.png){kind=icon}
.png){kind=icon}
Return
Singular value decomposition compressive ghost imaging based on multiple image prior information
.png){kind=icon}
PRE
AI
DOI：10.1016/j.optlaseng.2024.108471.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Ghost imaging, which utilizes structured illumination to reconstruct two-dimensional information of objects from one-dimensional bucket signals, has garnered significant attention. However, before introducing compressed sensing, how to obtain high fidelity images at low sampling ratios has always been a major challenge. Previous work on compressive ghost imaging lacks the incorporation of prior information into the image reconstruction process, leading to significant distortions in the reconstructed images at low sampling ratios. In this paper, to enhance the imaging quality of ghost imaging at low sampling ratios, we introduce two prior image information regarding smoothness and nonlocal self-similarity based on singular value decomposition compressive ghost imaging and propose a novel compressive ghost imaging. The scientific validity and feasibility of the proposed method are demonstrated both theoretically and experimentally. Furthermore, comparative analyses with other classical algorithms reveal significant advantages of our method in handling image smoothness and nonlocal self- similarity. Even at lower sampling ratios, the proposed method demonstrates impressive reconstruction capabilities, resulting in reconstructed images closely resembling the originals. Both global and local details of the images are effectively reconstructed, enhancing the fidelity of the reconstructed images.
Keywords:
Ghost imaging
Compressed sensing
Singular value decomposition
Journal
IF:
3.7
Papers:
7.1K
Citations:
1.7W