AI Enhanced SVD Based Dense Fusion for Visible Image Integration

DOI

10.2991/978-94-6463-852-3_25How to use a DOI?

Keywords

Image fusion; CNN; Singular Value Decomposition (SVD)

Abstract

Singular Value Decomposition (SVD) has long been a trusted tool for merging images from different sources into a unified whole. In our work, we set out to reinvent image fusion by pairing the strength of SVD with the intuitive learning abilities of Convolutional Neural Networks (CNNs). Imagine an approach where a specially designed encoder-decoder system digs deep into each image, gently uncovering and blending its most important features. This process not only keeps the essential structure and color intact but also quietly discards any repetitive clutter. Once these images have been thoughtfully combined, SVD steps in again to refine the final picture—enhancing what matters and smoothing out any distracting noise so that the result is clear and vibrant.

The enhanced images aren’t just technically improved; they’re easier for us to interpret and appreciate. This clarity benefits a range of real-world applications, from spotting objects in a busy scene and recognizing diverse environments to bolstering security measures. Thanks to the adaptive nature of CNNs, the system learns and evolves with every new condition it faces, proving its worth in fields as varied as remote sensing, autonomous navigation, security, and even medical diagnostics. Our tests show that this fresh blend of CNNs with SVD not only meets the highest standards but also points toward an exciting future for image enhancement. Quantitative evaluations on standard datasets reveal that our approach achieves higher values in key metrics such as Peak Signal-to-Noise Ratio (PSNR).

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  - Harshal Unde
AU  - Bhushan S. Deore
AU  - Ashwini Naik
AU  - Somdotta Roy Choudhary
PY  - 2025
DA  - 2025/10/07
TI  - AI Enhanced SVD Based Dense Fusion for Visible Image Integration
BT  - Proceedings of the MULTINOVA: First International Conference on Artificial Intelligence in Engineering, Healthcare and Sciences (ICAIEHS- 2025)
PB  - Atlantis Press
SP  - 398
EP  - 416
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-852-3_25
DO  - 10.2991/978-94-6463-852-3_25
ID  - Unde2025
ER  -