太空資料系統諮詢委員會(CCSDS)提出了應用於衛星遙測影像的壓縮標準。然而，當編碼過後的衛星影像從太空站被傳輸至地面接收站時，檔案可能會因為長距離的無線傳輸而遭到毀損。即使毀損產生的錯誤僅有單一位元，在解碼後仍然可能造成大範圍的影像破損。對於這樣的狀況，本論文提出了一個修復的演算法。福衛二號衛星擁有拍攝多頻譜影像的感測器，不同頻譜間的資訊是做為修復破損影像的重要資訊。首先利用線性回歸模型從其他未毀損的多光譜和全色圖像生成出參考圖像。接著再以深度學習的方法學習參考圖像和原始圖像之間端到端的映射關係。此映射關係是以卷積神經網絡（CNN）來表示，其將參考圖像作為輸入並輸出修復的圖像。在實驗結果中，所修復的衛星影像顯現了良好的視覺品質。

The Consultative Committee for Space Data Systems (CCSDS) provides an image compression standard for remote sensing image. When the coded CCSDS image has been transmitted from space to ground, the data can be damaged because of the long distance the signal has travelled. Even a single bit error can cause significant damage after decoding. A scheme is proposed to deal with such circumstance. The FORMOSAT-2 satellite is capable of capturing multi-spectral images which are essential information for restoring the corrupted images. First, the linear regression model is used to generate the reference image from other undamaged multi-spectral and panchromatic images. Then, a deep learning method directly learns an end-to-end mapping between the reference image and the ground truth image. The mapping is represented as a convolutional neural network (CNN) that takes the reference image as the input and outputs the restored one. The experimental results show that the proposed method can obtain better visual quality of the restored images.

[1] Mohamed Tolba, Hala Moushir Ebied, Mohamed H. Abdel-Aziz, “High Performance Hyperspectral Image Classification using Graphics Processing Units,” Thesis for: Master of Science, DOI: 10.13140/2.1.3025.4887, Jan. 2015.
[2] T. T. Cheung, C. R.S., and L. K.B., “DC error correction for image transmission over wireless channels,” Wireless Communications and Networking Conference, 2000. WCNC. 2000 IEEE, vol. 3, pp. 1237 – 1242, 23 Sep 2000-28 Sep 2000.
[3] Draft CCSDS recommended standard for image data compression, Consultative Committee for Space Data Systems Std.H. Gonçalves, L. Corte-Real, J. A. Gonçalves, “Automatic image registration through image segmentation and SIFT”, IEEE Trans. Geosci. Remote Sens., vol. 49, no. 7, pp. 2589-2600, Jul. 2011.
[4] K. Fukushima, “Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position,” Biol. Cybern., vol. 36, no. 4, pp. 193–202, Jan. 1980.
[5] Y. Lecun, L. Bottou, Y. Bengio, and P. Haffner, “Gradient-based leaning applied to document recognition,” Proceedings of the IEEE, vol. 86, no. 11, pp. 2278– 2324, Nov. 1998.
[6] A. Krizhevsky, I. Sutskever, and G. E. Hinton, “ImageNet classification with deep convolutional neural networks,” Commun. ACM, vol. 60, no. 6, pp. 84–90, Jun. 2017.
[7] W. Ouyang et al., “Deepid-net: Deformable deep convolutional neural networks for object detection,” in Proc. IEEE Conf. Comput. Vis. Pattern Recog., Boston, MA, USA, 2015, pp. 2403–2412.
[8] Y. Sun,Y. Chen, X.Wang, and X. Tang, “Deep learning face representation by joint identification-verification,” in Proc. 27th Int. Conf. Neural Inf. Process. Syst., Montreal, QC, Canada, 2014, pp. 1988–1996.
[9] C. Dong, C. C. Loy, K. He, and X. Tang, “Image super-resolution using deep convolutional networks,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 38, no. 2, pp. 295–307, Feb. 2016.
[10] Y. Chen, H. Jiang, C. Li, X. Jia, and P. Ghamisi, “Deep feature extraction and classification of hyperspectral images based on convolutional neural networks,” IEEE Trans. Geosci. Remote Sens., vol. 54, no. 10, pp. 6232– 6251, Oct. 2016.
[11] W. Li, G. Wu, F. Zhang, and Q. Du, “Hyperspectral image classification using deep pixel-pair features,” IEEE Trans. Geosci. Remote Sens., vol. 55, no. 2, pp. 844–853, Feb. 2017.
[12] [Online].https://commons.wikimedia.org/wiki/File:Rosenblattperceptron.png
[13] V. Nair and G. E. Hinton, “Rectified linear units improve restricted Boltzmann machines,” in Proc. 27th Int. Conf. Mach. Learn., Haifam, Israel, 2010, pp. 807–814.
[14] H.C. Burger, C.J. Schuler, and S. Harmeling, “Image denoising: Can plain neural networks compete with BM3D?,” IEEE Conference on Computer Vision and Pattern Recognition., pp. 2392–2399, 2012.
[15] C.J. Schuler, H.C. Burger, and S. Harmeling, “Scholkopf, B.: A machine learning approach for non-blind image deconvolution,” IEEE Conference on Computer Vision and Pattern Recognition, pp. 1067–1074, 2013.
[16] V. Jain, and S. Seung, “Natural image denoising with convolutional networks,” Advances in Neural Information Processing Systems, pp. 769–776, 2008.
[17] D. Eigen, D. Krishnan, and R. Fergus, “Restoring an image taken through a window covered with dirt or rain,” IEEE International Conference on Computer Vision, pp. 633–640, 2013.
[18] Z. Cui, H. Chang, S. Shan, B. Zhong, and X. Chen, “Deep network cascade for image super-resolution,” European Conference on Computer Vision, pp. 49–64, 2014.
[19] R. Ren, S. Guo, L. Gu, and H. Wang, “Automatic thick cloud removal for MODIS remote sensing imagery,” in Information Engineering and Computer Sci- ence, 2009. ICIECS 2009. International Conference on, 19-20 , pp. 1 – 4, Dec. 2009
[20] X. Zhu, F. Gao, D. Liu, and J. Chen, “A modified neighborhood similar pixel interpolator approach for removing thick clouds in landsat images,” in Geoscience and Remote Sensing Letters, IEEE, pp. 521 – 525, May 2012
[21] G. S. James Storey, and Pasquale Scaramuzza, “Landsat 7 scan line corrector-off gap- filled product,” Global Priorities in Land Remote Sensing, October 2005.
[22] F. Gao, J. Masek, M. Schwaller, and F. Hall, “On the blending of the landsat and modis surface reflectance: predicting daily landsat surface reflectance,” Geoscience and Remote Sensing, IEEE Transactions on, vol. 44, no. 8, pp. 2207–2218, Aug 2006.
[23] J. Chen, X. Zhu, J. E. Vogelmann, F. Gao, and S. Jin, “A simple and effective method for filling gaps in landsat etm+ slc-off images,” Remote Sensing of Environment, vol. 115, no. 4, pp. 1053 – 1064, 2011.
[24] W. Fan, J. Chen, and W. Ju, “A pixel missing patch inpainting method for remote sensing image,” in Geoinformatics, 2011 19th International Conference on, June 2011, pp. 1–4.
[25] W. Hu, M. Li, Y. Liu, Q. Huang, and K. Mao, “A new method of restoring etm+ slc-off images based on multi-temporal images,” in Geoinformatics, 2011 19th International Conference on, June 2011, pp. 1–4.
[26] Y. LeCun, L. Bottou, Y. Bengio, P. Haffner, “Gradient-based learning applied to document recognition, ” Proceedings of the IEEE 86(11), 2278–2324 (1998)
[27] W. Fan,W. Ju, and Z. Gu, “Method for reconstructing the pixel missing region on remote sensing images,” J. Appl. Remote Sens. 7, 073536 (2013).
[28] C. Zeng, H. Shen, and L. Zhang, “Recovering missing pixels for Landsat ETM + SLC-off imagery using multi-temporal regression analysis and a regularization method,” Remote Sens. Environ. 131(15), 182–194 (2013)
[29] S. C. Tai, Chiean-Yen Chao, “A Restoration Algorithm for Bit-Error-Caused Damages in CCSDS Satellite Images,” Master’s thesis, National Cheng Kung University, Tainan, Taiwan, R.O.C.,2014
[30] L. Wald, T. Ranchin, and M. Mangolini, “Fusion of satellite images of different spatial resolutions: assessing the quality of resulting images,” Photogramm. Eng. Remote Sens. 63(6), 691–699 (1997).
[31] M. Choi, “A new intensity–hue–saturation fusion approach to image fusion with a tradeoff parameter,” Geoscience and Remote Sensing, IEEE Transactions on, vol. 44, no. 6, pp. 1672 – 1682, jun 2006.
[32] J. N. Sarvaiya, S. Patnaik, and S. Bombaywala, “Image registration by template matching using normalized cross-correlation,” in Int. Conf. on Advances in Computing, Control, & Telecommunication Technologies (ACT ‘09), pp. 819–822 (2009).
[33] He, K., Sun, J.: Convolutional neural networks at constrained time cost. arXiv preprint arXiv:1412.1710 (2014)