Research On Blocked Compressed Sensing Ghost Imaging Method Based On Structure Group

Ghost imaging,also known as correlation imaging,is a new imaging method that uses the fluctuation correlation of the light field to obtain the image information of the object to be measured.How to reconstruct highquality images under low sampling conditions is a key problem in ghost imaging research.To solve this problem,compressed sensing ghost imaging emerges as the times require.Compressed sensing technology utilizes the sparsity of the signal to achieve non-adaptive signal coding at a sampling rate much lower than the Nyquist sampling rate,reducing the time and space complexity of the sampling process,and can completely restore image information.Reduce the number of samples and improve the imaging quality of the imaging system.However,the sparse basis of traditional compressed sensing associative imaging lacks adaptability and cannot decompose the original image sufficiently sparsely,resulting in the inaccurate effect of reconstructed images.This paper presents a novel associative imaging scheme that combines an experimental architecture for computational associative imaging and a compressive sensing reconstruction algorithm based on sparse representations of structural groups.We take the structure group composed of similar image patches as the basic unit of sparse representation,and each structure group has its own adaptive dictionary.This method can make good use of the prior knowledge of the image,fully characterize the local correlation and non-local self-similarity of the image,and solve the more accurate sparse coding.At the same time,the idea of blocking is added in the experiment to reduce the search domain of the structure group and further reduce the computational complexity.This paper further studies and discusses the performance of the proposed algorithm.The simulation and experimental results show that,compared with the traditional compressive sensing associative imaging scheme,the block compressive sensing associative imaging scheme based on the structure group can effectively improve the weight image quality.Even in the presence of noise interference,the scheme presented in this paper still has better imaging results.At the end of the article,the influence of the parameters of the algorithm on the reconstruction results of the proposed scheme is discussed,and the feasibility of the proposed scheme in practical application is demonstrated.