Title:Automated Colorization of Synthetic Aperture Radar Images Using Deep Learning Model
Authors:Chetna Sontaki, Dhruv Sharma, Divyanshu Aaliwal, Shruti Lashkari, Nidhi Nigam
Published in: Volume 2 Issue 2 July-December 2025, Page No. 40-45
DOI: https://doi.org/10.63844/IJAITR.v2.i1.2025.22-29
cite
Keywords: Convolution Neural Network, Noise
Reduction, Aperture Radar Images
Abstract:
This research explores the application of
deep learning models to colorize grayscale Synthetic
Aperture Radar images, improving their interpretability for remote sensing tasks. SAR is widely used in
fields such as geological studies and environmental
monitoring due to its ability to capture images in all
weather conditions. However, the grayscale nature of
SAR images limits their usability for visual analysis.
The study focuses on developing deep learning architectures, including Convolutional Neural Networks
(CNNs), Generative Adversarial Networks (GANs),
and autoencoders, to predict and apply appropriate
color schemes to SAR data. Key challenges such as
noise reduction, model design, and data limitations
are addressed through careful data pre-processing and
model training. The colorized images aim to enhance
feature distinction, enabling more accurate interpretation in applications like geological mapping and environmental monitoring. Validation methods include
both quantitative metrics like Mean Squared Error
(MSE) and qualitative evaluations by remote sensing
experts. The anticipated outcome is a more efficient
and visually interpretable SAR image analysis process,
allowing for better insights into natural phenomena
and improved decision-making in remote sensing applications. This research contributes to advancing the
state of SAR image analysis by introducing automated
deep learning-driven colorization techniques.
Download PDF |