Please wait a minute...
img

官方微信

遥感技术与应用  2019, Vol. 34 Issue (4): 793-798    DOI: 10.11873/j.issn.1004-0323.2019.4.0793
数据与图像处理     
基于结构性字典学习的毛儿盖遥感图像去噪研究
秦振涛1(),杨茹2
1. 攀枝花学院 数学与计算机学院,四川 攀枝花 617000
2. 攀枝花学院 土木与建筑工程学院,四川 攀枝花 617000
Remote Sensing Image of Mao'ergai Denoising based on Structured Dictionary Learning
Zhentao Qin1(),Ru Yang2
1. School of Mathematics and Computer Science, Panzhihua College, Panzhihua, 617000, China
2. School of Civil and Architecture Engineering, Panzhihua College, Panzhihua, 617000, China
 全文: PDF(5600 KB)   HTML
摘要:

遥感图像的噪声分析、评估和滤波作为遥感图像处理的研究重点而一直受到遥感应用领域的关注。为了进一步提高遥感图像的去噪能力,提出一种新的基于聚类的组稀疏字典学习多光谱遥感图像去噪算法,该算法能够综合利用多光谱遥感图像的空间局部性和光谱的全局性,对遥感图像像素进行聚类后划分为不同的组,然后通过字典学习获得多光谱遥感图像的空间、光谱字典和系数。经过阈值处理后,对空间相似的块进行平均处理,实现了对多光谱遥感图像的去噪。该算法用于岷江上游植被和土壤类型典型地区——毛儿盖实验区遥感图像的去噪,峰值信噪比相比band-wise K-SVD算法提高了7.6%左右,同时具有更好的视觉效果。

关键词: 遥感图像结构性字典学习去噪聚类    
Abstract:

The noise analysis, evaluation and denoising of remote sensing image are the focus of RSI processing. In order to improve the denoising ability of remote sensing image, presents a new structured dictionary-based method for multispectral image denoising based on cluster. This method incorporates both the locality of spatial and the correlation across spectrum of multispectral image. Remote sensing image was divided into different groups by clustering, and sparse representation coefficients of spatial and spectral and dictionary is obtained according to the dictionary learning algorithm. After threshold processing, the similar blocks are averaged and realized with multispectral remote sensing image denoising. This algorithm is applied to the denoising of remote sensing image of typical vegetation and soil types in the upper reaches of Minjiang river- Maoergai experimental area. Compared with the band-wise K-SVD algorithm, the PSNR of this algorithm can be improved by about 7.6%, with better visual effect.

Key words: Remote sensing image    Structured dictionary learning    Denoising    Cluster
收稿日期: 2018-12-11 出版日期: 2019-10-16
ZTFLH:  TP751  
基金资助: 国家自然科学基金项目(41372340);国土资源部地学空间信息技术重点实验室开放基金项目(KLGSIT2016?10);攀枝花市科技项目(2018CY?G?28)
作者简介: 秦振涛(1982-),男,陕西榆林人,博士,副教授,主要从事机器学习、3S技术以及计算机应用等方面的研究。E?mail:qinzt1982@163.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  
秦振涛
杨茹

引用本文:

秦振涛,杨茹. 基于结构性字典学习的毛儿盖遥感图像去噪研究[J]. 遥感技术与应用, 2019, 34(4): 793-798.

Zhentao Qin,Ru Yang. Remote Sensing Image of Mao'ergai Denoising based on Structured Dictionary Learning. Remote Sensing Technology and Application, 2019, 34(4): 793-798.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2019.4.0793        http://www.rsta.ac.cn/CN/Y2019/V34/I4/793

图1  研究区毛儿盖位置图(2007年TM321真彩色影像)
图2  高光谱遥感图像像素被划分不同的空间块
图3  各种去噪算法的去噪结果图
加噪图像固定k=5,噪声标准差δ
0.100.150.200.30
含噪图像14.3113.0311.739.31
Bw KSVD20.3919.5718.8017.38
Bw BM3D22.8022.2421.6920.69
Integral KSVD21.3420.4219.4117.80
3DNLM20.6620.3119.9317.72
PARAFAC14.6814.7314.7414.73
Zhao的算法20.9920.4819.9519.01
本文算法22.8922.2521.7020.86
表2  不同算法的去噪性能比较
方法BM3DIntegral KSVD3DNLMZhao算法PARAFAC本文算法
时间2.0430.21764.257.794.2025.12
表3  不同算法的执行时间比较
图4  去噪效果对比图
1 XuS, ZhouY, XiangH, et al. Remote Sensing Image Denoising Using Patch Grouping-based Nonlocal Means Algorithm[J]. IEEE Geoscience and Remote Sensing Letters, 2017:14(12):2275-2279.
2 WangR. Combining Interior and Exterior Characteristics For Remote Sensing Image Denoising[J]. Journal of Applied Remote Sensing, 2016, 10(2):025016. doi:10.1117/1.JRS.10/025016.
doi: 10.1117/1.JRS.10/025016
3 HuangF, LanB, TaoJ, et al. A Parallel Nonlocal Means Algorithm for Remote Sensing Image Denoising on an Intel Xeon Phi Platform[J]. IEEE Access, 2017:1-1. doi:10.1109/ACCESS.2017.2696362.
doi: 10.1109/ACCESS.2017.2696362
4 WangXianghai, ZhangHongwei, LiFang. A PDE-based Hybrid Model for De-noising Remote Sensing Image with Gaussian and Salt-pepper Noise [J]. Acta Geodaetica et Cartographica Sinica, 2010(3): 283-288.
4 王相海,张洪为,李放. 遥感图像高斯与椒盐噪声的PDE混合去噪模型研究[J]. 测绘学报, 2010(3): 283-288.
5 AharonM, EladM, BrucksteinA. K-SVD: An Algorithm for Designing Overcomplete Dictionaries for Sparse Representation[J]. IEEE Transactions on Signal Processing, 2006, 54(11): 4311-4322.
6 DabovK, FoiA, KatkovnikV, et al. Image Denoising by Sparse 3-D Transform-Domain Collaborative Filtering[J]. IEEE Transactions on Image Processing, 2007, 16(8):2080-2095.
7 ManjónJ, CoupéP, Martí-BonmatíL, et al. Adaptive Non-Local Means Denoising of MR Images with Spatially Varying Noise Levels[J]. Journal of Magnetic Resonance Imaging,2010, 1(31): 192-203.
8 LiuX F, BourennaneS, FossatiC. Denoising of Hyperspectral Images Using the Parafac Model and Statistical PerforMance Analysis[J]. IEEE Transactions on Geoscience and Remote Sensing, 2012, 10(50): 3717-3724.
9 ZhaoY, YangJ. Hyperspectral Image Denoising via Sparse Representation and Low-rank Constraint[J]. IEEE Transactions on Geoscience and Remote Sensing. 2015, 1(51): 296-308.
10 ZhouXia, ZhuoZhihong, YaoMin. Hyperspectral Remote Image Denoising based on Multi-task Nonnegative Dictionary Learning [J]. Control Engineering of China, 2017,24(12): 2544-2548.
10 周霞,卓志宏,姚敏. 多任务非负字典学习的高光谱遥感图像去噪[J]. 控制工程,2017,24(12):2544-2548.
11 WangXiaoyan, ChiTianhe. A Satellite Image Denoising Method based on MPSO and Dictionary Learning [J]. Computer Engineering and Science,2017,39(9):1675-1681.
11 王晓燕,池天河. 结合变异粒子群和字典学习的遥感影像去噪[J]. 计算机工程与科学,2017,39(9):1675-1681.
12 SongXiangfa, JiaoLicheng. Classification of Hyperspectral Remote Sensing Image based on Sparse Representation and Spectral Information[J]. Journal of Electronics & Information Technology, 2012(2): 268-272.
12 宋相法,焦李成. 基于稀疏表示及光谱信息的高光谱遥感图像分类[J]. 电子与信息学报, 2012(2): 268-272.
13 SongLin, ChengYongmei, ZhaoYongqiang. Hyper-spectromicroscope Classification based on Sparse Representation Model and Auto-regressive Model [J]. Acta Optica Sinica, 2012(3): 322-328.
13 宋琳,程咏梅,赵永强. 基于稀疏表示模型和自回归模型的高光谱分类[J]. 光学学报, 2012(3): 322-328.
14 XiaQin, XingShuai ,MaDongyang,et al. An Improved K-SVD-based Denoising Method for Remote Sensing Satellite Images [J]. Journal of Remote Sensing,2016,20(3):441-449.
14 夏琴,邢帅,马东洋,莫德林,等. 遥感卫星影像K-SVD稀疏表示去噪[J]. 遥感学报,2016,20(3):441-449.
15 ZhangCongmei, SunQuansen, WangChao,et al. Remote Sensing Image De-noising via Sparse Representation based on Non-local Self-similarity [J]. Remote Sensing Technology and Application, 2016,31(4):739-747.
15 张从梅,孙权森,王超,等. 基于非局部自相似性的遥感图像稀疏去噪算法[J]. 遥感技术与应用,2016, 31(4): 739-747.
16 PengY, MengD, XuZ, et al. Decomposable Nonlocal Tensor Dictionary Learning form Multispectral Image Denoising[C]∥ 2014 IEEE Conference on Computer Vision and Pattern Recognition, 2014.
17 CichockiA, MandicD P, PhanA H, et al. Tensor Decompositions for Signal Processing Applications from Two-way to Multiway Component Analysis[J]. IEEE Signal Processing Magazine, 2014, 32(2):145-163.
18 ArthurD. k-means++ : The Advantages of Careful Seeding[C]∥Proceedings of the eighteenth annual ACM-SIAM symposium on Discrete algorithms, Society for Industrial and Applied Mathematics, 2007.
[1] 王宇,杨艺,王宝山,王田,卜旭辉,王传云. 深度残差神经网络高分辨率遥感图像建筑物分割[J]. 遥感技术与应用, 2019, 34(4): 736-747.
[2] 赖日文,池毓锋,张泽均. 基于亮度恢复模型的Landsat 8数据山区阴影去除[J]. 遥感技术与应用, 2018, 33(3): 563-572.
[3] 秦振涛,杨茹,张靖,杨武年. 基于聚类结构自适应稀疏表示的高光谱遥感图像修复研究[J]. 遥感技术与应用, 2018, 33(2): 212-215.
[4] 万玲,尤红建,程跃兵,卢晓军. 合成孔径雷达图像分割研究进展[J]. 遥感技术与应用, 2018, 33(1): 10-24.
[5] 张霞,戚文超,孙伟超. 基于数学形态滤波的植被光谱去噪方法研究[J]. 遥感技术与应用, 2016, 31(5): 846-854.
[6] 王俊,秦其明,叶昕,王建华,秦雪彬,杨绣丞. 高分辨率光学遥感图像建筑物提取研究进展[J]. 遥感技术与应用, 2016, 31(4): 653-662.
[7] 张从梅,孙权森,王超,封磊,顾一禾. 基于非局部自相似性的遥感图像稀疏去噪算法[J]. 遥感技术与应用, 2016, 31(4): 739-747.
[8] 赵永光,李传荣,马灵玲,唐伶俐,王宁. 一种遥感图像太阳—观测几何归一化方法[J]. 遥感技术与应用, 2016, 31(2): 260-266.
[9] 汪骏,夏少波,王和平,王成,谭弘武. 基于直升机激光点云的分裂导线重建研究[J]. 遥感技术与应用, 2015, 30(6): 1189-1194.
[10] 肖新耀,许宁,尤红建. 一种基于à trous小波和联合稀疏表示的遥感图像融合方法[J]. 遥感技术与应用, 2015, 30(5): 1021-1026.
[11] 苏腾飞,李洪玉. 一种两阶段区域生长的遥感图像分割算法[J]. 遥感技术与应用, 2015, 30(3): 476-485.
[12] 侯鹏洋,季艳,高峰,胡蕾. 一种基于SIFT特征的快速逐层遥感图像配准方法[J]. 遥感技术与应用, 2014, 29(5): 873-877.
[13] 万智萍. 结合方向小波的多光谱与全色遥感图像融合算法[J]. 遥感技术与应用, 2014, 29(4): 660-668.
[14] 李青松,覃锡忠,贾振红,杨杰,胡英杰. 基于非下采样Shearlet和几何结构的遥感图像无监督变化检测[J]. 遥感技术与应用, 2014, 29(3): 482-488.
[15] 姜澒月,周坚华. 遥感图像薄云雾的梯度改正[J]. 遥感技术与应用, 2013, 28(4): 640-646.