Research on mechanism of joint-coding imaging based on generative adversarial neural network
Abstract
In the previous research of ghost imaging schemes based on joint-coding technology, to ensure the imaging quality, it is often carried out under the premise of full sampling or even super-sampling, which undoubtedly requires a long sampling time. In this paper, a joint coding imaging scheme based on the generative adversarial neural network is proposed. By introducing the generative adversarial neural network, the image is reconstructed under half sampling, which greatly saves the sampling time. At the same time, the joint coding technology is incorporated to ensure the imaging quality of the computational ghost imaging scheme. Comparison through experimental, it can be proved that the imaging quality of this scheme under half sampling can achieve the imaging effect of CSGI algorithm under full sampling, and in the case of different signal-to-noise ratio noise, the reconstruction of our scheme has stronger robustness compared with other schemes. Briefly our scheme provides a new imaging technology for the research in the imaging field, which has good theoretical significance.
- Publication:
-
Optics and Lasers in Engineering
- Pub Date:
- December 2023
- DOI:
- 10.1016/j.optlaseng.2023.107790
- Bibcode:
- 2023OptLE.17107790Y
- Keywords:
-
- Generate adversarial neural network;
- Wavelet transform;
- Computational ghost imaging