This paper proposed a framework based on joint multi-feature representation learning to reduce the inherent speckle phenomenon in Polarimetric Synthetic Aperture Radar (PolSAR) images interfere with the scattering characteristics of land objects. Firstly, the corresponding 6-dimensional real vector is obtained from the covariance matrix of PolSAR data and combined with the polarized feature vector obtained by the polarization decomposition method to improve the differentiation ability of similar features in images. Secondly, the stacked sparse autoencoder (SSAE) is employed, where the non-negative constraint method is incorporated to make the sparse features in the depth space robust by filtering the weights. Finally, a non-negative constrained SSAE model is proposed to effectively accomplish the classification task of PolSAR images. In the experiments, the proposed classification network is trained layer by layer using unlabeled data, the softmax classifier is trained with a small number of labeled pixels. The parameters obtained from the above steps are used as initial parameters to train the entire classification framework with labeled pixels, the resulting well-trained model is used to predict the labels corresponding to pixels in the datasets. Through experiments using the Flevoland and San Francisco Bay datasets, the results demonstrate that the proposed method yields superior classification results compared with traditional SVM, AE, and Gray Level Co-generation Matrix (GLCM) classification methods.

Chen Ruixia, Zhu Shisong, Wang Zhensong, Wang Jianlong. An Innovative PolSAR Image Classification Method Based on Non-Negative Constraints Stacked Sparse Autoencoder Network with Multi-Features Joint Representation Learning. Academic Journal of Computing & Information Science (2023), Vol. 6, Issue 8: 119-130. https://doi.org/10.25236/AJCIS.2023.060815.

[1] Moreira A., Prats-Iraola, et al. A tutorial on synthetic aperture radar [J]. Geoscience & Remote Sensing Magazine IEEE, 2013. 

[2] Yin Q, Hong W, Zhang F, et al. Optimal Combination of Polarimetric Features for Vegetation Classification in PolSAR Image[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2019, PP (99):1-13. 

[3] West R D, Riley R M. Polarimetric Interferometric SAR Change Detection Discrimination [J]. IEEE Transactions on Geoscience and Remote Sensing, 2019, 57(6). 

[4] Wang H, Xu F, Jin Y Q. A Review of Polsar Image Classification: from Polarimetry to Deep Learning: IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium, 2019. 

[5] Lim Y X, Burgin M S, Zyl J J V. An Optimal Nonnegative Eigenvalue Decomposition for the Freeman and Durden Three-Component Scattering Model [J]. IEEE Transactions on Geoscience & Remote Sensing, 2017, 55(4):2167-2176. 

[6] Lemme A ,René Felix Reinhart, Steil J J .Online learning and generalization of parts-based image representations by non-negative sparse autoencoders[J].Neural Networks, 2012, 33(9):194-203.DOI:10.1016/j.neunet.2012.05.003.

[7] Hua W, Wang S, Liu H, et al. Semisupervised PolSAR Image Classification Based on Improved Cotraining [J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2017, PP (11):1-16. 

[8] Yahia M, Khalfa F, Chabir M, et al. Bias compensation in h/a/Α polarimetric sar decomposition and its implication for the classification[J]. Progress in Electromagnetics Research B, 2016, 68(1):105-121. 

[9] Manickam S, Bhattacharya A, Singh G, et al. Estimation of Snow Surface Dielectric Constant From Polarimetric SAR Data [J]. IEEE Journal of Selected Topics in Applied Earth Observations & Remote Sensing, 2016. 

[10] Yan L L, Zhang J X, Gao J X, et al. Four-Component Model-Based Decomposition of Polarimetric SAR Data for Oriented Urban Buildings[J]. Acta Electronica Sinica, 2015, 43(1):203-208. 

[11] Wang Y, Liu H, Jiu B. PolSAR Coherency Matrix Decomposition Based on Constrained Sparse Representation[J]. IEEE Transactions on Geoscience & Remote Sensing, 2014, 52(9):5906-5922. 

[12] Cloude S R, Pottier E. An entropy based classification scheme for land applications of polarimetric SAR [J]. IEEE Trans. Geo. Remote Sens, 1997, 35(1):68-78. 

[13] Freeman A, Durden S L. A three-component scattering model for polarimetric SAR data [J]. IEEE Transactions on Geoscience & Remote Sensing, 1998, 36(3):963-973. 

[14] Krogager E. New decomposition of the radar target scattering matrix [J]. Electronics Letters, 2002, 26(18):1525-1527. 

[15] Van Zyl J J. Unsupervised classification of scattering behavior using radar polarimetry data [J]. IEEE Trans. geosci. remote Sens, 1989, 27(1):36-45. 

[16] Zhou Y, Wang H, Xu F, et al. Polarimetric SAR Image Classification Using Deep Convolutional Neural Networks [J]. IEEE Geoscience & Remote Sensing Letters, 2017, 13(12):1935-1939. 

[17] Hinton G E. Learning multiple layers of representation [J]. Trends in Cognitive Sciences, 2007, 11(10): 428-434. 

[18] Baldi P, Guyon G, Dror V, et al. Autoencoders, unsupervised learning and deep architectures: UTLW'11 Proceedings of the 2011 International Conference on Unsupervised and Transfer Learning workshop - Volume 27, 2011. 

[19] Yin Y, Gelenbe E. Nonnegative autoencoder with simplified random neural network[J].  2016. DOI:10.48550/arXiv.1609.08151.

[20] Bengio Y. Learning Deep Architectures for AI [J]. Foundations & Trends in Machine Learning, 2009, 2(1):1-127. 

[21] Xu J, Xiang L, Hang R, et al. Stacked Sparse Autoencoder (SSAE) for Nuclei Detection on Breast Cancer Histopathology Images. [J]. IEEE, 2014. 

[22] Zhou Y, Wang H, Xu F, et al. Polarimetric SAR Image Classification Using Deep Convolutional Neural Networks[J]. IEEE Geoscience & Remote Sensing Letters, 2017, 13(12):1935-1939. 

[23] Chen S W, Tao C S. PolSAR Image Classification Using Polarimetric-Feature-Driven Deep Convolutional Neural Network [J]. IEEE Geoscience and Remote Sensing Letters, 2018. 

[24] Geng J, Fan J, Wang H, et al. High-Resolution SAR Image Classification via Deep Convolutional Autoencoders [J]. IEEE Geoscience & Remote Sensing Letters, 2015, 12(11):2351-2355. 

[25] Zhang L, Ma W, Zhang D. Stacked Sparse Autoencoder in PolSAR Data Classification Using Local Spatial Information[J]. IEEE Geoscience & Remote Sensing Letters, 2017, 13(9):1359-1363. 

[26] Hou B, Kou H, Jiao L. Classification of Polarimetric SAR Images Using Multilayer Autoencoders and Superpixels [J]. IEEE Journal of Selected Topics in Applied Earth Observations & Remote Sensing, 2017, 9(7):3072 -3081. 

[27] Wang J, Hou B, Jiao L, et al. POL-SAR Image Classification Based on Modified Stacked Autoencoder Network and Data Distribution[J]. IEEE Transactions on Geoscience and Remote Sensing, 2019, PP(99):1-18. 

[28] Zhang L, Jiao L, Ma W, et al. PolSAR Image Classification Based on Multi-scale Stacked Sparse Autoencoder [J]. Neurocomputing, 2019, 351(JUL. 25):167-179. 

[29] Hu Y, Fan J, Wang J. Classification of PolSAR Images Based on Adaptive Nonlocal Stacked Sparse Autoencoder [J]. IEEE Geoence & Remote Sensing Letters, 2018:1-5. 

[30] Li Y, Chen Y, Liu G, et al. A Novel Deep Fully Convolutional Network for PolSAR Image Classification [J]. Remote Sensing, 2018, 10(12):1984. 

[31] Chen S W, Wang X S, Sato M. PolInSAR Complex Coherence Estimation Based on Covariance Matrix Similarity Test [J]. IEEE Transactions on Geoscience and Remote Sensing, 2012, 50(11):4699-4710. 

[32] Zhou Y, Wang H, Xu F, et al. Polarimetric SAR Image Classification Using Deep Convolutional Neural Networks[J]. IEEE Geoscience & Remote Sensing Letters, 2017, 13(12):1935-1939. 

[33] Cloude S R, Pottier E. A review of target decomposition theorems in radar polarimetry [J]. IEEE Transactions on Geoscience & Remote Sensing, 1996, 34(2):498-518. 

[34] Lee J S, Grunes M R, De Grandi G. Polarimetric SAR speckle filtering and its implication for classification [J]. IEEE Transactions on Geoscience & Remote Sensing, 1999, 37(5):2363-2373.