传统的合成孔径雷达(SAR)图像变化检测方法存在受散斑噪声影响大、图像深层信息难以利用、检测精度低等问题。针对以上问题,提出一种基于卷积神经网络和模糊聚类的SAR幅度图像变化检测算法。首先通过基于Gabor纹理的FLICM算法对差异图进行预分类,基于预分类结果自动选取可靠的训练样本,无需人工进行样本标注;引入多尺度通道注意力机制,并采用MSCA_WCNN完成二次分类,最终得到变化检测结果。本文算法在提取SAR图像不同尺度特征的同时能够对无关特征通道进行抑制,从而有效利用图像特征;小波卷积神经网络在保留图像有用信息的同时实现了去噪功能,增强了算法的鲁棒性。采用真实的星载SAR图像数据进行对比实验,实验结果表明,本文算法具有较高的检测精度,验证了算法的有效性。
Traditional synthetic aperture radar (SAR) image change detection methods have some problems, such as big impact by speckle noise, difficult to use deep information of the image, and low detection accuracy. To solve above problems, this paper presents an SAR amplitude image change detection algorithm based on convolution neural network and fuzzy clustering. Firstly, a hierarchical FLICM algorithm based on Gabor texture is used to pre-classify the difference images, and reliable training samples are automatically selected based on the pre-classification results without manual labeling. Then, a multiscale channel attention mechanism is introduced, and a MSCA_WCNN is used to complete the second classification, and the result of change detection is obtained. This algorithm extracts the different scale features of SAR images while suppressing the irrelevant feature channels to effectively utilize the image features. The wavelet convolution neural network achieves the denoising function while preserving the useful information of the image and enhances the robustness of the algorithm. The comparison experiments using real spaceborne SAR image data show that the algorithm has high detection accuracy and the effectiveness of the algorithm is verified.
[1] 吴一戎, 朱敏慧. 合成孔径雷达技术的发展现状与趋势[J]. 遥感技术与应用, 2000, 15(2):121-123. DOI:10.3969/j.issn.1004-0323.2000.02.012.
[2] 李麒崙, 张万昌, 易亚宁. 地震滑坡信息提取方法研究:以2017年九寨沟地震为例[J]. 中国科学院大学学报,2020,37(1):93-102.DOI:10.7523/j.issn.2095-6134.2020.01.011.
[3] Tarantino C, Adamo M, Lucas R, et al. Detection of changes in semi-natural grasslands by cross correlation analysis with WorldView-2 images and new Landsat 8 data[J]. Remote Sensing of Environment, 2016, 175:65-72. DOI:10.1016/j.rse.2015.12.031.
[4] Ban Y F, Yousif O A. Multitemporal spaceborne SAR data for urban change detection in China[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2012, 5(4):1087-1094. DOI:10.1109/JSTARS.2012.2201135.
[5] Jin S M, Yang L M, Danielson P, et al. A comprehensive change detection method for updating the National Land Cover Database to circa 2011[J]. Remote Sensing of Environment, 2013, 132:159-175. DOI:10.1016/j.rse.2013.01.012.
[6] 王峰萍, 王卫星, 高婷, 等. 基于离散小波变换和邻域模糊C均值的变化检测方法[J]. 西北工业大学学报, 2018,36(3):426-431.DOI:10.3969/j.issn.1000-2758.2018.03.004.
[7] Su L Z, Gong M G, Sun B, et al. Unsupervised change detection in SAR images based on locally fitting model and semi-EM algorithm[J]. International Journal of Remote Sensing, 2014, 35(2):621-650. DOI:10.1080/01431161.2013.871596.
[8] Zhang X H, Chen J W, Meng H Y. A novel SAR image change detection based on graph-cut and generalized Gaussian model[J]. IEEE Geoscience and Remote Sensing Letters, 2013, 10(1):14-18. DOI:10.1109/LGRS.2012.2189867.
[9] Bazi Y, Melgani F, Al-Sharari H D. Unsupervised change detection in multispectral remotely sensed imagery with level set methods[J]. IEEE Transactions on Geoscience and Remote Sensing, 2010, 48(8):3178-3187. DOI:10.1109/TGRS.2010.2045506.
[10] 黄平平, 段盈宏, 谭维贤, 等. 基于融合差异图的变化检测方法及其在洪灾中的应用[J]. 雷达学报, 2021, 10(1):143-158. DOI:10.12000/JR20118.
[11] 任永梅, 杨杰, 郭志强, 等. 基于多尺度卷积神经网络的自适应熵加权决策融合船舶图像分类方法[J].电子与信息学报, 2021, 43(5):1424-1431. DOI:10.11999/JEIT200102.
[12] Zhao Q N, Gong M G, Li H, et al. Three-class change detection in synthetic aperture radar images based on deep belief network[C]//Bio-Inspired Computing:Theories and Applications, 2015:696-705. DOI:10.1007/978-3-662-49014-3_62.
[13] Gong M G, Yang H L, Zhang P Z. Feature learning and change feature classification based on deep learning for ternary change detection in SAR images[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2017, 129:212-225. DOI:10.1016/j.isprsjprs.2017.05.001.
[14] Gong M G, Niu X D, Zhang P Z, et al. Generative adversarial networks for change detection in multispectral imagery[J]. IEEE Geoscience and Remote Sensing Letters, 2017, 14(12):2310-2314. DOI:10.1109/LGRS.2017.2762694.
[15] 徐真, 王宇, 李宁, 等. 一种基于CNN的SAR图像变化检测方法[J]. 雷达学报, 2017, 6(5):483-491. DOI:10.12000/JR17075.
[16] El Amin A M, Liu Q J, Wang Y H. Zoom out CNNs features for optical remote sensing change detection[C]//2017 2nd International Conference on Image, Vision and Computing (ICIVC). June 2-4, 2017, Chengdu, China. IEEE, 2017:812-817. DOI:10.1109/ICIVC.2017.7984667.
[17] Li M K, Li M, Zhang P, et al. SAR image change detection using PCANet guided by saliency detection[J]. IEEE Geoscience and Remote Sensing Letters, 2019, 16(3):402-406. DOI:10.1109/LGRS.2018.2876616.
[18] 刘梦岚, 杨学志, 贾璐, 等. 融合多尺度深度特征的SAR图像变化检测[J]. 合肥工业大学学报(自然科学版),2020,43(6):741-748.DOI:10.3969/j.issn.1003-5060.2020.06.005.
[19] Liu F, Jiao L C, Tang X, et al. Local restricted convolutional neural network for change detection in polarimetric SAR images[J]. IEEE Transactions on Neural Networks and Learning Systems, 2019, 30(3):818-833. DOI:10.1109/TNNLS.2018.2847309.
[20] Gao Y H, Gao F, Dong J Y, et al. SAR image change detection based on multiscale capsule network[J]. IEEE Geoscience and Remote Sensing Letters, 2021, 18(3):484-488. DOI:10.1109/LGRS.2020.2977838.
[21] Chen J, Yuan Z Y, Peng J, et al. DASNet:dual attentive fully convolutional Siamese networks for change detection in high-resolution satellite images[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020, 14:1194-1206. DOI:10.1109/JSTARS.2020.3037893.
[22] Manian V B,Vasquez R E. Texture analysis and synthesis:a review of recent advances[C]//AeroSense 2003. Proc SPIE 5108, Visual Information Processing XII, Orlando, Florida, USA. 2003, 5108:242-250. DOI:10.1117/12.488206.
[23] Krinidis S, Chatzis V. A robust fuzzy local information C-Means clustering algorithm[J]. IEEE Transactions on Image Processing:a Publication of the IEEE Signal Processing Society, 2010, 19(5):1328-1337. DOI:10.1109/TIP.2010.2040763.
[24] Gao F, Dong J Y, Li B, et al. Automatic change detection in synthetic aperture radar images based on PCANet[J]. IEEE Geoscience and Remote Sensing Letters, 2016, 13(12):1792-1796. DOI:10.1109/LGRS.2016.2611001.
[25] Lu H Y, Wang H F, Zhang Q Q, et al. A dual-tree complex wavelet transform based convolutional neural network for human thyroid medical image segmentation[C]//2018 IEEE International Conference on Healthcare Informatics. June 4-7, 2018, New York, NY, USA. IEEE, 2018:191-198. DOI:10.1109/ICHI.2018.00029.
[26] Yu F, Koltun V. Multi-scale context aggregation by dilated convolutions[EB/OL]. arXiv:1511.07122v1. (2016-04-30). https://arxiv.org/abs/1511.07122v1.
[27] Hu J, Shen L, Sun G. Squeeze-and-excitation networks[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. June 18-23, 2018, Salt Lake City, UT, USA. IEEE, 2018:7132-7141. DOI:10.1109/CVPR.2018.00745.
[28] Celik T. Unsupervised change detection in satellite images using principal component analysis and k-means clustering[J]. IEEE Geoscience and Remote Sensing Letters, 2009, 6(4):772-776. DOI:10.1109/LGRS.2009.2025059.
[29] Gao F, Wang X, Gao Y H, et al. Sea ice change detection in SAR images based on convolutional-wavelet neural networks[J]. IEEE Geoscience and Remote Sensing Letters, 2019, 16(8):1240-1244. DOI:10.1109/LGRS.2019.2895656.
