Method and apparatus for reducing artifacts in computed tomography (CT) image reconstruction

1. A method for reducing artifacts in Computed Tomography (CT) image reconstruction, the method comprising:

acquiring an original reconstructed image;

conducting total variation processing for the original reconstructed image, to generate a total variation reconstructed image;

conducting initial metal artifact reduction processing for the original reconstructed image, to generate an initial metal artifact reduction reconstructed image;

generating a weighted image based on the total variation reconstructed image and the initial metal artifact reduction reconstructed image, wherein the weighted image reflects that the original reconstructed image contains white-band artifacts; and

combining a portion of the original reconstructed image and a portion of the initial metal artifact reduction reconstructed image through the weighted image to generate a final image, wherein the final image does not contain white-band artifact.

2. The method for reducing artifacts in Computed Tomography image reconstruction according to claim 1 , wherein the generating of a weighted image based on the total variation reconstructed image and the initial metal artifact reduction reconstructed image further comprises:

subtracting the initial metal artifact reduction reconstructed image from a matrix of the total variation reconstruction image to generate an initial weighted image matrix;

applying Gaussian blur processing to the initial weighted image matrix, to generate a Gaussian weighted image matrix;

applying normalization processing to the Gaussian weighted image matrix, to generate the weighted image.

3. The method for reducing artifacts in Computed Tomography image reconstruction according to claim 2 , wherein the normalization processing is to divide a Gaussian weighted image matrix by a maximum of CT values of the Gaussian weighted image.

4. The method for reducing artifacts in Computed Tomography image reconstruction according to claim 1 , further comprising:

generating a final image by multipling the weighted image by a matrix of the initial metal artifact reduction reconstructed image, and

adding a product of a result of 1 minus the weighted image matrix and a matrix of the original reconstructed image.

5. The method for reducing artifacts in Computed Tomography image reconstruction according to claim 1 , wherein the initial metal artifact reduction processing comprises:

determining a metal projection area in the original reconstructed image;

conducting interpolation correction processing for projection values in the metal projection area; and

conducting restoration and reconstruction processing for the original reconstructed image.

6. An apparatus for reducing artifacts in Computed Tomography image reconstruction, the apparatus comprising:

an acquisition device, configured to acquire an original reconstructed image;

a total variation processor, configured to:

conduct a total variation processing for the acquired original reconstructed image, and

generate a total variation reconstructed image based on the total variation processing

an initial metal artifact reduction processor configured to:

conduct an initial metal artifact reduction processing for the original reconstructed image; and

generate an initial metal artifact reduction reconstructed image based on the initial meal artifact reduction processing;

a weighted image generator configured to:

generate a weighted image based on the total variation reconstructed image and the initial metal artifact reduction reconstructed image, wherein the weighted image reflects that the original reconstructed image contains white-band artifacts; and

a final image generator configured to:

combine a portion of the original reconstructed image and a portion of the initial metal artifact reduction reconstructed image through the weighted image to generate a final image, wherein the final image docs not contain white-band artifact.

7. The apparatus for reducing artifacts in Computed Tomography image reconstruction according to claim 6 , wherein the weighted image generator further comprises:

an initial weighted computing device configured to:

substract the initial metal artifact reduction reconstructed image from a matrix of the total variation reconstruction image to generate an initial weighted image matrix;

a Gaussian blur processor, configured to apply a Gaussian blur processing to the initial weighted image matrix, to generate a Gaussian weighted image matrix; and

a normalization processor, configured to:

conduct a normalization processing to the Gaussian weighted image matrix, to generate the weighted image.

8. The apparatus for reducing artifacts in Computed Tomography image reconstruction according to claim 7 , wherein the normalization processing comprises:

dividing the Gaussian weighted image matrix by the maximum of CT values of the Gaussian weighted image.

9. The apparatus for reducing artifacts in Computed Tomography image reconstruction according to claim 6 , wherein the final image generator is further configured to:

multiply the weighted image by a matrix of the initial metal artifact reduction reconstructed image, and

add a product of a result of 1 minus the weighted image matrix and a matrix of the original reconstructed image, thereby generating the final image.

10. The apparatus for reducing artifacts in Computed Tomography image reconstruction according to claim 6 , wherein the initial metal artifact reduction processor is further configured to:

determine a metal projection area in the original reconstructed image;

conduct interpolation correction processing for projection values in the metal projection area; and

conduct restoration and reconstruction processing for the original reconstructed image.

11. A Computed Tomography equipment, comprising:

a scanner configured to scan an object for obtaining raw data, so as to generate an original reconstructed image;

a processor, operably coupled to the scanner, and programmable to execute the steps of:

acquiring the original reconstructed image;

conducting a total variation processing on the original reconstructed image, to generate a total variation reconstructed image;

conducting an initial metal artifact reduction processing on the original reconstructed image, to generate an initial metal artifact reduction reconstructed image;

generating a weighted image based on the total variation reconstructed image and the initial metal artifact reduction reconstructed image, wherein the weighted image reflects that the original reconstructed image contains white-band artifacts; and

combining a portion of the original reconstructed image and a portion of the initial metal artifact reduction reconstructed image through the weighted image to generate final image, wherein the final image does not contain white-band artifact.