Lishun Wang, Zongliang Wu, Yong Zhong, Xin Yuan, "Snapshot spectral compressive imaging reconstruction using convolution and contextual Transformer," Photonics Res. 10, 1848 (2022)

Search by keywords or author
- Photonics Research
- Vol. 10, Issue 8, 1848 (2022)
![Reconstructed real data of Legoman, captured by snapshot SCI systems in Ref. [20]. We show reconstruction results of 12 spectral channels, and compare our proposed method with the latest self-supervised method (PnP-DIP-HSI [23]) and the method based on maximum a posteriori (MAP) estimation (DGSMP algorithm [24]). As can be seen from the purple and green areas in the plot, our method reconstructs a clearer image, the PnP-DIP-HSI method produces some artifacts, and the DGSMP method loses some details.](/richHtml/prj/2022/10/8/1848/img_001.jpg)
Fig. 1. Reconstructed real data of Legoman, captured by snapshot SCI systems in Ref. [20]. We show reconstruction results of 12 spectral channels, and compare our proposed method with the latest self-supervised method (PnP-DIP-HSI [23]) and the method based on maximum a posteriori (MAP) estimation (DGSMP algorithm [24]). As can be seen from the purple and green areas in the plot, our method reconstructs a clearer image, the PnP-DIP-HSI method produces some artifacts, and the DGSMP method loses some details.

Fig. 2. Schematic diagrams of CASSI system.

Fig. 3. Architecture of the proposed GAP-CCoT. (a) GAP-net with N stages; G ( · ) represents the operation of Eq. (6 ), D ( · ) represents a denoiser, and v ( 0 ) = H T g . (b) CCoT-net, the proposed denoising network plugged into GAP algorithm. (c) Convolution branch and Transformer branch; the output is connected with concatenation. (d) Convolution block with channel attention; c represents the output number of convolution channels. (e) Contextual Transformer block. (f) Pixelshuffle algorithm for fast upsampling.

Fig. 4. Reconstruction results of GAP-CCoT and other spectral reconstruction algorithms (λ -net, HSSP, TSA-net, GAP-net, DGSMP, PnP-DIP-HSI) in scene 3 and scene 9. Zoom in for better view.

Fig. 5. Architecture of the proposed Stacked CCoT. The input of the network is H T g , and CCoT-net is the same as in Fig. 3 (b).

Fig. 6. Effect of stage number on SCI reconstruction quality.

Fig. 7. Reconstruction results of GAP-CCoT and other spectral reconstruction algorithms (λ -net, TSA-net, GAP-net, DGSMP, PnP-DIP-HSI) in two real scenes (scene 1 and scene 2).

Fig. 8. Reconstructed frame of our method and other algorithms (GAP-TV, DeSCI, PnP-FFDNet, U-net, BIRNAT, RevSCI) on six benchmark datasets.
|
Table 1. Average PSNR in dB (upper entry in each cell) and SSIM (lower entry in each cell) of Different Algorithms on 10 Synthetic Datasetsa
|
Table 2. Computational Complexity and Average Reconstruction Quality of Several SOTA Algorithms on 10 Synthetic Datasets
|
Table 3. Average PSNR and SSIM Results on 10 Synthetic Data with Different Masks
|
Table 4. Ablation Study: Average PSNR and SSIM Values of Different Algorithms on 10 Synthetic Data
|
Table 5. Computational Complexity and Average Reconstruction Quality of GAP-CCoT on 10 Synthetic Data with Different Stages
|
Table 6. Average PSNR and SSIM Results on 10 Synthetic Data with Different Loss Functions
|
Table 7. Extending Our Method for Video Compressive Sensing: Average PSNR, SSIM, and Running Time per Measurement of Different Algorithms on Six Benchmark Datasets
|
Table 8. Computational Complexity and Average Reconstruction Quality of Several SOTA Algorithms on Six Grayscale Benchmark Datasets

Set citation alerts for the article
Please enter your email address