Research On The Spatial Scale Effect Of Ghost Imaging

Ghost imaging is a new imaging method different from traditional imaging.Compared with traditional imaging,the new ghost imaging technology has the following characteristics.Firstly,it is not difficult to meet the requirements of optical source in the imaging system.At the same time,it has high sensitivity to capture weak light,high spatial resolution and strong anti-noise performance.Most importantly,objects and images can be separated in the ghost imaging system.However,the current ghost imaging technology still has limitations.For example,in order to obtain sufficient imaging data,the detector needs repeated sampling,and a large number of sampling work affects the speed of image formation.At the same time,higher requirements are put forward for the hardware configuration of data storage space.In order to solve this problem,this paper proposes the idea of applying pooling in ghost imaging.The aggregation of image pixels can not only reduce the data dimension of the image,thus greatly reducing the data volume of the imaging system and improving the speed of image formation,but also reduce the over-fitting problem and optimize the imaging results.In this paper,the characteristics of the point spread function are studied and simulated.After pooling the image pixel resolution is compressed in ghost imaging,which makes the pixel resolution of the image close to the real spatial resolution in the imaging system.Correspondingly,the low-resolution reconstruction technique is studied,and appropriate interpolation algorithm is selected to reconstruct the compressed image with the same pixel number as the original target object.Do simulation experiments and analyze the results to explore spatial scale effects of ghost imaging.The experimental results show that the influence of repeated sampling and measurement on the imaging quality of ghost imaging can be divided into two sub-processes.One is to increase the image resolution,and the other is to improve the precision of the single pixel through repeated measurement and reduce the error caused in the process of image detection and sampling.The proportion of necessary samples required for the two sub-processes increases as the complexity of the target object increases.