Active single-pixel imaging method for strongly scattering media based on conditional generative adversarial networks

By designing a single-pixel optical imaging system based on Hadamard transform and an AMSPI-LSCGAN model of the least squares condition generation adversarial network, the imaging quality problems in low sampling rate and strong scattering environments are solved, and clear imaging under high turbidity conditions is achieved, which is suitable for single-pixel scattering imaging systems.

CN115601621BActive Publication Date: 2025-09-02HUBEI UNIV OF TECH
2 Cites 0 Cited by

Patent Information

Application Number
CN202211264651.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-10-17
Publication Date
2025-09-02
Estimated Expiration
2042-10-17

AI Technical Summary

Technical Problem

The existing single-pixel imaging methods have poor imaging quality in low sampling rates and strong scattering environments, especially when the turbidity is greater than 50 NTU, the target reconstruction takes a long time and the imaging quality is greatly affected by the scattering medium.

Method used

A single pixel optical imaging system based on Hadamard transformation is designed, and an AMSPI-LSCGAN model based on the least squares condition generated adversarial network is constructed. Combining the least squares loss, content loss and average structural similarity loss as joint loss functions, image reconstruction is carried out through a deep convolutional network, enhancing the imaging robustness under strong scattering conditions.

Benefits of technology

The image reconstruction quality is significantly improved at low sampling rate, and the object profile can be distinguished when the water turbidity reaches 144 NTU. It has high robustness and versatility. It is suitable for single-pixel scattering imaging systems and improves the target imaging quality in a strong scattering environment.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The present invention discloses an active single-pixel imaging method for strongly scattering media based on a conditional generative adversarial network. The method comprises the following steps: designing a single-pixel optical imaging system based on a Hadamard transform to obtain one-dimensional detection signal values ​​of a test target under a series of different scattering conditions and sampling rates; shaping the sampled one-dimensional detection signal values ​​under different scattering conditions and sampling rates into two-dimensional signal images through data averaging preprocessing, which are used as test set images; constructing a deep convolutional conditional generative adversarial network (AMSPI-LSCGAN) based on a least squares loss function model with a compression-excitation block and a residual block; combining the least squares loss with a content loss and an average structural similarity loss as the loss function of the AMSPI-LSCGAN network for training, thereby avoiding training collapse and obtaining reconstructed images with higher fidelity; and inputting the test set images into the trained AMSPI-LSCGAN network to reconstruct the test target.
Need to check novelty before this filing date? Find Prior Art