Correlated Imaging And Its Optimization For Moving Objects And Sliced Images

Correlated imaging,also known as ghost imaging,is a novel optical imaging technique.Unlike traditional optical imaging,a correlated imaging system includes two optical paths: a signal path and a reference path.The signal path passes through the object under test and is detected by a bucket detector that does not have spatial resolution,while the reference path,which does not interact with the object,propagates freely and is detected by an array detector with spatial resolution.By correlating the measurements from both paths,it is possible to obtain an image of the object in the reference path without the object itself.Neither path can independently form an image of the object.Correlated imaging has advantages such as non-locality,strong anti-interference ability,high detection sensitivity,and a broader detection bandwidth.However,the application of correlated imaging is limited by the significant time cost required to obtain spatial information.This paper mainly studies correlated imaging and its optimization for moving objects and sliced images.The specific contents and achievements are as follows:(1)Correlated imaging methods for objects with fast periodic moving/state-changed objects are proposed.Firstly,a Hadamard-based correlated imaging method for fast periodic moving/state-changed objects is proposed.By designing speckle patterns,the problem of multi-frame image imaging of objects with fast periodic moving/state-changed is transformed into correlated imaging of Hadamard speckle pattern to a static large image composed of image frames of various states of the object.This method allows imaging of fast periodic moving/state-changed objects using a slow bucket detector.Simulation and experimental results demonstrate the effectiveness of this method for imaging fast periodic moving/state-changed objects.Subsequently,a fast periodic moving/state-changed object imaging method based on compressed correlated imaging is proposed.This method improves the reconstruction quality of the object’s multi-frame images and reduces the number of measurements of the bucket detector.It provides a new approach for studying correlated imaging of fast periodic moving/state-changed objects.(2)Temporal and spatial correlated imaging methods for moving objects are proposed.Firstly,a method for temporal and spatial correlated imaging of moving objects is proposed for multi-frame images.The first frame image of the moving object is obtained through spatial correlated imaging,and the time signal encoding the relative position information of the moving object between frames is obtained by temporal correlated imaging.By combining the first frame and the encoded time signal,the remaining frames of the moving object can be obtained by shifting.Simulation results show that compared with existing correlated imaging methods for moving objects,this method can effectively reduce the data volume and reconstruction time of correlated imaging of multi-frame images of moving objects.Subsequently,a temporal and spatial correlated imaging method for real-time moving objects is proposed,which combines real-time position information acquisition of moving objects and a multi-resolution speckle patterns projection method.Simulation results show that compared to the traditional method,this approach can achieve high-quality imaging of the target of interest in a complex background,with the same number of measurements using bucket detectors.(3)A correlated imaging method for sliced images is proposed.This method combines the Radon transform and a structured light source array that periodically rotates around the target object.The structured speckle patterns are projected onto the target object,and the transmitted light is received by a bucket detector.Using the pre-set speckle information and the measurement results of the bucket detector,the sinogram of the sliced images can be obtained.Finally,the filtered back projection algorithm reconstructs the sliced images.The proposed methods have conducted research on the reconstruction problem of single-layer and multi-layer sliced images of objects,and their effectiveness has been demonstrated through simulation experiments.(4)The optimization methods for correlated imaging are proposed.Firstly,a random speckle orthogonal optimization computational correlated imaging method based on speckle modulation is proposed.By utilizing the properties of the real symmetric matrix formed by the random speckle patterns and combining them with a spatial mapping matrix,the original random speckle patterns are orthogonalized,thereby improving the reconstruction quality of target objects in correlated imaging.In addition,by incorporating the sparsity of object edge information,a compressed correlated imaging edge detection method is proposed to enhance the efficiency and quality of edge detection.Building upon this,a moving object position extraction method based on compressed correlated imaging is further introduced.This method improves the accuracy of extracting the positions of moving objects and simplifies the process of obtaining position information,thereby improving the practical application capabilities of real-time moving objects imaging based on temporal and spatial correlated imaging.