Design And Optimization Of The Deblurring Computational Ghost Imaging For Measurement Matrix

As a new imaging method,computational ghost imaging has the characteristics of strong anti-interference that traditional imaging does not have.In recent years,it has played a unique role in underwater imaging and remote sensing imaging.In the computational ghost imaging experiment in the scattering medium,such as underwater computational ghost imaging,the measurement matrix is very important to the imaging efficiency and the quality of the imaging results.This paper focuses on the computational ghost imaging with different measurement matrices in the presence of scattering media.The main contents of the work are as follows:1.By comparing different measurement matrices for computational ghost imaging applications in scattering media,the imaging speed and imaging quality of different measurement matrices are studied.By selecting the optimal structured light measurement matrix scheme,the difficulty of data storage and calculation in the computational ghost imaging is greatly reduced,the corresponding imaging efficiency is improved,and the imaging effect of the computational ghost imaging in the scattering medium is improved.2.A simulation scheme for computational ghost imaging in scattering media is proposed.Through theoretical analysis of common image degradation causes and experimental imaging results,the Gaussian fuzzy simulation is selected computational ghost imaging results under different measurement matrices,which greatly facilitates the analysis and comparison of the results of different measurement matrices after the quantization of the scattering medium,so as to optimize design.3.A scheme for computational ghost imaging in scattering media to remove the influence of scattering media is proposed.Analyzing the cause of noise in the computational ghost imaging of scattering media.Starting from multiple angles,optimizing the design of the measurement matrix in the emission path,and removing the blur caused by the scattering media in the imaging results with the generation countermeasure network and achieving good results.In this study,based on the theory of scattering medium,measurement matrices and computational ghost imaging,partial Hadamard matrix is selected as the measurement matrix,Gaussian blur is used in the emission path to simulate the scattering medium,and a reasonable computational ghost imaging model of the scattering medium is designed.Which greatly facilitates the simulation experiment of computational ghost imaging under the scattering medium.At the same time,the optimization of the computational ghost imaging under the scattering medium is studied,and the imaging effect of the computational ghost imaging under the scattering medium is further improved.