Scattering feature dimension reduction of multitemporal fully PolSAR image based on Stacked Sparse AutoEncoder

Henghui LI; Jiao GUO; Wenting HAN; Yanyang LIU; Jifeng NING

doi:10.11834/jrs.20209366

Technology and Methodology | Views : 0 下载量: 30 CSCD: 3

PDF
Export
Share
Collection
Album

Scattering feature dimension reduction of multitemporal fully PolSAR image based on Stacked Sparse AutoEncoder
Vol. 24, Issue 11, Pages: 1379-1391(2020)
DOI： 10.11834/jrs.20209366

扫描看全文

Quote
李恒辉，郭交，韩文霆，刘艳阳，宁纪锋.2020.栈式稀疏自编码网络的多时相全极化SAR散射特征降维.遥感学报，24（11）： 1379-1391

Li H H，Guo J，Han W T，Liu Y Y and Ning J F. 2020. Scattering feature dimension reduction of multitemporal fully PolSAR image based on Stacked Sparse AutoEncoder. Journal of Remote Sensing（Chinese）， 24（11）：1379-1391
李恒辉，郭交，韩文霆，刘艳阳，宁纪锋.2020.栈式稀疏自编码网络的多时相全极化SAR散射特征降维.遥感学报，24（11）： 1379-1391 DOI： 10.11834/jrs.20209366.

Li H H，Guo J，Han W T，Liu Y Y and Ning J F. 2020. Scattering feature dimension reduction of multitemporal fully PolSAR image based on Stacked Sparse AutoEncoder. Journal of Remote Sensing（Chinese）， 24（11）：1379-1391 DOI： 10.11834/jrs.20209366.

摘要

利用极化合成孔径雷达（PolSAR）能够实现地物的识别和分类，而多时相全极化SAR可以获取地物更多的散射特征，提升地物识别精度，但高维散射特征的引入会带来严重的维数灾难问题。为了实现对高维散射特征的有效降维，本文提出一种基于栈式稀疏自编码网络S-SAE（Stacked Sparse AutoEncoder）的多时相PolSAR散射特征降维方法。该方法首先对PolSAR数据进行极化目标分解以获取高维散射特征；然后使用S-SAE对获取的多维特征进行降维处理，其中S-SAE降维方法首先采用无监督训练方式进行逐层贪婪训练；再结合Sigmod分类器，利用监督训练的方式对S-SAE进行参数优化，实现高维特征的有效降维；最后以降维后的特征作为支持向量机（SVM）和卷积神经网络（CNN）分类器的输入，实现地物分类。通过仿真和实测的两组多时相Sentinel-1数据处理结果表明，双隐层的S-SAE降维方法在各分类器上均取得最优的降维效果；对比各降维方法在SVM分类器上的分类精度，S-SAE较于局部线性嵌入（LLE）与主成分分析（PCA）降维方法，总体分类精度分别至少提升了9%和14%；在CNN分类器上，S-SAE较于LLE与PCA降维方法，总体分类精度分别至少提升了7%和9%。

Abstract

Polarimetric Synthetic Aperture Radar (PolSAR) has been proven to recognize and classify various ground objects, and multitemporal fully PolSAR can acquire many scattering features to improve the accuracy of recognition and classification. However, the decomposed scattering features with high dimensionality can cause serious problems of “curse of dimensionality.” This paper proposes a multitemporal PolSAR scattering feature dimension reduction method based on Stacked Sparse AutoEncoder (S-SAE) to effectively reduce the dimensionality of high-dimensional scattering features. The proposed method firstly decomposes the PolarSAR data to obtain high-dimensional scattering features and adopts S-SAE to reduce the dimensionality of the acquired multidimensional features. For the S-SAE construction, unsupervised layer-by-layer greedy training is performed to optimize the main parameters. Combined with a sigmoid classifier, the parameters of S-SAE are finely tuned through supervised training to achieve effective dimension reduction of high-dimensional features. The reduced low-dimensional features are taken as the input of Support Vector Machine (SVM) and Convolutional Neural Network (CNN) classifiers to evaluate the performance of feature dimension reduction. The performance of the proposed method is validated on two datasets of multitemporal simulated and real Sentinel-1 data. Experimental results show that the S-SAE method with two hidden layers achieves the best performance of feature dimension reduction. Compared with traditional Locally Linear Embedding (LLE) and Principal Component Analysis (PCA) dimension reduction methods, the overall classification accuracy of S-SAE for the SVM classifier is raised by at least 9% and 14%. The overall classification accuracy of S-SAE for the CNN classifier is at least 7% and 9% higher than that of LLE and PCA, respectively.

关键词

特征降维作物分类极化合成孔径雷达多时相栈式稀疏自编码网络卷积神经网络

Keywords

dimension reductioncrop classificationPolarimetric Synthetic Aperture Radar (PolSAR)multi-temporalStack Sparse AutoEncoder(S-SAE)CNN

references

Alberga V, Satalino G and Staykova D K. 2008. Comparison of polarimetric SAR observables in terms of classification performance. International Journal of Remote Sensing, 29(14): 4129-4150 [DOI: 10.1080/01431160701840182http://dx.doi.org/10.1080/01431160701840182]

Caves R. 2011. AgriSAR 2009-Final Report. MDA System. https://earth.esa.int/c/document_library/get_file?folderId=87248&name=DLFE-1316.pdfhttps://earth.esa.int/c/document_library/get_file?folderId=87248&name=DLFE-1316.pdf

Chen S W and Tao C S. 2017. Multi-temporal PolSAR crops classification using polarimetric-feature-driven deep convolutional neural network//2017 International Workshop on Remote Sensing with Intelligent Processing (RSIP). Shanghai, China: IEEE: 1-4 [DOI: 10.1109/RSIP.2017.7958818http://dx.doi.org/10.1109/RSIP.2017.7958818]

Chen S W, Wang X S and Sato M. 2014a. Uniform polarimetric matrix rotation theory and its applications. IEEE Transactions on Geoscience and Remote Sensing, 52(8): 4756-4770 [DOI: 10.1109/TGRS.2013.2284359http://dx.doi.org/10.1109/TGRS.2013.2284359]

Chen Y S, Lin Z H, Zhao X, Wang G and Gu Y F. 2014b. Deep learning-based classification of hyperspectral data. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 7(6): 2094-2107 [DOI: 10.1109/JSTARS.2014.2329330http://dx.doi.org/10.1109/JSTARS.2014.2329330]

Cloude S R and Pottier E. 1997. An entropy based classification scheme for land applications of polarimetric SAR. IEEE Transactions on Geoscience and Remote Sensing, 35(1): 68-78 [DOI: 10.1109/36.551935http://dx.doi.org/10.1109/36.551935]

Cortes C and Vapnik V. 1995. Support-vector networks. Machine Learning, 20(3): 273-297 [DOI: 10.1007/BF00994018http://dx.doi.org/10.1007/BF00994018]

Dong G G, Liao G S, Liu H W and Kuang G Y. 2018. A review of the AutoEncoder and its variants: a comparative perspective from target recognition in synthetic-aperture radar images. IEEE Geoscience and Remote Sensing Magazine, 6(3): 44-68 [DOI: 10.1109/MGRS.2018.2853555http://dx.doi.org/10.1109/MGRS.2018.2853555]

Freeman A, Van Zyl J J, Klein J D, Zebker H A and Shen Y. 1992. Calibration of Stokes and scattering matrix format polarimetric SAR data. IEEE Transactions on Geoscience and Remote Sensing, 30(3): 531-539 [DOI: 10.1109/36.142931http://dx.doi.org/10.1109/36.142931]

Goodfellow I, Bengio Y and Courville A. 2016. Deep Learning. Cambridge, MA: MIT Press: 505-507

Guo J, Wei P L, Liu J, Jin B, Su B F and Zhou Z S. 2018. Crop classification based on differential characteristics of H/α scattering parameters for multitemporal quad- and dual-polarization SAR images. IEEE Transactions on Geoscience and Remote Sensing, 56(10): 6111-6123 [DOI: 10.1109/TGRS.2018.2832054http://dx.doi.org/10.1109/TGRS.2018.2832054]

Hinton G E and Salakhutdinov R R. 2006. Reducing the dimensionality of data with neural networks. Science, 313(5786): 504-507 [DOI: 10.1126/science.1127647http://dx.doi.org/10.1126/science.1127647]

Huynen J R. 1970. Phenomenological Theory of Radar Targets. The Netherlands: Technical University Delft: 653-712

Lee J S, Grunes M R and Kwok R. 1994. Classification of multi-look polarimetric SAR imagery based on complex Wishart distribution. International Journal of Remote Sensing, 15(11): 2299-2311 [DOI: 10.1080/01431169408954244http://dx.doi.org/10.1080/01431169408954244]

Liu X, Jiao L C, Tang X, Sun Q G and Zhang D. 2019. Polarimetric convolutional network for polsar image classification. IEEE Transactions on Geoscience and Remote Sensing, 57(5): 3040-3054 [DOI: 10.1109/TGRS.2018.2879984http://dx.doi.org/10.1109/TGRS.2018.2879984]

Paul S and Kumar D N. 2018. Spectral-spatial classification of hyperspectral data with mutual information based segmented stacked AutoEncoder approach. ISPRS Journal of Photogrammetry and Remote Sensing, 138: 265-280 [DOI: 10.1016/j.isprsjprs.2018.02.001http://dx.doi.org/10.1016/j.isprsjprs.2018.02.001]

Shi L, Zhang L F, Zhao L L, Yang J, Li P X and Zhang L P. 2013. The potential of linear discriminative Laplacian eigenmaps dimensionality reduction in polarimetric SAR classification for agricultural areas. ISPRS Journal of Photogrammetry and Remote Sensing, 86: 124-135 [DOI: 10.1016/j.isprsjprs.2013.09.013http://dx.doi.org/10.1016/j.isprsjprs.2013.09.013]

Shi L, Zhang L F, Zhao L L, Zhang L P, Li P X and Wu D. 2016. Adaptive laplacian eigenmap-based dimension reduction for ocean target discrimination. IEEE Geoscience and Remote Sensing Letters, 13(7): 902-906 [DOI: 10.1109/LGRS.2016.2553046http://dx.doi.org/10.1109/LGRS.2016.2553046]

Shi Z, Liang Z Z, Yang Y Y and Guo Y. 2015. Research status and prospect of agricultural remote sensing. Transactions of the Chinese Society for Agricultural Machinery, 46(2): 247-260

史周, 梁宗正, 杨媛媛, 郭燕. 2015. 农业遥感研究现状与展望. 农业机械学报, 46(2): 247-260 [DOI: 10.6041/j.issn.1000-1298.2015.02.037http://dx.doi.org/10.6041/j.issn.1000-1298.2015.02.037]

Tamiminia H, Homayouni S, McNairn H and Safari A. 2017. A particle swarm optimized kernel-based clustering method for crop mapping from multi-temporal polarimetric L-band SAR observations. International Journal of Applied Earth Observation and Geoinformation, 58: 201-212 [DOI: 10.1016/j.jag.2017.02.010http://dx.doi.org/10.1016/j.jag.2017.02.010]

Wang X S, Chen E X, Li Z Y, Yao W Q and Zhao L. 2015. Multi-temporal and dual-polarization interferometric SAR for land cover type classification. Acta Geodaetica et Cartographica Sinica, 44(5): 533-540

王馨爽, 陈尔学, 李增元, 姚顽强, 赵磊. 2015. 多时相双极化合成孔径雷达干涉测量土地覆盖分类方法. 测绘学报, 44(5): 533-540 [DOI: 10.11947/j.AGCS.2015.20130244http://dx.doi.org/10.11947/j.AGCS.2015.20130244]

Xu Q, Zhang X, Yu S H, Chen Q H and Liu X G. 2019. Multi-feature-based classification method using random forest and superpixels for polarimetric SAR images. Journal of Remote Sensing, 23(4): 685-694

徐乔, 张霄, 余绍淮, 陈启浩, 刘修国. 2019. 综合多特征的极化SAR图像随机森林分类算法. 遥感学报, 23(4): 685-694 [DOI: 10.11834/jrs.20197475http://dx.doi.org/10.11834/jrs.20197475]

Yamaguchi Y, Moriyama T, Ishido M and Yamada H. 2005. Four-component scattering model for polarimetric SAR image decomposition. IEEE Transactions on Geoscience and Remote Sensing, 43(8): 1699-1706 [DOI: 10.1109/TGRS.2005.852084http://dx.doi.org/10.1109/TGRS.2005.852084]

Yekkehkhany B, Homayouni S, McNairn H and Safari A. 2014. Multi-temporal full polarimetry L-band SAR data classification for agriculture land cover mapping//2014 IEEE Geoscience and Remote Sensing Symposium. Quebec, QC: IEEE: 2770-2773 [DOI: 10.1109/IGARSS.2014.6947050http://dx.doi.org/10.1109/IGARSS.2014.6947050]

Yuan Y X, Sun L and Zhang Q. 2018. Authentication technology via stack sparse AutoEncoder and micro-motion feature. Journal of Air Force Engineering University (Natural Science Edition), 19(4): 48-53

袁延鑫, 孙莉, 张群. 2018. 基于堆栈稀疏自编码器和微动特征的身份认证技术. 空军工程大学学报(自然科学版), 19(4): 48-53 [DOI: 10.3969/j.issn.1009-3516.2018.04.009http://dx.doi.org/10.3969/j.issn.1009-3516.2018.04.009]

Zhang K, Hei B Q, Zhou Z and Li S Y. 2018. CNN with coefficient of variation-based dimensionality reduction for hyperspectral remote sensing images classification. Journal of Remote Sensing, 22(1): 87-96

张康, 黑保琴, 周壮, 李盛阳. 2018. 变异系数降维的CNN高光谱遥感图像分类. 遥感学报， 22(1): 87-96 [DOI: 10.11834/JRS.20187075http://dx.doi.org/10.11834/JRS.20187075]

Zhao L L, Yang J, Li P X, Lang F K and Shi L. 2013. Statistical classification of weak backscattering scatterers of PolSAR image. Journal of Remote Sensing, 17(2): 306-319

赵伶俐，杨杰，李平湘，郎丰铠，史磊. 2013. 极化SAR影像弱散射地物统计分类．遥感学报, 17(2): 306-319 [DOI: 10.11834/jrs.20132018http://dx.doi.org/10.11834/jrs.20132018]

Zhu X X, Tuia D, Mou L C, Xia G S, Zhang L P, Xu F and Fraundorfer F. 2017. Deep learning in remote sensing: a comprehensive review and list of resources. IEEE Geoscience and Remote Sensing Magazine, 5(4): 8-36

DOI 10.1109/MGRS.2017.2762307

Alert me when the article has been cited

提交

Crop classification based on two-dimensional representation and CNN model from remote sensing

Mapping cropland at metric resolution using the spatiotemporal information from multi-source GF satellite data

Separating land surface component temperatures from Low Earth Orbit （LEO） satellite data by coupling of dual-time and multi-pixel data

CNN with local binary patterns for hyperspectral images classification

Enhanced multi-temporal cloud detection algorithm for optical remote sensing images