Fractional Optimization Model for Infrared and Visible Image Fusion
Kang Zhang (Nanjing University of Science and Technology), Shiwei Wu (Nanjing University of Science and Technology), Zhiliang Wu (Nanjing University of Science and Technology), Xia Yuan (Nanjing University of Science and Technology),* ChunXia Zhao (Nanjing university of science and technology)
The 33rd British Machine Vision Conference
Abstract
Infrared and visible image fusion is a fundamental task for image processing to enhance the image quality. To highlight target and retain effective details, different from previous methods using integer gradients, we use the fractional gradient to well represent image features and propose a novel fractional optimization model to fuse infraredand visible images. Specially, a fractional optimization function is designed with global contrast fidelity and fractional gradient constraint to obtain the pre-fused image. Then, the base layer of the pre-fused image obtained by multi-level decomposition latent low-rank representation is taken as the fused base layer, while for the fusion of detail layers, a fractional gradient energy function is designed to evaluate the importance of detail information to generate the fused detail layers. Compared with 10 state-of-the-art image fusion methods qualitatively and quantitatively on two public datasets (TNO and RoadScene), our method generally shows superior performance.Video
Citation
@inproceedings{Zhang_2022_BMVC,
author = {Kang Zhang and Shiwei Wu and Zhiliang Wu and Xia Yuan and ChunXia Zhao},
title = {Fractional Optimization Model for Infrared and Visible Image Fusion},
booktitle = {33rd British Machine Vision Conference 2022, {BMVC} 2022, London, UK, November 21-24, 2022},
publisher = {{BMVA} Press},
year = {2022},
url = {https://bmvc2022.mpi-inf.mpg.de/0458.pdf}
}
Copyright © 2022
The British Machine Vision Conference is organised by