Haitao Yin, Wei Zhou. Multi-Scale Dilated Convolutional Neural Network Based Multi-Focus Image Fusion Algorithm[J]. Laser & Optoelectronics Progress, 2023, 60(2): 0210003
- Laser & Optoelectronics Progress
- Vol. 60, Issue 2, 0210003 (2023)
Fig. 1. Residual block
Fig. 2. SE attention module
Fig. 3. Architecture of MDF-Net
Fig. 4. Multi-scale dilated block
Fig. 5. Dilated convolutional with different dilation rates.（a）Dilation rate is 1；（b）dilation rate is 2；
Fig. 6. Examples of simulated multi-focus images
Fig. 7. Fused results of “children” image. (a)(b) Source images; (c) NSCT; (d) SR; (e) IMF; (f) MWGF; (g) CNN; (h) DeepFuse; (i) DenseFuse-ADD; (j) DenseFuse-L1; (k) IFCNN-MAX; (l) MDF-Net
Fig. 8. Fusion results of “monkey” image. (a)(b) Source images; (c) NSCT; (d) SR; (e) IMF; (f) MWGF; (g) CNN; (h) DeepFuse; (i) DenseFuse-ADD; (j) DenseFuse-L1; (k) IFCNN-MAX; (l) MDF-Net
Fig. 9. Fusion results of “gymnasium” image. (a)(b) Source images; (c) NSCT; (d) SR; (e) IMF; (f) MWGF; (g) CNN; (h) DeepFuse; (i) DenseFuse-ADD; (j) DenseFuse-L1; (k) IFCNN-MAX; (l) MDF-Net
Fig. 10. Fusion results of “statue” image. (a)(b) Source images; (c) NSCT; (d) SR; (e) IMF; (f) MWGF; (g) CNN; (h) DeepFuse; (i) DenseFuse-ADD; (j) DenseFuse-L1; (k) IFCNN-MAX; (l) MDF-Net
|
Table 1. Indexes values of various fusion algorithms on "gymnasium" image
|
Table 2. Average indexes values of various fusion algorithms on Lytro dataset
|
Table 3. Indexes results of different variants of MDF-Net
|
Table 4. Ablation experiment on number of MDB modules
|
Table 5. Ablation experiment on number of dilated convolution branches in MDB
|
Table 6. Ablation experiment on number of feature channels in MDB
Set citation alerts for the article
Please enter your email address
© Copyright 2018-2021 | Chinese Laser Press. All Rights Reserved 沪ICP备15018463号-20