Abstract

Polarimetric Synthetic Aperture Radar (PolSAR) data presents significant advantages compared to optical remote sensing, particularly due to its capability to obtain consistent imagery irrespective of solar illumination or atmospheric conditions. Although optical satellites are proficient in frequent data acquisition, their utility is frequently hindered by cloud cover and reliance on sunlight. In contrast, PolSAR operates independently of radiance and possesses the ability to penetrate cloud layers, thereby facilitating dependable, all-weather, day-and-night observations. Nonetheless, the intricate nature of PolSAR data and the expertise required for its analysis have posed significant challenges to widespread adoption. To mitigate this issue, the current study proposes a user-centric land cover classification tool aimed at simplifying the classification and interpretation of ALOS2-PALSAR L-band imagery. The tool utilizes a Random Forest classifier, which has been trained on labelled data from Mumbai, to categorize five distinct land cover types—water, settlements, forests, wetlands, and mangroves—across five designated regions: Mumbai, New Delhi, Ahmedabad, San Francisco, and California. The classification accuracy across these locations varies from 65% to over 95%, indicative of regional discrepancies in landscape structure and land cover categories. The tool is scalable for additional training sets with the possibility of global classification coverage and different classification techniques in its final form. The tool provides segmented images along with the area estimates for each land cover class.

References

• H. Parikh, S. Patel, and V. Patel, "Classification of SAR and PALSAR images using deep learning: a review," Int. J. Image Data Fusion, vol. 11, no. 1, pp. 1–32, 2019, doi: 10.1080/19479832.2019.1655489.
• E. Hossain, "App Designer and Graphical User Interface in MATLAB," in MATLAB and Simulink Crash Course for Engineers, Springer, Cham, 2022, doi: 10.1007/978-3-030-89762-8_11.
• B. Waske, S. van der Linden, C. Oldenburg, B. Jakimow, A. Rabe, and P. Hostert, "imageRF – A user-oriented implementation for remote sensing image analysis with Random Forests," Environ. Model. Softw., vol. 35, pp. 192–193, 2012, doi: 10.1016/j.envsoft.2012.01.014.
• M. Pal, "Random Forest classifier for remote sensing classification," Int. J. Remote Sens., vol. 26, no. 1, pp. 217–222, 2005, doi: 10.1080/01431160412331269698.
• S. S. Shinde and S. S. Patil, "Automatic land cover classification with SAR imagery and Machine learning using Google Earth Engine," International Journal of Electrical and Computer Engineering Systems, vol. 13, no. 10, pp. 909–916, Dec. 2022. doi: 10.32985/ijeces.13.10.6.
• R. Garg, A. Kumar, M. Prateek, K. Pandey, and S. Kumar, "Land cover classification of spaceborne multifrequency SAR and optical multispectral data using machine learning," Advances in Space Research, vol. 69, no. 5, pp. 1984–1997, Feb. 2022.
• J. Zhang, Y. Liu, Z. Wang, and Y. Li, "Land cover classification of SAR images based on ascending and descending orbit feature optimization and data fusion," 2025.
• L. Wang et al., "Research on land cover classification of multi-source remote sensing data based on improved U-net network," Scientific Reports, vol. 13, no. 1, Sep. 2023.
• S. Shetty et al., “MATSAR: A COMPREHENSIVE MACHINE LEARNING APPROACH FOR POLSAR DATA PROCESSING”, International Journal of Computer Applications. 187. 23-29. 10.5120/ijca2025924824.
• P. O. Gislason, J. A. Benediktsson, and J. R. Sveinsson, "Random Forests for land cover classification," Pattern Recognition Letters, vol. 27, no. 4, pp. 294 300, Mar. 2006, doi: 10.1016/j.patrec.2005.08.011.
• A. Lapini, S. Pettinato, E. Santi, S. Paloscia, G. Fontanelli, and A. Garzelli, "Comparison of machine learning methods applied to SAR images for forest classification in Mediterranean areas," Remote Sensing, vol. 12, no. 3, p. 369, 2020. DOI: 10.3390/rs12030369.
• F. Zhang and X. Yang, "Improving land cover classification in an urbanized coastal area by random forests: The role of variable selection," Remote Sensing of Environment, vol. 251, p. 112105, 2020. DOI: 10.1016/j.rse.2020.112105.
• D. Ming, T. Zhou, M. Wang, and T. Tan, "Land cover classification using random forest with genetic algorithm-based parameter optimization," Journal of Applied Remote Sensing, vol. 10, no. 3, p. 035021, 2016. DOI: 10.1117/1.jrs.10.035021.
• M. Belgiu and L. Drăguţ, "Random forest in remote sensing: A review of applications and future directions," ISPRS Journal of Photogrammetry and Remote Sensing, vol. 114, pp. 24–31, Apr. 2016. doi : 10.1016/j.isprsjprs.2016.01.011
• F. F. Camargo, E. E. Sano, C. M. Almeida, J. C. Mura, and T. Almeida, "A comparative assessment of machine-learning techniques for land use and land cover classification of the Brazilian tropical savanna using ALOS 2/PALSAR-2 polarimetric images," Remote Sensing, vol. 11, no. 13, p. 1600, 2019. DOI: 10.3390/rs11131600.
• V. Turkar, Juhi Checker, Shaunak De, and Gulab Singh, “Impact of G4U and 7-component target decomposition on PALSAR image semantic segmentation,” Advances in Space Research, vol. 70, no. 12, pp. 3798-3810, 2022, doi: 10.1016/j.asr.2022.08.004.
• G. Singh, R. Malik, S. Mohanty, V. S. Rathore, K. Yamada, M. Umemura, and Y. Yamaguchi, "Seven-Component Scattering Power Decomposition of POLSAR Coherency Matrix," IEEE Transactions on Geoscience and Remote Sensing, vol. 57, no. 11, pp. 8371–8382, Nov. 2019, doi: 10.1109/TGRS.2019.2920762.
• H. Fan, S. Quan, D. Dai, X. Wang, and S. Xiao, "Seven-component model-based decomposition for PolSAR data with sophisticated scattering models," Remote Sens., vol. 11, p. 2802, Nov. 2019
• W. Wang, X. Yang, G. Liu, H. Zhou, W. Ma, Y. Yu, and Z. Li, "Random forest classification of sediments on exposed intertidal flats using ALOS-2 quad-polarimetric SAR data," Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., vol. XLI-B8, pp. 1191–1194, 2016, doi: 10.5194/isprs-archives-XLI-B8-1191-2016.
• R. Kotru, V. Turkar, S. Simu, S. De, M. Shaikh, S. Banerjee, G. Singh, and A. Das, "Development of a generalized model to classify various land covers for ALOS-2 L-Band images using semantic segmentation," Advances in Space Research, vol. 70, no. 12, pp. 3811–3821, 2022, doi: 10.1016/j.asr.2022.07.078.
• V. Turkar, A. Masurkar, A. Das, and R. Daruwala, "Development of Generalized Machine Learning Model to Classify PALSAR Data," in IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium, Pasadena, CA, USA, 2023, pp. 7475-7478, doi: 10.1109/IGARSS52108.2023.10281706.
• F. F. Camargo, E. E. Sano, C. M. Almeida, J. C. Mura, and T. Almeida, "A Comparative Assessment of Machine-Learning Techniques for Land Use and Land Cover Classification of the Brazilian Tropical Savanna Using ALOS-2/PALSAR-2 Polarimetric Images," Remote Sensing, vol. 11, no. 13, Art. no. 1600, 2019, doi: 10.3390/rs11131600.