超像素分割和波段分割的高光谱图像去噪

doi:10.19665/j.issn1001-2400.20240502

西安电子科技大学学报 ›› 2024, Vol. 51 ›› Issue (5): 122-135.doi: 10.19665/j.issn1001-2400.20240502

• 计算机科学与技术 & 网络空间安全 • 上一篇下一篇

超像素分割和波段分割的高光谱图像去噪

李华君¹(), 蒋俊正²^,³(), 周芳⁴(), 全英汇³()

1.桂林电子科技大学信息与通信学院,广西壮族自治区桂林 541004
2.西安电子科技大学杭州研究院,浙江杭州 311231
3.西安电子科技大学电子工程学院,陕西西安 710071
4.中国计量大学信息工程学院,浙江杭州 310018

收稿日期:2023-11-23 出版日期:2024-05-23 发布日期:2024-05-23
通讯作者: 周芳(1984—),女,副教授,E-mail:zhoufang1026@guet.edu.cn
作者简介:李华君(1999—),男,桂林电子科技大学硕士研究生,E-mail: 22022303063@mails.guet.edu.cn；
蒋俊正(1983—),男,教授,E-mail: jiangjunzheng@xidian.edu.cn；
全英汇(1981—),男,教授,E-mail:yhquan@mail.xidian.edu.cn
基金资助:
广西自然科学杰出青年基金(2021GXNSFFA220004);广西科技基地和人才专项(桂科AD21220112);国家自然科学基金(62261014);国家自然科学基金(62171146);广西创新驱动发展专项(桂科AA21077008);广西研究生教育创新计划项目(YCBZ2023137)

Hyperspectral image denoising based on superpixel segmentation and band segmentation

LI Huajun¹(), JIANG Junzheng²^,³(), ZHOU Fang⁴(), QUAN Yinghui³()

1. School of Information and Communication,Guilin University of Electronic Technology,Guilin 541004,China
2. Hangzhou Institute of Technology,Xidian University,Hangzhou 311231,China
3. School of Electronic Engineering,Xidian University,Xian 710071,China
4. College of Information Engineering,ChinaJiliang University,Hangzhou 310018,China

Received:2023-11-23 Online:2024-05-23 Published:2024-05-23

摘要/Abstract

摘要：

针对现有的高光谱图像去噪算法采用逐波段或者全波段方式去噪,未能充分利用高光谱图像波段相似性的问题,提出了超像素分割和波段分割的高光谱图像去噪算法。文中将构建双层图模型,包括上层图和下层图模型。首先,对高光谱图像应用超像素分割技术,得到一系列的超像素。对超像素内的像素建模为节点,像素之间用边连接,构建一系列下层图,从而充分利用高光谱图像的空间信息和保留边界信息。根据超像素分割结果,沿着波段维分割,形成超像素体,以充分利用高光谱图像的波段相似性。将超像素体建模为节点,超像素体之间用边连接,构建上层图。基于构建的图结构和图分割方式,将高光谱图像去噪问题归结为一系列的优化问题,在优化问题中利用克罗内克乘积图重新定义了图拉普拉斯正则项。最后,实验结果表明,与现有算法相比,文中所提算法具有更高的平均峰值信噪比、平均结构相似性和光谱差异性。

关键词: 高光谱图像去噪, 图信号处理, 超像素分割, 波段分割, 图拉普拉斯正则项

Abstract:

Existing hyperspectral image denoising algorithms adopt a band-by-band or full-band approach to denoising,which fails to make full use of the similarity of hyperspectral image bands.To address this problem,this paper proposes a hyperspectral image denoising algorithm based on superpixel segmentation and band segmentation.In this paper,we construct a two-layer graph,including the upper and lower layer graphs.First,superpixel segmentation is applied to the hyperspectral image to obtain a series of superpixels.In order to utilize the spatial information on the hyperspectral image and retain the boundary information,the pixels within the superpixels are modeled as nodes with the pixels connected with edges to construct a series of lower layer graphs.In order to utilize the band similarity of the hyperspectral image,superpixel volumes are formed by segmenting along the band dimension based on the superpixel segmentation results with the superpixel volumes modeled as nodes,and the superpixel volumes connected with edges to construct an upper layer graph.Based on the graph structure and graph segmentation,the hyperspectral image denoising problem is reduced to a series of optimization problems,in which the graph Laplacian regularization is redefined using the Kronecker graph product.Finally,experimental results show that the proposed algorithm has a higher mean signal-to-noise ratio,mean structural similarity index measure and erreur relative globale adimensionnelle de synthese compared with the existing algorithms.

Key words: hyperspectral image denoising, graph signal processing, superpixel segmentation, band segmentation, graph Laplacian regularization

中图分类号:

TP751

李华君, 蒋俊正, 周芳, 全英汇. 超像素分割和波段分割的高光谱图像去噪[J]. 西安电子科技大学学报, 2024, 51(5): 122-135.

LI Huajun, JIANG Junzheng, ZHOU Fang, QUAN Yinghui. Hyperspectral image denoising based on superpixel segmentation and band segmentation[J]. Journal of Xidian University, 2024, 51(5): 122-135.

图/表 20

图1

图2

图3

图4

图5

图6

表1

图7

图8

图9

图10

图11

图12

图13

图14

图15

图16

图17

图18

图19

参考文献 24

[1]	KHAN M J, KHAN H S, YOUSAF A, et al. Modern Trends in Hyperspectral Image Analysis:A Review[J]. IEEE Access, 2018, 6:14118-14129.
[2]	CAI W Y, JIANG J Z, OUYANG S. Hyperspectral Image Denoising Using Adaptive Weight Graph Total Variation Regularization and Low-Rank Matrix Recovery[J]. IEEE Geoscience and Remote Sensing Letters, 2022, 19:1-5.
[3]	BIOUCAS-DIAS J M, PLAZA A, CAMPS-VALLS G, et al. Hyperspectral Remote Sensing Data Analysis and Future Challenges[J]. IEEE Geoscience and Remote Sensing Magazine, 2013, 1(2):6-36.
[4]	ZENG H J, HUANG S G, CHEN Y Y, et al. All of Low-Rank and Sparse:A Recast Total Variation Approach to Hyperspectral Denoising[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2023, 16:7357-7373.
[5]	ATKINSON I, KAMALABADI F, JONES D L. Wavelet-Based Hyperspectral Image Estimation[C]// IGARSS 2003.2003 IEEE International Geoscience and Remote Sensing Symposium.Proceedings(IEEE Cat.No.03CH37477).Piscataway:IEEE, 2003:743-745.
[6]	ZHANG H. Hyperspectral Image Denoising with Cubic Total Variation Model[J]. ISPRS Annals of the Photogrammetry,Remote Sensing and Spatial Information Sciences, 2012:95-98.
[7]	WANG Y, PENG J, ZHAO Q, et al. Hyperspectral Image Restoration via Total Variation Regularized Low-Rank Tensor Decomposition[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2018, 11(4):1227-1243.
[8]	ZHANG H, HE W, ZHANG L, et al. Hyperspectral Image Restoration Using Low-Rank Matrix Recovery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2014, 52(8):4729-4743.
[9]	蔡明娇, 蒋俊正, 蔡万源, 等. 张量分解和自适应图全变分的高光谱图像去噪[J]. 西安电子科技大学学报, 2024, 51(2):157-169.
	CAI Mingjiao, JIANG Junzheng, CAI Wanyuan, et al. Hyperspectral Image Denoising Based on Tensor Decomposition and Adaptive Weight Graph Total Variation[J]. Journal of Xidian University, 2024, 51(2):157-169.
[10]	HE C, SUN L, HUANG W, et al. TSLRLN:Tensor Subspace Low-Rank Learning with Non-Local Prior for Hyperspectral Image Mixed Denoising[J]. Signal Processing, 2021, 184:108060.
[11]	HE W, YAO Q, LI C, et al. Non-Local Meets Global:An Iterative Paradigm for Hyperspectral Image Restoration[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44(4):2089-2107.
[12]	LIU Y, SHAN C, GAO Q, et al. Hyperspectral Image Denoising via Minimizing the Partial Sum of Singular Values and Superpixel Segmentation[J]. Neurocomputing, 2019, 330:465-482.
[13]	BEHROOZI Y, YAZDI M, ASLI A Z. Hyperspectral Image Denoising Based on Superpixel Segmentation Low-Rank Matrix Approximation and Total Variation[J]. Circuits Systems and Signal Processing, 2022, 41(6):3372-3396.
[14]	陈俊杰, 邓洪高, 马谋, 等. GCN-GRU:一种无线传感器网络故障检测模型[J]. 西安电子科技大学学报, 2022, 49(5):60-67.
	CHEN Junjie, DENG Honggao, MA Mou, et al. GRN-GRU:AFault Detection Model for Wireless Sensor Networks[J]. Journal of Xidian University, 2022, 49(5):60-67.
[15]	DUDGEON D E, MERSEREAU R M. Multidimensional Digital Signal Processing[M]. Englewood Cliffs: Prentice-Hall,1983.
[16]	TAN Z, ZHU Y, LIU B. Learning Spatial-Temporal Feature with Graph Product[J]. Signal Processing, 2023, 210:109062.
[17]	HU W, PANG J, LIU X, et al. Graph Signal Processing for Geometric Data and Beyond:Theory and Applications[J]. IEEE Transactions on Multimedia, 2022, 24:3961-3977.
[18]	ACHANTA R, SHAJI A, SMITH K, et al. SLIC Superpixels Compared to State-of-the-Art Superpixel Methods[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(11):2274-2282. pmid: 22641706
[19]	DONG W, WOZNIAK M, WU J S, et al. Denoising Aggregation of Graph Neural Networks by Using Principal Component Analysis[J]. IEEE Transactions on Industrial Informatics, 2023, 19(3):2385-2394.
[20]	CAI J F, CANDèS E J, SHEN Z. A Singular Value Thresholding Algorithm for Matrix Completion[J]. SIAM Journal on Optimization, 2010, 20(4):1956-1982.
[21]	ZHUANG L, FU X, NG M K, et al. Hyperspectral Image Denoising Based on Global and Nonlocal Low-Rank Factorizations[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 59(12):10438-10454.
[22]	HYDICE Image of Washington DC Mall(2023)[DB/OL].[2023-08-01]. https://engineering.purdue.edu/-biehl/MultiSpec/hyperspectral.html.
[23]	AVIRIS Image of Cuprite(2023)[DB/OL].[2023-08-01]. https://www.ehu.eus/ccwintco/index.php/Hyperspectral_Remote_Sensing_Scenes.
[24]	ZHANG H Y, CHEN H Y, YANG G Y, et al. LR-Net:Low-Rank Spatial-Spectral Network for Hyperspectral Image Denoising[J]. IEEE Transactions on Image Processing, 2021, 30:8743-8758.

不同情史	Level	Indexes	LRMR	LRTDGTV	NGmeet	GLF	TSLRLN	KGLRSSBS
情况1	G=0.025	MPSNR	39.80	36.41	32.33	32.16	32.22	42.06
	P=0.05	MSSIM	0.981 0	0.954 8	0.939 1	0.936 9	0.938 5	0.988 7
		ERGAS	44.05	63.34	105.90	104.38	104.77	34.36
情况2	G=0.1	MPSNR	30.45	30.82	21.38	21.42	21.53	32.01
	P=0.2	MSSIM	0.883 0	0.847 8	0.702 8	0.715 7	0.734 6	0.911 9
		ERGAS	128.25	125.56	398.51	378.88	378.70	105.48
情况3	Gaussian+	MPSNR	30.83	31.10	25.08	25.46	25.47	32.66
	impulse	MSSIM	0.886 8	0.856 1	0.806 4	0.821 5	0.824 9	0.920 6
		ERGAS	128.70	126.60	282.46	268.04	278.18	101.20
情况4	Gaussian	MPSNR	30.72	30.78	25.10	25.39	25.51	32.40
	+impulse+	MSSIM	0.885 9	0.851 1	0.805 0	0.818 0	0.824 0	0.919 1
	deadlines	ERGAS	129.47	134.10	281.60	268.52	277.27	104.07

超像素分割和波段分割的高光谱图像去噪

Hyperspectral image denoising based on superpixel segmentation and band segmentation

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 20

参考文献 24

相关文章 2

Metrics

本文评价

推荐阅读 0

[1]	蔡明娇, 蒋俊正, 蔡万源, 周芳. 张量分解和自适应图全变分的高光谱图像去噪[J]. 西安电子科技大学学报, 2024, 51(2): 157-169.
[2]	杨杰,蒋俊正. 利用联合图模型的传感器网络数据修复方法[J]. 西安电子科技大学学报, 2020, 47(1): 44-51.