Ground Object Information Recovery for Thin Cloud Contaminated Remote Sensing Images by Combining Classification with Transfer Learning
HU Gen-sheng1,2,3, ZHA Hui-min1,2, LIANG Dong1,2, BAO Wen-xia1,2,3
1. Key Laboratory of Intelligent Computing and Signal Processing, Ministry of Education, Anhui University, Hefei, Anhui 230039, China;
2. School of Electronics and Information Engineering, Anhui University, Hefei, Anhui 230601, China;
3. Anhui Key Laboratory of Polarization Imaging Detection Technology, Hefei, Anhui 230031, China
Abstract：By using multi-source and multi-temporal remote sensing images,a ground object information recovery algorithm for thin cloud contaminated remote sensing images is proposed by combining classification with transfer learning.Firstly,multi-resolution decomposition of multi-source and multi-temporal remote sensing images is performed by using multi-directional nonsubsampled dual-tree complex wavelet transform.The decomposed high frequency coefficients of the ground objects of the thin cloud images are primarily classified by using Bayesian method.Then the transfer least square support vector regression model is trained to obtain the model parameters by using the domain adaptive learning of the low frequency coefficients of each class of ground objects.Finally,the low frequency coefficients of the thin cloud-contaminated images are predicted by using those of the cloudless reference images.The thin clouds are removed and the ground object information of the thin cloud contaminated images is recovered.Experimental results show that the ground objects recovered by the proposed algorithm have clear spatial details and small spectral distortion.Especially for the thin cloud contaminated remote sensing images with seasonal variation of ground objects,the proposed algorithm can effectively recover the ground object information contaminated by thin clouds.
[1] Rasti B,Sveinsson J R,Ulfarsson M O.Wavelet-based sparse reduced-rank regression for hyperspectral image restoration[J].IEEE Transactions on Geoscience and Remote Sensing,2014,52(10):6688-6698.
[2] Wang X,Li M M,Tang H M.A modified homomorphism filtering algorithm for cloud removal[A].2010 International Conference on Computational Intelligence and Software Engineering[C].Wuhan,China:IEEE,2010.1-4.
[3] Wu X P,Yang W N,Li G M.Thin cloud removal of ZY-3 image based on improved homomorphism filtering method[A].21st International Conference on Geoinformatics[C].Kaifeng,China:IEEE,2013.1-4.
[4] Leung Y,Liu J,Zhang J S.An improved adaptive intensity-hue-saturation method for the fusion of remote sensing images[J].IEEE Geoscience and Remote Sensing Letters,2014,11(5):985-989.
[5] Gao C,Zhu H J,Guo Y C.Analysis and improvement of SUSAN algorithm[J].Signal Processing,2012,92(10):2552-2559.
[6] 卢力,王勇涛,田金文,柳健.基于SUSAN算法的遥感图像去云[J].通信学报,2006,27(8):160-164. Lu Li,Wang Yongtao,Tian Jinwen,Liu Jian.Clould removal from satellite imagery based on SUSAN algorithm[J].Journal of Communication,2006,27(8):160-164.(in Chinese)
[7] Zhu X F,Wu F,Zhuang Y P.An improved approach to remove cloud and mist from remote sensing digital images based on Mallat algorithm[J].Journal of Remote Sensing,2007,11(2):241-246.
[8] Zhu X F,Jiang X L,Li.F Removing thin cloud in color remote sensing images[J].Journal of Applied Optics,2007,28(6):698-701.
[9] 李新宇,刘扬阳,蒋雪娜,王守觉.基于高维空间几何信息学的遥感图像去薄云算法[J].电子学报,2011,39(5):1002-1006. Li Xinyu,Liu Yangyang,Jiang Xuena,Wang Shoujue.An noval algorithm to remove the thin cloud from remote sensing images based on high-dimensional space geometrical informatics[J].Acta Electronica Sinica,2011,39(5):1002-1006.(in Chinese)
[10] Luo B,Khan M M,Bienvenu T.Decision-based fusion for pansharpening of remote sensing images[J].IEEE Geoscience and Remote Sensing Letters,2013,10(1):19-23.
[11] Chien C L,Tsai W H.Image fusion with no gamut problem by improved nonlinear IHS transforms for remote sensing[J].IEEE Transactions on Geoscience and Remote Sensing,2014,52(1):651-663.
[12] Shi C,Liu F, Li L L,Hao H X.Pan-sharpening algorithm to remove thin cloud via mask dodging and nonsampled shift-invariant shearlet transform[J].Journal of Applied Remote Sensing,2014,8(1):083658.
[13] Makarau A,Richter R,Schl? pfer D,Reinartz P.Combined haze and cirrus removal for multispectral imagery[J].IEEE Geoscience and Remote Sensing Letters,2016,13(3):379-383.
[14] Lorenzi L,Melgani F,Mercier G.Missing-area reconstruction in multispectral images under a compressive sensing perspective[J].IEEE Transactions on Geoscience and Remote Sensing,2013,51(7):3998-4008.
[15] Hu G S,Li X Y,Liang D.Thin cloud removal from remote sensing images using multidirectional dual tree complex wavelet transform and transfer least square support vector regression[J].Journal of Applied Remote Sensing,2015,9(1):095053,1-19.
[16] Pan S J,Yang Q.A survey on transfer learning[J].IEEE Transactions on Knowledge and Data Engineering,2010,22(10):1345-1359.
[17] Pan S J, Tsang I W,Kwok J T,Yang Q.Domain adaption via transfer component analysis[J].IEEE Transactions on Neural Networks,2011,22(2):199-210.
[18] L X Duan,I W Tsang,D Xu.Domain transfer multiple kernel learning[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2012,34(3):465-479.
[19] Duan L X,Xu D,Tsang I W.Domain adaptation from multiple sources:a domain-dependent regularization approach[J].IEEE Transactions on Neural Networks and Learning Systems,2012,23(3):504-518.
[20] 韩敏,杨雪.改进贝叶斯ARTMAP的迁移学习遥感影像分类算法[J].电子学报,2016,44(9):2248-2253. Han Min,Yang Xue.Transfer learning using improved bayesian ARTMAP for remote sensing image classification[J].Acta Electronica Sinica,2016,44(9):2248-2253.(in Chinese)
[21] Yang J,Yan R,Hauptmann A G.Cross-domain video concept detection using adaptive svms[A].Proceedings of the 15th ACM International Conference on Multimedia[C].Augsburg,Germany:ACM,2007,188-197.
[22] Scholkopt B,Smola A J.Learning with Kernels[M].Cambridge,MA:The M IT Press,2002.
2017, Vol. 45 
