Research On Computational Ghost Imaging Based On Gold Sequence Coding

Ghost imaging is a novel optical imaging technology.Compared with traditional optical imaging technology,it has the advantages of anti-interference,high sensitivity of low-light imaging,and non-line-of-sight vision.It has many potential applications in biological,security,military and other imaging fields.It has been nearly 30 years since the concept of ghost imaging was proposed,and researchers at home and abroad have carried out a lot of fruitful research work in this field.At present,the practical progress of ghost imaging technology is still in the stage of principle verification.Although the laboratory environment can achieve real-time high-quality ghost imaging,the actual environmental imaging quality still has certain limitations.In the actual complex imaging environment,the imaging quality is not only greatly affected by noise,but also directly related to the measurement matrix used by the imaging system.The measurement matrices commonly used in ghost imaging systems include random matrices and orthogonal matrices.The random matrix has good pseudo-randomness and is more uniformly affected by noise.Its imaging quality is more robust to noise than orthogonal matrices,such as the Hadamard matrix;and the Hadamard matrix as an orthogonal matrix has good orthogonality.Only the number of measurements(full sampling)equal to the number of pixels of the object to be measured is needed to obtain a high-quality reconstructed image.The two types of measurement matrices have different advantages in terms of improving ghost imaging performance.In this paper,a new measurement matrix is ??redesigned by combining the respective characteristics of the above two types of measurement matrices to improve the imaging efficiency and imaging quality of ghost imaging under actual environmental conditions.The main research contents of this article are as follows:(1)Based on the widely used pseudo-random sequence Gold sequence,a Gold matrix suitable for calculating ghost imaging is designed and constructed.This matrix has the advantages of random measurement matrix and orthogonal measurement matrix.(2)Based on the theoretical analysis to achieve the full reconstruction of the perfect sampling calculation of ghost imaging on the basis of the mathematical properties of the measurement matrix,using the feature matrix to analyze the imaging performance of the Gold matrix and Random matrix,Hadamard matrix,Toeplitz matrix.(3)Carry out numerical simulation research on the calculation of ghost imaging based on the Gold matrix.The results show that under ideal conditions,the Gold matrix can achieve perfect reconstruction at full sampling,and the Gold matrix can still achieve better imaging quality under the influence of noise.(4)Carry out numerical simulation and simulation research based on Gold matrix combined with compressed sensing ghost imaging.The results show that using the TVAL3 algorithm to achieve image reconstruction,the imaging quality of the Gold matrix under low compression ratio is superior to the Random matrix and Hadamard matrix,which effectively reduces the sampling rate and achieves high-quality image reconstruction at low sampling numbers.(5)Completing the experimental study of computational ghost imaging based on the Gold matrix in the laboratory environment.The results show that,compared with the Random matrix,Hadamard matrix and Toeplitz matrix,the reconstruction quality of the Gold matrix has obvious advantages.