Abstract
Polarimetric synthetic aperture radar (PolSAR) images with multiple polarimetric channels have high discrimination ability. So, they are appropriate data for classification applications. The suitable use of both scattering (polarimetric) and spatial features in a PolSAR image is important to provide an accurate classification map. Many basic classifiers such as support vector machine (SVM) only works based on feature vectors and ignore the spatial features. To convert a basic classifier such as SVM to a powerful classifier for PolSAR image classification, the polarimetric and spatial base iterative (PSI) classifier is proposed in this work. The PSI method tries to refine the labels of an initial classification map obtained by a simple classifier (SVM in this work). To this end, three terms of similarity based on spatial, polarimetric and class-specific features are computed in neighborhood regions. According to the computed similarity values, the new labels are generated and the scattering features of the PolSAR images are refined. This simple process is repeated several times to a pre-determined number of iterations. The result is a clean and accurate classification map. The experiments are done on two AIRSAR PolSAR images. Compared to the base classifier (SVM), the proposed method provides about 10% and 6% increase in classification accuracy of Flevoland and Sanfrancisco datasets with 19 and 50 iterations, respectively. Moreover, the proposed PSI method shows desirable performance compared to several complex and powerful classification methods.
Similar content being viewed by others
Data availability
No new datasets were generated in this paper. The datasets used for the experiments are publicly available datasets.
References
Chen Y et al (2019) A Novel Semicoupled Projective Dictionary Pair Learning Method for PolSAR Image Classification. IEEE Trans Geosci Remote Sens 57(4):2407–2418
Chen Y, Li L, Jiao L, Li Y, Liu X, Chai X (2021) Nonlinear Projective Dictionary Pair Learning for PolSAR Image Classification. IEEE Access 9:70650–70661
Chen J, Ou J, Zeng H, Cai C (2021) A fast algorithm based on gray level co-occurrence matrix and Gabor feature for HEVC screen content coding. J Vis Commun Image Represent 78:103128
Chen B, Wang S, Jiao L, Stolkin R, Liu H (2015) A Three-Component Fisher-Based Feature Weighting Method for Supervised PolSAR Image Classification. IEEE Geosci Remote Sens Lett 12(4):731–735
Chen L, Yang W, Liu Y, Sun H (2011) Polsar scene classification based on fast approximate neareast neighbours search, 2011 IEEE International Geoscience and Remote Sensing Symposium, Vancouver, BC, Canada, pp. 2137–2140
Garg R, Kumar A, Bansal N et al (2021) Semantic segmentation of PolSAR image data using advanced deep learning model. Sci Rep 11:15365
Ge S, Lu J, Gu H, Yuan Z, Su W (2017) Polarimetrie SAR image classification based on deep belief network and superpixel segmentation," 2017 3rd International Conference on Frontiers of Signal Processing (ICFSP), pp. 114–119
Golpardaz M, Helfroush MS, Danyali H, Ghaffari R (2021) Fully Statistical, Wavelet-based conditional random field (FSWCRF) for SAR image segmentation. Expert Syst Appl 168:114370
Guan H, Lei X, Yu Y, Zhao H, Peng D, Junior JM, Li J (2022) Road marking extraction in UAV imagery using attentive capsule feature pyramid network. Int J Appl Earth Obs Geoinf 107:102677
Guo J, Wang L, Zhu D Zhang G (2022) Semisupervised Classification of PolSAR Images Using a Novel Memory Convolutional Neural Network, IEEE Geosci Remote Sens Lett, 19, 1–5, Art no. 4007605
Han K, Jiang M, Wang M, Liu G (2018) Compact Polarimetric SAR Interferometry Target Decomposition With the Freeman-Durden Method. IEEE J Sel Top Appl Earth Obs Remote Sens 11(8):2847–2861
Hou B, Yang C, Ren B, Jiao L (2018) Decomposition-Feature-Iterative-Clustering-Based Superpixel Segmentation for PolSAR Image Classification. IEEE Geosci Remote Sens Lett 15(8):1239–1243
Hua W, Xie W, Jin X (2020) Three-Channel Convolutional Neural Network for Polarimetric SAR Images Classification. IEEE J Sel Top Appl Earth Observ Remote Sens 13:4895–4907
Imani M (2021) A Random Patches Based Edge Preserving Network for Land Cover Classification Using Polarimetric Synthetic Aperture Radar Images. Int J Remote Sens 42(13):4946–4964
Imani M (2022) Low Frequency and Radar’s Physical Based Features for Improvement of Convolutional Neural Networks for PolSAR Image Classification. Egypt J Remote Sens Space Sci 25:55–62
Imani M, Ghassemian H (2013) The Investigation of Sensitivity of SVM Classifier Respect to The Number of Features and The Number of Training Samples, 2nd International Conference on Sensors and Models in Photogrammetry and Remote Sensing, SMPR 2013, Tehran, Iran, 209–214
Imani M, Ghassemian H (2018) Discriminant analysis in morphological feature space for high-dimensional image spatial–spectral classification. J Appl Remote Sens 12(1):016024–28
Jamali, Mahdianpari M, Mohammadimanesh F, Bhattacharya A, Homayouni S (2022) PolSAR Image Classification Based on Deep Convolutional Neural Networks Using Wavelet Transformation, IEEE Geosci Remote Sens Lett, 19, 1–5, Art no. 4510105
Jamali A, Roy SK, Bhattacharya A, Ghamisi P (2023) Local Window Attention Transformer for Polarimetric SAR Image Classification. IEEE Geosci Remote Sens Lett, 20, 1–5, Art no. 4004205
Lê TT, Atto AM, Trouvé E, Solikhin A, Pinel V (2015) Change detection matrix for multitemporal filtering and change analysis of SAR and PolSAR image time series. ISPRS J Photogramm Remote Sens 107:64–76
Li L, Ma L, Jiao L, Liu F, Sun Q, Zhao J (2020) Complex Contourlet-CNN for polarimetric SAR image classification. Pattern Recogn 100:107110
Liu B, Zhang Z, Liu X, Yu W (2014) Edge Extraction for Polarimetric SAR Images Using Degenerate Filter With Weighted Maximum Likelihood Estimation. IEEE Geosci Remote Sens Lett 11(12):2140–2144
Luo J, Lv Y, Guo J (2022) Multi-temporal PolSAR Image Classification Using F-SAE-CNN, 2022 3rd China International SAR Symposium (CISS), Shanghai, China, pp. 1–5
Mahrooghy M, Aanstoos JV, Nobrega RAA, Hasan K, Prasad S, Younan NH (2015) A Machine Learning Framework for Detecting Landslides on Earthen Levees Using Spaceborne SAR Imagery. IEEE J Sel Top Appl Earth Obs Remote Sens 8(8):3791–3801
Mirzapour F, Ghassemian H (2013) Using GLCM and Gabor filters for classification of PAN images. 21st Iranian Conference on Electrical Engineering (ICEE 2013), Mashhad, Iran
Molaei S, Abadi MESA (2020) Maintaining filter structure: A Gabor-based convolutional neural network for image analysis. Appl Soft Comput 88:105960
Mousavi H, Imani M, Ghassemian H (2022) Deep Curriculum Learning for PolSAR Image Classification, 2022 International Conference on Machine Vision and Image Processing (MVIP), Ahvaz, Iran, pp. 1–5
Moya L, Zakeri H, Yamazaki F, Liu W, Mas E, Koshimura S (2019) 3D gray level co-occurrence matrix and its application to identifying collapsed buildings. ISPRS J Photogramm Remote Sens 149:14–28
Raj AJ, Idicula SM, Paul B (2022) A novel sarnede method for real-time ship detection from synthetic aperture radar image. Multimed Tools Appl 81:16921–16944
Ruggeri S, Henao-Cespedes V, Garcés-Gómez YA, Uzcátegui AP (2021) Optimized unsupervised CORINE Land Cover mapping using linear spectral mixture analysis and object-based image analysis. Egypt. J. Remote. Sens. Space Sci 24, 3 Part 2:1061–1069
Samat A, Li E, Du P, Liu S, Miao Z (2021) Improving Deep Forest via Patch-Based Pooling Morphological Profiling, and Pseudo Labeling for Remote Sensing Image Classification. IEEE J Sel Top Appl Earth Observ Remote Sens 14:9334–9349
Shang R, He J, Wang J, Xu K, Jiao L, Stolkin R (2020) Dense connection and depthwise separable convolution based CNN for polarimetric SAR image classification. Knowl-Based Syst 194:105542
Song W, Wu Y, Xiao X (2021) Nonstationary PolSAR Image Classification by Deep-Features-Based High-Order Triple Discriminative Random Field. IEEE Geosci Remote Sens Lett 18(8):1406–1410
Wang Y, Cheng J, Zhou Y, Zhang F, Yin Q (2022) A Multichannel Fusion Convolutional Neural Network Based on Scattering Mechanism for PolSAR Image Classification. IEEE Geosci Remote Sens Lett 19(1–5):4007805
Wang J, Hou B, Ren B, Zhang Y, Yang M, Wang S, Jiao L (2022) Parameter selection of Touzi decomposition and a distribution improved autoencoder for PolSAR image classification. ISPRS J Photogramm Remote Sens 186:246–266
Wang Y, Zhou GQ, Li MM (2019) Multiscale SAR image segmentation by combining curvelet transform and GMTRJ algorithms. Digital Signal Processing 95:102583
Xie W, Xie Z, Zhao F, Ren B (2018) POLSAR Image Classification via Clustering-WAE Classification Model. IEEE Access 6:40041–40049
Xie W, Yang X, Wang R, Zhao F (2022) Polsar Image Classification Via Auxiliary Classifier Generative Adversarial Network, IGARSS 2022–2022 IEEE International Geoscience and Remote Sensing Symposium. Kuala Lumpur, Malaysia, pp 1205–1208
Zhang L, Ma W, Zhang D (2016) Stacked Sparse Autoencoder in PolSAR Data Classification Using Local Spatial Information. IEEE Geosci Remote Sens Lett 13(9):1359–1363
Zhang Z, Pan X, He K et al (2021) SAR image segmentation with parallel region merging. Multimed Tools Appl 80:5701–5721
Zhang Z, Wang H, Xu F, Jin Y (2017) Complex-Valued Convolutional Neural Network and Its Application in Polarimetric SAR Image Classification. IEEE Trans Geosci Remote Sens 55(12):7177–7188
Zhao F, Ma G, Xie W, Liu H (2019) Semi-supervised recurrent complex-valued convolution neural network for polsar image classification, IGARSS 2019–2019 IEEE International Geoscience and Remote Sensing Symposium. Yokohama, Japan, pp 1–4
Zhu L, Ma X, Wu P, Xu J (2021) Multiple Classifiers Based Semi-Supervised Polarimetric SAR Image Classification Method. Sensors 21(9):3006
Zhu F, Zhang Y, Li D (2016) A novel interpretation of the PolSAR coherency matrix data, 2016 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Beijing, China, pp. 4730–4733
Zou B, Lu D, Zhang L, Moon WM (2017) Independent and Commutable Target Decomposition of PolSAR Data Using a Mapping From SU(4) to SO(6). IEEE Trans Geosci Remote Sens 55(6):3396–3407
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
There is no conflict of interest.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Imani, M. An iterative PolSAR image classification method with utilizing scattering and contextual information. Multimed Tools Appl 83, 16605–16621 (2024). https://doi.org/10.1007/s11042-023-16205-z
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-023-16205-z