摘要
为提高影像云识别精度,提出一种多尺度膨胀卷积深层神经网络云识别方法。结合卫星影像特征,设计云识别卷积神经网络结构,该结构包含深层特征编码模块、局部多尺度膨胀感知模块以及云区预测解码模块。首先,编码模块中通过基础卷积层获取深度特征;其次,联合多尺度膨胀卷积和池化层共同感知,每层操作连接非线性函数,以提升网络模型的表达能力;最后,云区预测解码模块中融合对应编码模块的特征,再利用L1正则化上采样算法实现端对端的像素级云识别结果。选用典型云遮挡区域影像进行云识别实验,并与Otsu算法和FCN-8S算法进行对比。结果表明,本文所提算法的检测精度较高,Kappa系数显著提升。
To improve the accuracy of cloud detection, we propose a multi-scale dilation convolutional neural network method. Combining with the characteristic of satellite images, we design the deep convolution network structure, which includes a deep-feature coding module, a local dilation perception module, and a cloud-dense decoding module. First, the deep-features of cloud are obtained by the basic convolutional layer in conjunction with the coding module. Second, multi-scale dilation convolution layers jointed with pooling layers are used to perceive corporately. A nonlinear function is employed in each block to improve the effectiveness of network model expression. Finally, the cloud-dense decoding module integrate the features corresponding to those included in the coding module and then utilize the L1 regularization upsampling algorithm to accomplish the end-to-end pixel-level cloud detection task. Cloud detection experiments are performed in the typical cloud mask areas; the results are compared with those of the Otsu algorithm and the FCN-8 S method. The results indicate that the accuracy of proposed method is higher and the Kappa coefficient is effectively improved.
引文
[1] Pan H B, Zhang G, Tang X M, et al. The geometrical model of sensor corrected products for ZY-3 satellite[J].Acta Geodaetica et Cartographica Sinica, 2013, 42(4):516-522.潘红播,张过,唐新明,等.资源三号测绘卫星传感器校正产品几何模型[J].测绘学报,2013, 42(4):516-522.
[2] Chen Y, Fan R S, Wang J X, et al. Cloud detection of ZY-3 satellite remote sensing images based on deep learning[J]. Acta Optica Sinica, 2018, 38(1):0128005.陈洋,范荣双,王竞雪,等.基于深度学习的资源三号卫星遥感影像云检测方法[J].光学学报,2018, 38(1):0128005.
[3] Chen P Y, Srinivasan R, Fedosejevs G, et al. An automated cloud detection method for daily NOAA-14AVHRR data for Texas, USA[J]. International Journal of Remote Sensing, 2002, 23(15):2939-2950.
[4] Otsu N. A threshold selection method from gray-level histograms[J]. IEEE Transactions on Systems, Man,and Cybernetics, 1979, 9(1):62-66.
[5] Zhu Z, Woodcock C E. Object-based cloud and cloud shadow detection in Landsat imagery[J]. Remote Sensing of Environment, 2012, 118:83-94.
[6] Panem C, Baillarin S, Latry C, et al. Automatic cloud detection on high resolution images[C]//Proceedings of IEEE International Geoscience and Remote Sensing Symposium. Seoul:IEEE, 2005:1-4.
[7] Qin Y, Fu Z L, Zhou F, et al. A method of automatic cloud detection using TM images[J]. Geomatics and Information Science of Wuhan University, 2014, 39(2):234-238秦永,付仲良,周凡,等.一种利用TM影像自动检测云层的方法[J].武汉大学学报(信息科学版),2014,39(2):234-238.
[8] Liu J. Improvement of dynamic threshold value extraction technic in FY-2 cloud detection[J]. Journal of Infrared and Millimeter Waves, 2010, 29(4):288-292.刘健.FY-2云检测中动态阈值提取技术改进方法研究[J].红外与毫米波学报,2010, 29(4):288-292.
[9] Shan N, Zheng T Y, Wang Z S. High-speed and highaccuracy algorithm for cloud detection and its application[J]. Journal of Remote Sensing, 2009, 13(6):1138-1155.单娜,郑天垚,王贞松.快速高准确度云检测算法及其应用[J].遥感学报,2009,13(6):1138-1155.
[10] Harb M, Gamba P, Dellacqua F. Automatic delineation of clouds and their shadows in Landsat and CBERS(HRCC)data[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2016,9(4):1532-1542.
[11] Braaten J D, Cohen W B, Yang Z. Automated cloud and cloud shadow identification in Landsat MSS imagery for temperate ecosystems[J]. Remote Sensing of Environment, 2015, 169:128-138.
[12] Cheng Q, Shen H F, Zhang L P, et al. Cloud removal for remotely sensed images by similar pixel replacement guided with a spatio-temporal MRF model[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2014,92:54-68.
[13] Ozkan S, Efendioglu M, Demirpolat C. Cloud detection from RGB color remote sensing images with deep pyramid networks[J/OL]. Computer Science, 2018.[2018-08-08]. https://arxiv.org/pdf/1801.08706.pdf.
[14] Scaramuzza P L, Bouchard M A, Dwyer J L.Development of the Landsat data continuity mission cloud-cover assessment algorithms[J].IEEE Transactions on Geoscience&Remote Sensing, 2012,50(4):1140-1154.
[15] Murtagh F, Barreto D, Marcello J. Decision boundaries using bayes factors:the case of cloud masks[J]. IEEE Transactions on Geoscience and Remote Sensing, 2003,41(12):2952-2958.
[16] Chen L C, Papandreou G, Kokkinos I, et al. DeepLab:semantic image segmentation with deep convolutionalnets, atrous convolution, and fully connected CRFs[J].IEEE Transactions on Pattern Analysis&Machine Intelligence, 2016, 40(4):834-848.
[17] Lu Y S, Li Y X, Liu B, et al. Hyperspectral data haze monitoring based on deep residual network[J]. Acta Optica Sinica, 2017, 37(11):1128001.陆永帅,李元祥,刘波,等.基于深度残差网络的高光谱遥感数据霾监测[J].光学学报,2017, 37(11):1128001.
[18] Badrinarayanan V, Kendall A, Cipolla R. SegNet:a deep convolutional encoder-decoder architecture for image segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(12):2481-2495.
[19] Ronneberger O, Fischer P, Brox T. U-Net:convolutional networks for biomedical image segmentation[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention.Cham:Springer, 2015:234-241.
[20] Krizhevsky A, Sutskever I, Hinton G E. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017, 60(6):84-90.
[21] LeCun Y, Bengio Y, Hinton G. Deep learning[J].Nature, 2015, 521(7553):436-444.
[22] Hamaguchi R, Fujita A, Nemoto K, et al. Effective useof dilated convolutions for segmenting small object instances in remote sensing imagery[C]. IEEE Winter Conference on Applications of Computer Vision, 2018:1442-1450.
[23] Hinton G E. Learning multiple layers of representation[J]. Trends in Cognitive Sciences, 2007, 11(10):428-434.
[24] Kingma D, Ba J. Adam:a method for stochastic optimization[J/OL]. Computer Science, 2014.[2018-08-08]. https://arxiv. org/pdf/1412.6980. pdf.
[25] Luo W, Li Y, Urtasun R, et al. Understanding the effective receptive field in deep convolutional neural networks[J/OL]. Computer Science, 2017.[2018-08-08]. https://arxiv.org/pdf/1701.04128.pdf.
[26] Shelhamer E, Long J, Darrell T. Fully convolutional networks for semantic segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(4):640-651.
[27] Gao X J, Wang Y C, Zheng S Y, et al. Real-time automatic cloud detection during the process of taking aerial photographs[J]. Spectroscopy and Spectral Analysis, 2014, 34(7):1909-1913.高贤君,万幼川,郑顺义,等.航空摄影过程中云的实时自动检测[J].光谱学与光谱分析,2014, 34(7):1909-1913.