Fig 1.
Schematic illustration of image fusion.
(a) IR, (b) Vis, (c) GTF, (d) FusionGAN, (e) MDLatLRR, (f) Our.
Fig 2.
Example of the multi-scale decomposition (a) source image, (b) the proposed multi-scale decomposition.
Fig 3.
A surf plot of the proposed clustering algorithm versus ADF clustering results.
The first row includes, from left to right: infrared image, ADF clustering results, and the proposed algorithm clustering results; the second row includes, from left to right: infrared image partial area surf map, ADF clustering partial area surf map, and the proposed clustering algorithm partial area surf map. The abscissa of the Surf map represents the spatial position of the pixel, and the ordinate represents the size of the pixel value.
Fig 4.
Multi-scale decomposed images for visible-light and infrared images.
I(ir) represents the infrared image, I(vis) represents the visible image, and the yellow cube represents the proposed smoothing method. The processed infrared and visible images are denoted as IM(ir) and IM(vis), respectively. Through Eq (8), the visible image is decomposed into bright detail layer IH(vis) and dark detail layer IL(vis), while, through Eq (9), the infrared image is decomposed into bright detail layer IH(ir) and dark detail layer IL(ir).
Fig 5.
Formulation process of IZ at intermediate frequency layer.
Fig 6.
The fusion process to obtain the final result IF.
Fig 7.
Fusion results according to the clustering window dimension.
(a) n = 0, (b) n = 50, (c) n = 100, (d) n = 150, (e) n = 200.
Fig 8.
Fusion results according to mean filter window dimension.
(a)m = 3, (b) m = 5, (c) m = 7, (d) m = 9, (e) m = 11.
Table 1.
The flow chart of the proposed algorithm.
Fig 9.
The diagram of the proposed algorithm.
Fig 10.
Qualitative fusion results for eight typical infrared and visible image pairs from the TNO database.
(The first row: Infrared images; The second row: Visible images; The third row: FusionGAN; The fourth row: GANMcC; The fifth row: MDLatLRR; The sixth row: NestFuse; The seventh row: RESNetFusion; The eighth row: SEDRFuse; The ninth row: STDFusionNet; The tenth row: OUR).
Fig 11.
Quantitative comparisons in terms of the eight metrics: AG, H, SD, SF, EI, Qab/f, Lab/f, and Nab/f, for ten image pairs from the TNO database.
Fig 12.
Quantitative comparisons in terms of the eight metrics; AG, H, SD, SF, EI, Qab/f, Lab/f, and Nab/f, for the Nato_campsequence from the TNO dataset.
Fig 13.
Qualitative comparison for 20th frame of Trees and runner from the INO dataset.
(a) the infrared and visible images.(b) the fusion results of FusionGAN, GANMcC, MDLatLRR, and NestFuse. (c) the fusion results of ResNetFusion, SEDRFuse, STDFusionNet, and OUR.
Fig 14.
Quantitative comparisons in terms of the eight metrics, AG, H, SD, SF, EI, Q ab/f, Lab/f, and Nab/f, for the Nato_campsequence from the Trees and runner sequence from the INO dataset.
Table 2.
Computational time comparison.
All the experiments were performed on the CPU. (unit: s).