Advanced Photonics Nexus, Volume. 3, Issue 2, 026010(2024)
Physics-constrained deep-inverse point spread function model: toward non-line-of-sight imaging reconstruction
Fig. 1. The NLOS system and reconstruction principle. (a) A confocal NLOS imaging system with a CMOS camera to capture the image. (b) The imaging equation in an optical system with PSF and (c) propagation process from object to image in the NLOS system.
Fig. 2. Light path in the NLOS system. (a) Wavefront propagation process of diffuse reflection and (b) definition of diffuse reflection parameters.
Fig. 3. Flowchart of PCIN algorithm for NLOS imaging reconstruction. The speckle image captured by the camera is put into CNN, and PCIN iteratively updates the parameters in CNN using the loss function constructed by the speckle image and forward physical model. The optimized parameters are utilized to obtain a high-quality reconstructed image.
Fig. 4. Back and front of the experimental scene. Light passes from the laser, to the collimator, to the wall, to the hidden object, and finally to the camera.
Fig. 5. Comparison of the reconstructed images of various exposure time from the proposed PCIN method. (a) Speckle images of different exposure time captured by the camera. (b) Ground truth. (c) Reconstructed images of different exposure time.
Fig. 6. Comparison of the reconstructed images of various exposure time from the proposed PCIN method. (a) Speckle images of different exposure time captured by the camera. (b) Ground truth. (c) Reconstructed images of different exposure time.
Fig. 7. Comparison of the reconstructed cartoon images and Chinese characters of various exposure time from the proposed PCIN method. (a) Speckle images of different exposure time captured by the camera. (b) Ground truth. (c) Reconstructed images of different exposure time.
Fig. 8. Comparison of the reconstructed images of convex, concave, and wavy walls.
Fig. 9. Comparison of the reconstructed images of PR, CNN, and PCIN methods at 40 ms exposure time.
Fig. 10. Comparison of the reconstructed images of PR, CNN, and PCIN methods after a 10 deg change in image plane inclination.
Fig. 11. Comparison of the reconstructed images of PR, CNN, and PCIN methods at 20 ms exposure time.
Get Citation
Copy Citation Text
Su Wu, Chan Huang, Jing Lin, Tao Wang, Shanshan Zheng, Haisheng Feng, Lei Yu. Physics-constrained deep-inverse point spread function model: toward non-line-of-sight imaging reconstruction[J]. Advanced Photonics Nexus, 2024, 3(2): 026010
Category: Research Articles
Received: Nov. 8, 2023
Accepted: Feb. 27, 2024
Published Online: Mar. 19, 2024
The Author Email: Yu Lei (yulei@aiofm.ac.cn)