CPC G06T 5/002 (2013.01) [G06N 3/045 (2023.01); G06N 3/088 (2013.01); G06T 3/4046 (2013.01); G06T 3/4053 (2013.01); G06T 5/20 (2013.01); G06T 5/50 (2013.01); G06T 2207/10024 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01)] | 18 Claims |
1. An image denoising method, comprising:
acquiring a first data set and a second data set, wherein the first data set comprises a plurality of first images without noise, the second data set comprises a plurality of second images with real noise, contents of each first image and each second image are different;
training, by using the first data set and the second data set, a first network to obtain a noise generation model;
inputting the first image into the noise generation model, and outputting a third image with simulated noise, wherein a plurality of third images forms a third data set; and
training, by using the first data set and the third data set, an image denoising network to obtain an image denoising model;
wherein the image denoising model is configured to convert an original image with noise into an output image without noise;
wherein prior to the inputting the first image into the noise generation model, the method further comprises:
converting the first image into a first training sample image;
the inputting the first image into the noise generation model and outputting the third image with simulated noise, comprises:
inputting the first training sample image into the noise generation model and outputting the third image, wherein a resolution of the first image is larger than a resolution of the first training sample image, and a resolution of the third image is the same as the resolution of the first training sample image;
wherein the image denoising model is further configured to convert the original image with noise and having a first resolution into the output image without noise and having a second resolution, and the first resolution is smaller than the second resolution.
|