Granted Patent B2

US 12,462,448 · App. 18/279,093 · Granted Nov 4, 2025

Inventors: Bin Xiao (Shenzhen, CN); Yu Wang (Shenzhen, CN); Tao Shao (Shenzhen, CN); Congchao Zhu (Shenzhen, CN)

Assignee: HONOR DEVICE CO., LTD.

G06T11/001G06T7/11G06T7/90G06V10/56G06T2207/10024G06T2207/20081G06T2207/20084

Abstract

Claims (55)

1 . An image processing method, wherein the method comprises:

displaying a first interface, and determining, on the first interface, a first file selected by a user;

displaying a second interface, and displaying a generated lookup table on the second interface; wherein, the lookup table is generated by a preset deep neural network model or a polynomial fitting algorithm;

determining, on the second interface, a target lookup table selected by the user from the lookup table, and performing color adjustment on the first file by using the target lookup table; and

displaying a color-adjusted first file,

wherein the method further comprises:

determining a first image selected by the user and a second image corresponding to the first image, wherein the second image is an image obtained through first color adjustment on the first image; and

generating, based on the first image, the second image, and the deep neural network model, a lookup table corresponding to the first color adjustment; and generating, based on the first image, the second image, and the polynomial fitting algorithm, a lookup table corresponding to the first color adjustment, wherein the lookup table corresponding to the first color adjustment is used for enabling the first file and the second image to have same color adjustment effect during the color adjustment on the first file.

2 . The method according to claim 1 , wherein the lookup table comprises a global lookup table and a local lookup table.

3 . The method according to claim 1 , wherein the generating, based on the first image, the second image, and the deep neural network model, the lookup table corresponding to the first color adjustment comprises:

inputting the first image and the second image to a preset deep neural network model, to obtain a global lookup table output by the deep neural network model.

4 . The method according to claim 1 , wherein the generating, based on the first image, the second image, and the deep neural network model, the lookup table corresponding to the first color adjustment comprises:

segmenting the first image and the second image into N subblocks in a same segmentation manner;

combining a first subblock in the first image and a second subblock in the second image into an input data pair to obtain N input data pairs, wherein a location of the first subblock of the input data pair in the first image is the same as a location of the second subblock in the second image;

inputting the N input data pairs to the deep neural network model to obtain N subblock lookup tables; and

combining the N subblock lookup tables to obtain a local lookup table, wherein N is a positive integer.

5 . The method according to claim 4 , wherein the combining the N subblock lookup tables to obtain a local lookup table comprises:

smoothing each subblock lookup table; and

combining smoothed N subblock lookup tables to obtain the local lookup table.

6 . The method according to claim 5 , wherein the smoothing each subblock lookup table comprises:

determining a weight corresponding to each subblock lookup table;

obtaining a first product of an output value of an N th subblock lookup table and a weight corresponding to the N th subblock lookup table, and a second product of an output value of an adjacent lookup table and a weight corresponding to a adjacent lookup table, wherein the adjacent lookup table is a subblock lookup table adjacent to the N th subblock lookup table; and

replacing the output value of the N th subblock lookup table with a sum of the first product and the second product.

7 . The method according to claim 1 , wherein before the generating, based on the first image, the second image, and the deep neural network model, the lookup table corresponding to the first color adjustment, the method further comprises:

obtaining a first sample and a plurality of second samples obtained through different color adjustment on the first sample;

combining the first sample and the second sample of the first sample into a sample pair, wherein each sample pair comprises a first sample and a second sample of the first sample; and

inputting the sample pair to the deep neural network model to train the deep neural network model, so that a trained deep neural network model can output a global lookup table corresponding to each sample pair.

8 . The method according to claim 1 , wherein the generating, based on the first image, the second image, and the polynomial fitting algorithm, the lookup table corresponding to the first color adjustment comprises:

using a color value of the first image as an input value, and using a color value of the second image as an output value;

determining a target polynomial between the input value and the output value by using a polynomial fitting algorithm; and

outputting, by using the target polynomial, the lookup table corresponding to first color adjustment effect.

9 . The method according to claim 1 , wherein before the displaying a second interface, and displaying a generated lookup table on the second interface, the method further comprises:

displaying a third interface, and determining, on the third interface, a lookup table type selected by the user, wherein the lookup table type comprises a global lookup table and a local lookup table; and

obtaining a lookup table corresponding to the lookup table type, to display the lookup table corresponding to the lookup table type on the second interface.

10 . The method according to claim 1 , wherein the first file comprises at least one of an image or a video.

11 . An electronic device, comprising:

a display;

one or more processors; and

a memory, wherein

the display is configured to display an image generated by the processor, the memory stores one or more computer programs, including instructions that, when executed by the electronic device, enable the electronic device to perform image processing operations comprising:

displaying a first interface, and determining on the first interface, a first file selected by a user;

displaying a second interface, and displaying a generated lookup table on the second interface; wherein, the lookup table is generated by a preset deep neural network model or a polynomial fitting algorithm;

determining, on the second interface, a target lookup table selected by the user from the lookup table, and performing color adjustment on the first file by using the target lookup table; and

displaying a color-adjusted first file;

wherein the image processing operations further comprise:

determining a first image selected by the user and a second image corresponding to the first image, wherein the second image is an image obtained through first color adjustment on the first image; and

generating, based on the first image, the second image, and the deep neural network model, a lookup table corresponding to the first color adjustment; and generating, based on the first image. the second image, and the polynomial fitting algorithm, a lookup table corresponding to the first color adjustment, wherein the lookup table corresponding to the first color adjustment is used for enabling the first file and the second image to have same color adjustment effect during the color adjustment on the first file.

12 . A non-transitory computer-readable storage medium, storing instructions that, when run on an electronic device, enable the electronic device to perform image processing operations comprising:

displaying a first interface, and determining, on the first interface, a first file selected by a user,

displaying a second interface, and displaying a generated lookup table on the second interface, wherein, the lookup table is generated by a preset deep neural network model or a polynomial fitting algorithm;

determining, on the second interface, a target lookup table selected by the user from the lookup table, and performing color adjustment on the first file by using the target lookup table;

and displaying a color-adjusted first file;

wherein the image processing operations further comprise:

determining a first image selected by the user and a second image corresponding to the first image, wherein the second image is an image obtained through first color adjustment on the first image; and

generating, based on the first image, the second image, and the deep neural network model, a lookup table corresponding to the first color adjustment; and generating, based on the first image, the second image, and the polynomial fitting algorithm, a lookup table corresponding to the first color adjustment, wherein the lookup table corresponding to the first color adjustment is used for enabling the first file and the second image to have same color adjustment effect during the color adjustment on the first file.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (24)