Please wait a minute...
img

官方微信

遥感技术与应用  2021, Vol. 36 Issue (5): 1092-1099    DOI: 10.11873/j.issn.1004-0323.2021.5.1092
数据与图像处理     
基于类内拟合的遥感影像薄云雾校正方法
张弛1,4(),姜红涛2,谢成3,赖少川3,沈焕锋4,5()
1.广州市城市规划勘测设计研究院,广东 广州 510060
2.仲恺农业工程学院,广东 广州 510130
3.国家石油天然气管网集团公司华南分公司,广东 广州 510130
4.武汉大学资源与环境科学学院,湖北 武汉 430079
5.地球空间信息技术协同创新中心,湖北 武汉 430079
An Intra-class Linear Regression based Thin Cloud Removal Method for Visible Remote Sensing Images
Chi Zhang1,4(),Hongtao Jiang2,Cheng Xie3,Shaochuan Lai3,Huanfeng Shen4,5()
1.Guangzhou Urban Planning & Design Survey Research Institute,Guangzhou 510060,China
2.Zhongkai University of Agriculture and Engineering,Guangzhou 510130,China
3.South China Branch of Sinopec Sales Co. ,Ltd. ,Guangzhou 510130,China
4.School of Resource and Environmental Sciences,Wuhan University,Wuhan 430079,China
5.Collaborative Innovation Center of Geospatial Technology,Wuhan 430079,China
 全文: PDF(4336 KB)   HTML
摘要:

光学遥感观测极易受到云雾影响,降低数据质量并限制其后续应用潜力。由此,提出了一种基于类内拟合的遥感影像薄云雾校正方法。首先,利用滑动窗口逐波段地搜索局部最小值,称之为暗目标,通过拟合不同波段的暗目标样本估计出薄云雾辐射的相关性。基于此,联合云雾波段相关性与成像模型,生成不含云雾干扰的合成假彩色影像,利用K均值分类自动得到地表覆被类型。利用地类信息,进一步选取晴空区像元获取不同地类在不同波段对间的线性关系。最后,将上述两种线性关系进行联立,求解出各地表类型在不同波段上的值,从而完成影像校正。通过模拟与真实实验对方法有效性和场景适用性进行测试,并从定性目视与定量评估两方面对结果进行检验。实验结果表明:提出方法可有效去除薄云雾干扰,适用于不同地表覆被类型场景,获得高光谱保真的校正地表。

关键词: 薄云雾校正类内拟合可见光影像暗目标法    
Abstract:

Earth observations of satellite or airborne sensors are easily interfered by the atmospheric conditions, thereby resulting in the frequent cloud contamination in the acquired images, reducing the availability and validity of data. In this paper, a thin cloud removal method based on intra-class linear regression is proposed for visible remote sensing images, which mainly consists three steps. Firstly, the local dark object (minimum in local) is searched band by band with a certain window size. The dark objects samples are then used to regress the linear correlation of clouds among bands. Secondly, the cloud correlations among bands are combined with the cloudy image model to generate the synthetic image without cloud contamination, and the K-means classification is performed on it to obtain the land cover types. Based on that, the linear relationship of different land covers can be estimated using the corresponding clear samples. Thirdly, by integrating the linear correlations of clouds and various land covers, the clear surface information can finally be solved from the cloudy image model. Both the simulated and real data were collected to validate the effectiveness of the method from visual and quantitative aspects. Experimental results demonstrate that the thin clouds in various scenes can be totally removed by the method and the degraded information can be recovered satisfactory.

Key words: Thin cloud removal    Intra-class linear regression    Visible remote sensing images    Dark object searching
收稿日期: 2020-06-22 出版日期: 2021-12-07
ZTFLH:  P208  
基金资助: 成品油管道巡线无人机影像的质量改善技术研究与应用(30251731?19?ZC0607?0024)
通讯作者: 沈焕锋     E-mail: zhangchi9502@outlook.com;shenhf@whu.edu.cn
作者简介: 张弛(1995-),男,江西抚州人,博士研究生,主要从事遥感影像质量改善工作。E?mail: zhangchi9502@outlook.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  
张弛
姜红涛
谢成
赖少川
沈焕锋

引用本文:

张弛,姜红涛,谢成,赖少川,沈焕锋. 基于类内拟合的遥感影像薄云雾校正方法[J]. 遥感技术与应用, 2021, 36(5): 1092-1099.

Chi Zhang,Hongtao Jiang,Cheng Xie,Shaochuan Lai,Huanfeng Shen. An Intra-class Linear Regression based Thin Cloud Removal Method for Visible Remote Sensing Images. Remote Sensing Technology and Application, 2021, 36(5): 1092-1099.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2021.5.1092        http://www.rsta.ac.cn/CN/Y2021/V36/I5/1092

图1  薄云雾可见光影像
图2  波段对暗目标散点回归图
图3  K均值分类结果对比
图4  不同地表覆被类别在不同波段对间地表辐射的线性散点示意图
图5  模拟实验校正结果比较
TDCPDSP本文方法
RMSE14.512 37.354 22.117 4
SA1.711 51.214 10.645 5
R20.824 10.891 60.942 8
表 1  定量指标评价结果
图6  第一组真实实验校正结果比较
图7  第二组真实实验校正结果比较
1 Zhao Zhongmin, Zhu Chongguang. Approach to removing cloud cover from satellite imagery[J].Journal of Remote Sensing,1996,1(3):195-199.
1 赵忠明, 朱重光. 遥感图像中薄云的去除方法[J]. 遥感学报,1996,1(3):195-199.
2 Cao Shuang, Li Hao, Ma Wen. Removing thin cloud arithmetic based on mathematic morphology for remote sensing image [J]. Geography and Geo-Information Science, 2015,31(4): 34-37.
2 曹爽, 李浩, 马文. 基于数学形态学的遥感影像薄云处理方法[J]. 地理与地理信息科学,2015,31(4):34-37.
3 Shen H, Li H, Qian Y, et al. An effective thin cloud removal procedure for visible remote sensing images[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2014, 96: 224-235.
4 Chen S, Chen X, Chen J, et al. An iterative haze optimized transformation for automatic Cloud/Haze detection of Landsat imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2015, 54(5): 2682-2694.
5 Du Y, Guindon B, Cihlar J. Haze detection and removal in high resolution satellite image with wavelet analysis[J]. IEEE Transactions on Geoscience and Remote Sensing, 2002, 40(1): 210-217.
6 Li H, Zhang L, Shen H, et al. A variational gradient-based fusion method for visible and SWIR imagery[J]. Photogrammetric Engineering & Remote Sensing, 2012, 78(9): 947-958.
7 Jiang Hou, Lü Ning, Yao Ling. HOT-transform based method to remove haze or thin cloud for Landsat 8 OLI satellite data[J]. Journal of Remote Sensing, 2016, 20(4): 620-631.
7 姜侯, 吕宁, 姚凌. 改进 HOT 法的 Landsat 8 OLI 遥感影像雾霾及薄云去除[J]. 遥感学报, 2016, 20(4): 620-631.
8 Jiang Haoyue, Zhou Jianhua. Removing thin cloud cover from remote sensing images[J]. Remote Sensing Technology and Application, 2013, 28(4): 640-646.
8 姜灏月, 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.
9 Li Gang, Yang Wunian, Weng Tao. Homomorphic filtering based thin cloud removal method for remote sensing images [J]. Science of Surveying and Mapping, 2007, 32(3): 47-48.
9 李刚, 杨武年, 翁韬. 一种基于同态滤波的遥感图像薄云去除算法[J]. 测绘科学, 2007, 32(3): 47-48.
10 Liu J, Wang X, Chen M, et al. Thin cloud removal from single satellite images[J]. Optics Express, 2014, 22(1): 618-632.
11 Chavez Jr P S. An improved Dark-Object subtraction technique for atmospheric scattering correction of multispectral data[J]. Remote Sensing of Environment, 1988, 24(3): 459-479.
12 Makarau A, Richter R, Müller R, et al. Haze detection and removal in remotely sensed multispectral imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2014, 52(9): 5895-5905.
13 He K, Sun J, Tang X. Single image haze removal using dark channel prior[J]. IEEE Transactions On Pattern Analysis and Machine Intelligence, 2010, 33(12): 2341-2353.
14 Li J, Wu Z, Hu Z, Zhang J, Molinier M. Thin cloud removal in optical remote sensing images based on generative adversarial networks and physical model of cloud distortion[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2020, 166: 373-389.
15 Wla B, Ying L A, Di C A. Thin cloud removal with residual symmetrical concatenation network[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2019, 153: 137-150.
16 Zi Y, Xie F, Zhang N, et al. Thin cloud removal for multispectral remote sensing images using convolutional neural networks combined with an imaging model[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021,99: 1-1.
17 Chen H, Chen R, Li N. Attentive generative adversarial network for removing thin cloud from a single remote sensing image[J]. IET Image Processing, 2021, 15(4): 856-867.
18 Lü H, Wang Y, Shen Y. An empirical and radiative transfer model based algorithm to remove thin clouds in visible bands[J]. Remote Sensing of Environment, 2016, 179: 183-195.
19 Kaufman Y J, Wald A E, Remer L A, et al. The MODIS 2.1-/Spl Mu/M channel-correlation with visible reflectance for use in remote sensing of aerosol[J]. IEEE Transactions On Geoscience and Remote Sensing,1997,35(5):1286-1298.
20 Xie F, Chen J, Pan X, et al. Adaptive haze removal for single remote sensing image[J]. IEEE Access, 2018,6: 67982-67991.
21 Muhammad K, Ahmad J, Rho S, et al. Image steganography for authenticity of visual contents in social networks[J]. Multimedia Tools and Applications, 2017, 76(18): 18985-19004.
[1] 赵志强,李爱农,边金虎,郭文静,刘倩楠,赵伟,黄成全. 山区可见光—近红外遥感影像浓密植被暗像元自动识别方法研究[J]. 遥感技术与应用, 2015, 30(1): 58-67.