The Key Technology Research For Improving The Reconstruction Quality Of Ghost Imaging With Pseudo-thermal Light

Correlated imaging,also known as ghost imaging(GI),has become a research hotspot in the field of quantum imaging and classical optical imaging in the recent years due to its advantages of non-localized imaging,lensless imaging,anti-turbulence interference and so on.Compared with the conventional imaging technologies,GI can obtain the object information through the second-order correlation of two light beams.In some cases,the object information can even be obtained by only one bucket detector without spatial resolution.Therefore,ghost imaging has attained wide attention in the fields of remote sensing,national defense,biomedicine,and optical information encryption.However,the low signal-to-noise ratio(SNR)and the huge number of measurements are still the key issues which limit its application in practice and urgent to be solved currently.To solve these problems,we have proposed some feasible research schemes to improve the reconstruction quality of GI in the directions of improving the SNR and reducing the number of measurements.The feasibility and effectiveness of the proposed schemes are demonstrated by the method of mathematical modeling,numerical simulation,and experimental verification.The main research contents and innovations of this paper are summarized as follows:Firstly,a novel iterative denoising ghost imaging scheme based on the adaptive threshold algorithm is proposed.The proposed scheme uses an iterative threshold to calculate the ideal threshold associated with the correlated noise in GI.It should be noted that the threshold can be obtained without needing the prior knowledge of the object.Afterward,we can construct the correlated noise in our denoising model.Besides,we apply the adaptive threshold method to binarize the initial value of each iteration to make it closer to the original object's transmission coefficient.After three iterations,we can make the reconstruction quality much better.The considerable simulations and experimental results show that the proposed scheme has the ability to remove the background noise of the reconstruction image.Especially,the visual effects and PSNR values are improved in comparisons with the existing GI methods.This method provides a novel technical solution for suppressing the correlated noise in GI and can be employed for improving GI quality.Secondly,a novel mask-based denoising scheme for GI is proposed.After analyzing the advantages and disadvantages of scalar-matrix-structured ghost imaging(SMGI),and the iterative threshold is employed to optimize this method.The proposed method can be named as iterative threshold ghost imaging(ITGI).Afterward,the mask-based denoising ghost imaging(MDGI)is proposed based on ITGI.A mask is designed through the maximum between-class variance(OTSU)and morphological dilation method,and to remove the correlated noise in GI.The simulation and experimental results demonstrate MDGI have the advantage of improving the reconstruction quality of ghost imaging.The proposed method provides a novel solution for effectively improving the reconstruction quality of ghost imaging and reducing the number of measurements.Finally,we proposed a novel image enhancement scheme for GI by fuzzy c-means(FCM)clustering method term as FGI.After analyzing the influence of the light intensity average on ghost imaging visibility,the reference light total intensity and bucket intensity are counted and classified by the FCM clustering method.Referring to the clustering results,the corresponding frames are selected to participate in reconstruction in GI.The simulation and experiment results demonstrate that this proposed scheme can effectively improve the reconstruction quality of GI with reduced measurement data.In addition,the proposed method provides new ideas and technical means for the selection of measurement data in GI.