Infrared and Visible Image Fusion Using Multi-level Adaptive Fractional Differential


Kang Zhang (Nanjing University of Science and Technology), Xinnian Guo (Suqian University)
The 35th British Machine Vision Conference
PDFPosterVideo (Right click to download)

Abstract

Good texture detail retention is important for image fusion. To more fully extract the multi-level image features and maintain good texture details, we put forward a multi-level adaptive fractional differential method for infrared and visible image fusion. In particular, fractional differential has global correlation, which can well reflect the dependence of image changes and take full advantage of the detail features. In the proposed method, we adopt multi-level decomposition method (MDLatLRR) to decompose the source images into base layers and detail layers. Specially, we design a new fractional order calculation method through modified spatial frequency and average gradient. Based on this, an adaptive fractional differential model is designed to fuse the detail layers, in which a fractional gradient energy function is proposed to weigh the importance of the corresponding detail features. Compared with other image fusion methods qualitatively and quantitatively on two public datasets, the proposed method generally shows superior performance.

Citation

@inproceedings{Zhang_2024_BMVC,
author    = {Kang Zhang and Xinnian Guo},
title     = {Infrared and Visible Image Fusion Using Multi-level Adaptive Fractional Differential},
booktitle = {35th British Machine Vision Conference 2024, {BMVC} 2024, Glasgow, UK, November 25-28, 2024},
publisher = {BMVA},
year      = {2024},
url       = {https://papers.bmvc2024.org/0201.pdf}
}


Copyright © 2024 The British Machine Vision Association and Society for Pattern Recognition
The British Machine Vision Conference is organised by The British Machine Vision Association and Society for Pattern Recognition. The Association is a Company limited by guarantee, No.2543446, and a non-profit-making body, registered in England and Wales as Charity No.1002307 (Registered Office: Dept. of Computer Science, Durham University, South Road, Durham, DH1 3LE, UK).

Imprint | Data Protection