Sparse Angle Reconstruction Method and Device of CT Image Based on Total Curvature Joint Total Variation

Abstract

The invention relates to the field of CT (Computed Tomography) image reconstruction, and discloses a CT image sparse-view reconstruction method based on total curvature joint total variation. The CT image sparse-view reconstruction method comprises the steps of setting a weighting factor; building a total curvature joint total variation minimization model; deducing a final CT image reconstruction algorithm by using an alternating direction method; and carrying out the final CT image reconstruction algorithm, and realizing and acquiring a final reconstruction result. The invention further discloses a CT image sparse-view reconstruction device based on total curvature joint total variation, which comprises a weighting factor setting module, a total curvature joint total variation minimization model building model, a final CT image reconstruction algorithm deducing module and a final reconstruction result obtaining module. The CT image sparse-view reconstruction method and device are high in efficiency, can adapt to less acquisition data and improve the reconstructed image quality.

Description

technical field [0001] The invention relates to the field of CT image reconstruction, in particular to a CT image sparse angle reconstruction method and device based on total curvature combined with total variation. Background technique [0002] As a modern imaging technology, computed tomography (Computed Tomography, CT) has been widely used in medicine, industry and other fields. However, on the one hand, the harm of large doses of ionizing radiation to the human body has been medically proven; Data are obtained at fewer projection angles, which all belong to the sparse-view problem (Sparse-view Problem). Improving the quality of CT image reconstruction under sparse angle scanning has important theoretical research and engineering practical significance. How to design a high-precision method for CT image reconstruction under sparse angle scanning is also a hot and difficult issue in research. [0003] The sparse angle problem of cone beam CT is essentially an inverse pro...

AI Technical Summary

Problems solved by technology

Although theoretically these methods have certain advantages over total variational methods, in actual calculations, curvature-based variational models often encounter challenges.

This is because the second-order variational problem needs to obtain a result by solving a fourth-order partial differential equation, and the fourth-order parabolic partial differential equation does not satisfy the maximum principle, and is often unstable in numerical calculations

Images