Home / Current Issue / Paper 1705252
Polarimetric SAR Data Denoising using SOFM
Subject area: Science,Engineering and Technology · Area of research: Image Processing
Abstract
Synthetic Aperture Radar (SAR) data plays a critical role in remote sensing applications, providing valuable information for various fields, including environmental monitoring, disaster management, and defense. However, SAR data is often contaminated by noise, which can degrade the quality of the information extracted from it. In this research, we address the problem of denoising polarimetric SAR data using Self-Organizing Feature Maps (SOFM), a neural network-based approach. The motivation behind this work stems from the necessity to enhance the quality and accuracy of SAR data for improved interpretation and analysis. We propose a methodology that leverages the unsupervised learning capabilities of SOFM to reduce noise in polarimetric SAR images. The key idea is to find the best-matching unit (BMU) for each pixel in the data and use it to update the noisy pixel with information from its corresponding BMU. In the literature review, we discuss the significance of SAR data and review existing denoising techniques, highlighting the advantages and limitations of each. We present the design strategy, including the choice of SOFM as the denoising tool and the selection of parameters. The methodology section offers a detailed description of the denoising process, emphasizing the calculation of BMUs and the update of noisy data based on these BMUs. We also provide insights into the design of the SOFM network and its training process. While partial results from the validation process are presented, the project remains a work in progress, with ongoing experiments and further analysis. Our approach shows promise in improving the quality of polarimetric SAR data, but comprehensive validation and fine-tuning are required to assess its full potential. In conclusion, this project aims to contribute to the advancement of SAR data processing by introducing a novel denoising technique based on Self-Organizing Feature Maps. By reducing noise in polarimetric SAR data, our methodology has the potential to enhance the accuracy and reliability of information derived from SAR images, thus benefiting a wide range of applications in remote sensing.
Keywords
Polarimetric SAR Data, Denoising, Self-Organizing Feature Maps (SOFM), Remote Sensing, Synthetic Aperture Radar (SAR), Noise Reduction
References
[1] Shitole, Sanjay, et al. "Selection of suitable window size for speckle reduction and deblurring using SOFM in polarimetric SAR images." Journal of the Indian society of remote sensing 43 (2015): 739-750.
[2] Shitole, S., De, S., Rao, Y. S., Krishna Mohan, B., & Das, A. (2015). Selection of suitable window size for speckle reduction and deblurring using SOFM in polarimetric SAR images. Journal of the Indian society of remote sensing, 43, 739-750.
[3] Shitole, Sanjay, Shaunak De, Y. S. Rao, B. Krishna Mohan, and Anup Das. "Selection of suitable window size for speckle reduction and deblurring using SOFM in polarimetric SAR images." Journal of the Indian society of remote sensing 43 (2015): 739-750.
[4] Shitole S, Rao YS, Mohan BK, Das A. Self-organizing feature map based polarimetric SAR data denoising. In2013 IEEE International Geoscience and Remote Sensing Symposium-IGARSS 2013 Jul 21 (pp. 2373-2376). IEEE.
[5] Gierszewska, Monika, and Tomasz Berezowski. "On the role of polarimetric decomposition and speckle filtering methods for C-Band SAR wetland classification purposes." IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 15 (2022): 2845-2860.
[6] Jain, V., Shitole, S., & Rahman, M. (2023). Performance evaluation of DFT based speckle reduction framework for synthetic aperture radar (SAR) images at different frequencies and image regions. Remote Sensing Applications: Society and Environment, 101001.
[7] Parhad, S.V., Aher, S.A. and Warhade, K.K., 2021, October. A Comparative Analysis of Speckle Noise Removal in SAR Images. In 2021 2nd Global Conference for Advancement in Technology (GCAT) (pp. 1-4). IEEE.
[8] Chouhan, Siddharth Singh, Ajay Kaul, and Uday Pratap Singh. "Image segmentation using computational intelligence techniques." Archives of Computational Methods in Engineering 26 (2019): 533-596.
[9] Gierszewska M, Berezowski T. On the role of polarimetric decomposition and speckle filtering methods for C-Band SAR wetland classification purposes. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 2022 Mar 28;15:2845-60.
[10] Jain, V., Shitole, S. and Rahman, M., 2023. Performance evaluation of DFT based speckle reduction framework for synthetic aperture radar (SAR) images at different frequencies and image regions. Remote Sensing Applications: Society and Environment, p.101001.
How to cite this paper
@article{1705252,
author = {Amit Kumar Pandey, Mithilesh Vishwakarma, Bipin Yadav, Gopal Rajbhar},
title = {Polarimetric SAR Data Denoising using SOFM},
journal = {Iconic Research And Engineering Journals},
year = {2023},
volume = {7},
number = {6},
pages = {42-48},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1705252.pdf},
abstract = {Synthetic Aperture Radar (SAR) data plays a critical role in remote sensing applications, providing valuable information for various fields, including environmental monitoring, disaster management, and defense. However, SAR data is often contaminated by noise, which can degrade the quality of the information extracted from it. In this research, we address the problem of denoising polarimetric SAR data using Self-Organizing Feature Maps (SOFM), a neural network-based approach. The motivation behind this work stems from the necessity to enhance the quality and accuracy of SAR data for improved interpretation and analysis. We propose a methodology that leverages the unsupervised learning capabilities of SOFM to reduce noise in polarimetric SAR images. The key idea is to find the best-matching unit (BMU) for each pixel in the data and use it to update the noisy pixel with information from its corresponding BMU. In the literature review, we discuss the significance of SAR data and review existing denoising techniques, highlighting the advantages and limitations of each. We present the design strategy, including the choice of SOFM as the denoising tool and the selection of parameters. The methodology section offers a detailed description of the denoising process, emphasizing the calculation of BMUs and the update of noisy data based on these BMUs. We also provide insights into the design of the SOFM network and its training process. While partial results from the validation process are presented, the project remains a work in progress, with ongoing experiments and further analysis. Our approach shows promise in improving the quality of polarimetric SAR data, but comprehensive validation and fine-tuning are required to assess its full potential. In conclusion, this project aims to contribute to the advancement of SAR data processing by introducing a novel denoising technique based on Self-Organizing Feature Maps. By reducing noise in polarimetric SAR data, our methodology has the potential to enhance the accuracy and reliability of information derived from SAR images, thus benefiting a wide range of applications in remote sensing.},
keywords = {Polarimetric SAR Data, Denoising, Self-Organizing Feature Maps (SOFM), Remote Sensing, Synthetic Aperture Radar (SAR), Noise Reduction},
month = {December},
}