Iterative Filtered Anti-Correlation Imaging

Correlation imaging,also known as ghost imaging(GI),quantum imaging or two-photon imaging,has become a hot research direction in the field of quantum optical imaging in recent decades due to its unique imaging mode.Compared with other imaging techniques such as traditional imaging,correlation imaging has many practical application values and advantages.For example,its imaging is non-local;lensless imaging can be achieved,and the light source can be more selective;even in the presence and influence of noise turbulence or scattering medium,high resolution images can be obtained.In the development of correlation imaging,most research directions and contents focus on improving the imaging speed,signal-to-noise ratio(SNR),resolution and other aspects,but there are few anti-correlation phenomena involving classical light fields.In the early stage,through threshold regulation of the intensity of the reference light field and the signal light field in the experimental system,our group observed the spatial anti-correlation phenomenon of photons for the first time in the field of correlation imaging,and then proposed the iterative filtered ghost imaging scheme,which significantly improved the SNR of the reconstructed images.This article will work in the early stage of the team work conducted on the basis of the following two aspects: the first on the one hand,on the strength of the correlation imaging experimental scheme on the basis of the optimization and improvement,combined with iterative method,through to the threshold selection and regulation,physical experiments,and this experiment used threshold selection scheme is summarized.The experimental results show that the image can be restored well by selecting and adjusting the two threshold values;compared with traditional ghost imaging,the imaging quality is improved obviously.In the second aspect,based on the experimental scheme of iterative filtered ghost imaging(FGI)and combined with compressed sensing(CS),an experimental scheme of iterative filtered ghost imaging with compressed sensing is proposed.Experimental results show that the proposed scheme can reduce the sampling number can be reduced to less than 1% of the Nyquist limit compared with the traditional ghost imaging scheme and the correlation imaging experimental scheme with iterative filtered ghost imaging.In order to further explore the quality of reconstructed images,the mean square error(MSE),peak signal-to-noise ratio(PSNR)and structural similarity(SSIM)of reconstructed images were investigated by comparing with the iterative filtered ghost imaging scheme.The results show that with the increase of the number of iterations,the three indexes to measure the quality of reconstructed images are gradually improved,which also indicates that the quality of reconstructed images is improved.