourneret J Y. 2015. Hyperspectral and multispectral image fusion based on a sparse representation. IEEE Transactions on Geoscience and Remote Sensing, 53(7): 3658-3668 [DOI: 10.1109/TGRS.2014.2381272http://dx.doi.org/10.1109/TGRS.2014.2381272]

Xu N, Xiao X Y, Geng X R, You H J and Cao Y G. 2016. Spectral-spatial constrained sparse unmixing of hyperspectral imagery using a hybrid spectral library. Remote Sensing Letters, 7(7): 641-650 [DOI: 10.1080/2150704X.2016.1177240http://dx.doi.org/10.1080/2150704X.2016.1177240]

Xu Y H, Du B, Zhang F and Zhang L P. 2018. Hyperspectral image classification via a random patches network. ISPRS Journal of Photogrammetry and Remote Sensing, 142: 344-357 [DOI: 10.1016/j.isprsjprs.2018.05.014http://dx.doi.org/10.1016/j.isprsjprs.2018.05.014]

Yokoya N, Grohnfeldt C and Chanussot J. 2017. Hyperspectral and multispectral data fusion: a comparative review of the recent literature. IEEE Geoscience and Remote Sensing Magazine, 5(2): 29-56 [DOI: 10.1109/MGRS.2016.2637824http://dx.doi.org/10.1109/MGRS.2016.2637824]

Yokoya N, Yairi T and Iwasaki A. 2012. Coupled nonnegative matrix factorization unmixing for hyperspectral and multispectral data fusion. IEEE Transactions on Geoscience and Remote Sensing, 50(2): 528-537 [DOI: 10.1109/TGRS.2011.2161320http://dx.doi.org/10.1109/TGRS.2011.2161320]

Yuan Y, Zheng X T and Lu X Q. 2017. Hyperspectral image superresolution by transfer learning. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 10(5): 1963-1974 [DOI: 10.1109/JSTARS.2017.2655112http://dx.doi.org/10.1109/JSTARS.2017.2655112]

Zhang K, Wang M, Yang S Y, Xing Y H and Qu R. 2016. Fusion of panchromatic and multispectral images via coupled sparse non-negative matrix factorization. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 9(12): 5740-5747 [DOI: 10.1109/JSTARS.2015.2475754http://dx.doi.org/10.1109/JSTARS.2015.2475754]

Zhao T, Zhang Y F, Xue X Q and He M Y. 2016. Hyperspectral and multispectral image fusion using collaborative representation with local adaptive dictionary pair//2016 IEEE International Geoscience and Remote Sensing Symposium. Beijing: IEEE: 7212-7215 [DOI: 10.1109/IGARSS.2016.7730881http://dx.doi.org/10.1109/IGARSS.2016.7730881]

Zhu J, Zhou L F and Zhang D R. 2011. Identification for building surface material based on hyperspectral remote sensing//2011 19th International Conference on Geoinformatics. Shanghai: IEEE: 1-5 [DOI: 10.1109/GeoInformatics.2011.5980687http://dx.doi.org/10.1109/GeoInformatics.2011.5980687]

Zhu X X, Spiridonova S and Bamler R. 2012. A pan-sharpening algorithm based on joint sparsity//2012 Tyrrhenian Workshop on Advances in Radar and Remote Sensing (TyWRRS). Naples: IEEE: 177-184 [DOI: 10.1109/TyWRRS.2012.6381125http://dx.doi.org/10.1109/TyWRRS.2012.6381125]

文章被引用时，请邮件提醒。

提交

具有分类器机制的高光谱图像特征提取方法

基于时间序列叶面积指数稀疏表示的作物种植区域提取

主题学习和稀疏表示的MODIS图像超分辨率重建

遥感卫星影像K-SVD稀疏表示去噪