Geometric Deviation Correction Method For Orthogonal Translational CL System

Orthogonal Translational Computed Laminography(OTCL)is a kind of detection technology for plate-like components,which scans the detected object from two orthogonal directions in a plane parallel with the plane of component to obtain information and reconstruct images.This method can realize high-resolution imaging of the detected object in two orthogonal directions,and has the advantages of easy implementation,high resolution and good imaging quality.OTCL requires that the system must satisfy an ideal geometric relationship,that is,the rotation angle of platform is equals to 90°,and the rotation center is located on the line through focal point of X-ray source and center of detector.The two geometric parameter deviations of actual system will lead to artifacts in CL images,and reduce the quality of images.In this paper,the geometric parameter deviations correction of the OTCL system was studied.The main research work is as follows:Firstly,the difference between the ideal OTCT system and the actual OTCT system as well as the influence of geometric parameter deviations on the projection coordinates of points was analyzed.What’s more,the influence of different geometric parameter deviations on the reconstruction results was simulated.Secondly,a calibration correction method of geometrical parameter deviations based on the point model was proposed.Based on the known calibration model,we theoretically analyzed the geometric relationship between feature points and projection data.The equations for solving the geometric parameters of system were obtained to calibrate geometric parameters.Simulation verified the feasibility and correctness of the point phantom.Thirdly,on the basis of obtaining the geometric parameters of system,the deviations of platform were equivalent to the translation and rotation of X-ray source and detector.A modified simultaneous iterative reconstruction technique(SIRT)algorithm was used to calibrate the geometric errors to reconstruct images,which can reduce aliasing and artifacts,and improve the quality of reconstructed results.Simulation shows that the point model can correct geometric parameter deviations effectively.Fourthly,using the actual OTCL experimental system,the actual scanning experiments were carried out.In practice,the method presented in this paper is feasible to calibrate a single deviation or several deviations.Because there are two kinds of deviations in the actual system,only correcting one kind of deviation can’t completely eliminate image dislocation and artifacts.After using the method in this paper to comprehensively correct various deviations,the misalignment and artifacts in reconstructed images were reduced,and the image’s quality was improved significantly,which meets the practical application requirements.This paper starts from the principle of OTCL imaging,aims at correcting geometric parameter deviations,explores the calibration and correction method of geometric deviations,and carries out simulation and imaging experiment research.The research results have certain theoretical significance and practical value to promote the image’s quality of OTCL system reconstruction.