Scatter correction algorithm for cone-beam computed tomography based on convolution and total variation model

In order to solve the problem of scattering artifacts in the computed tomography images, a modified Poisson distribution function is presented and used as the scatter model. A set of scatter basis images are obtained from scatter estimations with different parameter values in the model. Then the corrected image is obtained by adding the basis images to the original image. The coefficients for the linear combination are determined by minimizing the total variation of the combined images. The method requires low computational effort and allows flexible application to different scanned objects. Experiment results indicate that cupping and dark bands artifacts due to scattering are effectively depressed by the proposed method， and the image contrast is enhanced.