TU Darmstadt / ULB / TUbiblio

An Adaptive Two-Scale Image Fusion of Visible and Infrared Images

Han, Xiyu and Lv, Tao and Song, Xiangyu and Nie, Ting and Liang, Huaidan and He, Bin and Kuijper, Arjan (2019):
An Adaptive Two-Scale Image Fusion of Visible and Infrared Images.
In: IEEE Access, pp. 56341-56352, 7, ISSN 2169-3536,
DOI: 10.1109/ACCESS.2019.2913289,
[Online-Edition: https://doi.org/10.1109/ACCESS.2019.2913289],
[Article]

Abstract

In this paper, we proposed an adaptive two-scale image fusion method using latent low-rank representation (LatLRR). Firstly, both IR and VI images are decomposed into a two-scale representation using LatLRR to generate low-rank parts (the global structure) and saliency parts (the local structure). The algorithm denoises at the same time. Then, the guided filter is used in the saliency parts to make full use of the spatial consistency, which reduces artifacts effectively. With respect to the fusion rule of the low-rank parts, we construct adaptive weights by adopting fusion global-local-topology particle swarm optimization (FGLT-PSO) to obtain more useful information from the source images. Finally, the resulting image is reconstructed by adding the fused low-rank part and the fused saliency part. Experimental results validate that the proposed method outperforms several representative image fusion algorithms on publicly available datasets for infrared and visible image fusion in terms of subjective visual effect and objective assessment.

Item Type: Article
Erschienen: 2019
Creators: Han, Xiyu and Lv, Tao and Song, Xiangyu and Nie, Ting and Liang, Huaidan and He, Bin and Kuijper, Arjan
Title: An Adaptive Two-Scale Image Fusion of Visible and Infrared Images
Language: English
Abstract:

In this paper, we proposed an adaptive two-scale image fusion method using latent low-rank representation (LatLRR). Firstly, both IR and VI images are decomposed into a two-scale representation using LatLRR to generate low-rank parts (the global structure) and saliency parts (the local structure). The algorithm denoises at the same time. Then, the guided filter is used in the saliency parts to make full use of the spatial consistency, which reduces artifacts effectively. With respect to the fusion rule of the low-rank parts, we construct adaptive weights by adopting fusion global-local-topology particle swarm optimization (FGLT-PSO) to obtain more useful information from the source images. Finally, the resulting image is reconstructed by adding the fused low-rank part and the fused saliency part. Experimental results validate that the proposed method outperforms several representative image fusion algorithms on publicly available datasets for infrared and visible image fusion in terms of subjective visual effect and objective assessment.

Journal or Publication Title: IEEE Access
Volume: 7
Uncontrolled Keywords: Image fusion, Infrared light, Image processing, Consistency, Spatial data
Divisions: 20 Department of Computer Science
20 Department of Computer Science > Mathematical and Applied Visual Computing
Date Deposited: 19 Jun 2019 11:08
DOI: 10.1109/ACCESS.2019.2913289
Official URL: https://doi.org/10.1109/ACCESS.2019.2913289
Export:
Suche nach Titel in: TUfind oder in Google

Optionen (nur für Redakteure)

View Item View Item