Fig 1.
Schematic illustration of our proposed FDNet through the comparsion with others popular algorithms.
From left to right: the infrared image, the visible image, the fusion results of the CNN [18], the Deeplearning approach [19], the ResNet50 approsch [20], and our proposed FDNet.
Fig 2.
Block diagram of attention mechanism.
Fig 3.
Depthwise separable convolution process.
Fig 4.
General framework diagram.
Fig 5.
The structure diagram of multi-scale feature extraction map.
Table 1.
Depthwise separable process parameter settings.
Fig 6.
I-CBAM overall structure diagram.
Fig 7.
Decomposition network framework.
Fig 8.
Schematic diagram of adaptive weight block.
Fig 9.
Module ablation experimental results.
Table 2.
Objective evaluation results of ablation experiments.
Fig 10.
Decomposition network ablation experiment.
Table 3.
Objective evaluation results of ablation experiments.
Fig 11.
Intensity loss ablation experiment.
Fig 12.
Gradient loss ablation experiment.
Table 4.
Objective evaluation results of ablation experiments.
Fig 13.
Nato camp fusion results.
Fig 14.
Helicopter fusion results.
Fig 15.
Marne-04 fusion results.
Fig 16.
Movie-01 fusion results.
Fig 17.
Movie-18 fusion results.
Fig 18.
Bench fusion results.
Table 5.
The quality evaluation results of the EN.
Table 6.
The quality evaluation results of the AG.
Table 7.
The quality evaluation results of the SD.
Table 8.
The quality evaluation results of the SF.
Table 9.
The quality evaluation results of the MI.
Table 10.
The quality evaluation results of the VIFF.
Table 11.
The quality evaluation results of the SNR.
Table 12.
The quality evaluation results of the CC.