US 12,469,116 B2
Method for training image processing model, and method for generating high dynamic range image
Mengdi Sun, Beijing (CN); and Guannan Chen, Beijing (CN)
Assigned to BEIJING BOE TECHNOLOGY DEVELOPMENT CO., LTD., Beijing (CN)
Appl. No. 17/910,020
Filed by BOE TECHNOLOGY GROUP CO., LTD., Beijing (CN)
PCT Filed Sep. 17, 2021, PCT No. PCT/CN2021/119180
§ 371(c)(1), (2) Date Sep. 8, 2022,
PCT Pub. No. WO2023/039863, PCT Pub. Date Mar. 23, 2023.
Prior Publication US 2024/0202891 A1, Jun. 20, 2024
Int. Cl. G06T 5/92 (2024.01); G06T 5/50 (2006.01); G06T 5/60 (2024.01)
CPC G06T 5/92 (2024.01) [G06T 5/50 (2013.01); G06T 5/60 (2024.01); G06T 2207/20081 (2013.01); G06T 2207/20208 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for training an image processing model, comprising:
inputting a low dynamic range image into a first initial image processing model, so that a high dynamic range reconstruction processing is performed on the low dynamic range image to generate a first high dynamic range image;
inputting the low dynamic range image into a second initial image processing model to generate a first coefficient;
generating a second high dynamic range image according to the first high dynamic range image and the first coefficient;
generating a loss function according to data pairs of the second high dynamic range image and a real high dynamic range image, wherein the real high dynamic range image corresponds to the low dynamic range image; and
training the first initial image processing model and the second initial image processing model with the loss function, so that a trained first target image processing model and a trained second target image processing model are obtained,
wherein the first coefficient is a weight coefficient generated by the second initial image processing model performing a structural feature analysis on the low dynamic range image.
 
14. An electronic device comprising: a processor, a memory, and a computer program stored on the memory and executable on the processor, wherein
the processor is configured to input a low dynamic range image into a first initial image processing model, so that a high dynamic range reconstruction processing is performed on the low dynamic range image to generate a first high dynamic range image;
the processor is further configured to input the low dynamic range image into a second initial image processing model to generate a first coefficient;
the processor is further configured to generate a second high dynamic range image according to the first high dynamic range image and the first coefficient;
the processor is further configured to generate a loss function according to data pairs of the second high dynamic range image and a real high dynamic range image; wherein the real high dynamic range image is a real high dynamic range image corresponding to the low dynamic range image;
the processor is further configured to train the first initial image processing model and the second initial image processing model with the loss function, so that a trained first target image processing model and a trained second target image processing model are obtained; and
the memory is configured to store data of the first high dynamic range image, the first coefficient, and data of the second high dynamic range image,
wherein the first coefficient is a weight coefficient generated by the second initial image processing model performing a structural feature analysis on the low dynamic range image.
 
20. A method for training an image processing model, comprising:
inputting a low dynamic range image into a first initial image processing model, so that a high dynamic range reconstruction processing is performed on the low dynamic range image to generate a first high dynamic range image;
inputting the low dynamic range image into a second initial image processing model to generate a first coefficient;
generating a second high dynamic range image according to the first high dynamic range image and the first coefficient;
generating a loss function according to data pairs of the second high dynamic range image and a real high dynamic range image, wherein the real high dynamic range image corresponds to the low dynamic range image; and
training the first initial image processing model and the second initial image processing model with the loss function, so that a trained first target image processing model and a trained second target image processing model are obtained;
wherein generating the second high dynamic range image according to the first high dynamic range image and the first coefficient includes:
multiplying the first high dynamic range image by the first coefficient to generate the second high dynamic range image.