SAR Image Colorization For Comprehensive Insight Using Deep Learning Model (H)

DOI

10.2991/978-94-6463-858-5_188How to use a DOI?

Keywords

Deep neural networks; SAR picture colorization; and polarimetric virtual aperture radar

Abstract

Many theories concerning radar polarimetry and polarimetric synthetic aperture radar (PolSAR) processing techniques have been established. Nevertheless, most of the SAR photographs are not fully polarimetric (full-pol). This paper suggests radar image colorization as a feasible way of reconstructing a full-pol image from a non-full-pol image, so that contemporary PolSAR techniques, such as model-based decomposition and unsupervised classification, can be readily used on the regenerated full-pol SAR images. It is proposed to train a dedicating deep neural network that is intended for the task of converting a single polarization gray-scale SAR images in full-pol. The neural network is composed of two parts: a feature extraction network that develops multi-scale spatial results, which are obtained from a grayscale SAR image, and a figure translation network that converts spatial features into polarimetric features, which enables the reconstruction of the polarimetric covariance matrix for each pixel. Both qualitative and quantitative experiments with actual full-pol datasets are carried out to support the effectiveness of this technique. The generated full-polarimetric synthetic aperture radar image is consistent with actual full-polarimetric images not only in terms of their visual likeness but also in real-world PolSAR uses, such as terrain classification and target decomposition.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - J. Pallavi
AU  - B. Shubha Deepika
AU  - Jayalakshmi Machiraju
AU  - M. Manisha
PY  - 2025
DA  - 2025/11/04
TI  - SAR Image Colorization For Comprehensive Insight Using Deep Learning Model (H)
BT  - Proceedings of International Conference on Computer Science and Communication Engineering (ICCSCE 2025)
PB  - Atlantis Press
SP  - 2257
EP  - 2267
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-858-5_188
DO  - 10.2991/978-94-6463-858-5_188
ID  - Pallavi2025
ER  -