CPC G06T 5/002 (2013.01) [G06N 20/00 (2019.01); G06T 2207/10081 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/30004 (2013.01)] | 16 Claims |
1. A method for noise reduction in an image, comprising:
providing an input image, the input image being a reconstructed computed tomography (CT) image;
de-noising the input image and producing a de-noised input image; and
adapting a noise texture of the de-noised input image via an adaptation method to produce an adapted image, the adaptation method including replacing the noise texture of the de-noised input image by a noise texture of the adapted image via an algorithm which adapts the de-noised input image based on an adaptation function, the noise texture of the de-noised input image being hyper-Gaussian, the noise texture of the adapted image being Gaussian,
wherein a noise amplitude of the de-noised input image is retained during the adaptation method.
|