Study Of CT Image Reconstruction Algorithm Based On High Order Total Variation

As a traditional imaging technique,Computed Tomography(CT)is widely used in nondestructive testing.In recent years,low-dose CT has become a focus of attention,which is mainly realized by sparse reconstruction algorithm.At present,most commercial CT still adopts the analytical method as the main algorithm,which produces block artifacts when the sparse projection data is used for reconstruction,and the block artifacts directly affect the quality of the image.With the introduction of compressed sensing theory,the traditional total variation(TV)minimization algorithm can accurately reconstruct images from sparse data.The total variation(TV)minimization algorithm has been used in practical breast CT and has shown good reconstruction performance.However,sometimes it may lead to block artifacts.Recently,researches in the field of image processing show that the high order total variation(HOTV)can effectively suppress that artifacts.However,the researches of HOTV reconstruction algorithm have not been carried out deeply and extensively in image reconstruction.In view of this,this paper studies the HOTV image reconstruction algorithm,and that mainly include:(1)In this paper,we construct the data fidelity constrained,second order TV minimization model,designed the ASD-POCS algorithm.And the reconstruction algorithm was verified and evaluated by the Shepp-Logan simulation phantom of wavy background,the gray-changing simulation phantom and the chest X-ray image simulation phantom.Experimental results show that compared with TV algorithm,HOTV algorithm can suppress block artifacts effectively and achieve higher accuracy.(2)In order to further solve the parameter adjustment problem in ASD-POCS algorithm,we proposed to use Chambolle-Pock(CP)algorithm framework to solve the HOTV algorithm model.The CP algorithm outstanding advantage is that all the algorithm parameters can be calculated by formula without empirical adjustment.The HOTV-CP algorithm is a better reconstruction algorithm than the TV-CP algorithm in image reconstruction if the piecewise constant feature of the reconstructed image is not so obvious but the grayscale fluctuation feature is prominent.This paper starts from the principle of CT imaging,aims at improving the accuracy of reconstruction algorithm,explores the HOTV reconstruction model and its solution algorithm,and systematically evaluates its reconstruction performance.The results of this paper have certain theoretical significance and practical value to promote theapplication of HOTV algorithm in CT reconstruction.