Search

CN-121999727-A - Backlight type distribution map acquisition method, image compensation method, device and system

CN121999727ACN 121999727 ACN121999727 ACN 121999727ACN-121999727-A

Abstract

The invention relates to the technical field of image display and discloses a backlight type distribution map acquisition method, an image compensation method, a device and a system, wherein the method comprises the steps of acquiring display picture information, wherein the display picture information comprises a brightness value of each pixel in a display picture; the method comprises the steps of carrying out backlight extraction processing on display picture information to obtain a backlight gray level map, carrying out interpolation processing on the backlight gray level map to obtain a target size to-be-processed image, and processing the to-be-processed image based on a target mapping relation to generate a backlight light pattern distribution map, wherein the target mapping relation is a mapping relation between the target size to-be-processed image and the backlight light pattern distribution map of the display device acquired through a sensor. The invention improves the precision of the obtained backlight type distribution image, and further improves the effect of image pixel compensation through the backlight type distribution image.

Inventors

  • Liao Jiancan

Assignees

  • 广州视源电子科技股份有限公司
  • 广州视睿电子科技有限公司

Dates

Publication Date
20260508
Application Date
20241101

Claims (14)

  1. 1. A backlight pattern generation method, wherein the method is applied to a display device, the method comprising: Acquiring display screen information, wherein the display screen information comprises a brightness value of each pixel in a display screen; Performing backlight extraction processing on the display picture information to obtain a backlight gray scale map, wherein the values of pixels in the backlight gray scale map are used for controlling the backlight brightness of backlight partitions of the display equipment; performing interpolation processing on the backlight gray level image to obtain an image to be processed with a target size; and processing the image to be processed based on a target mapping relation to generate a backlight pattern distribution map, wherein the target mapping relation is the mapping relation between the image to be processed with the target size and the backlight pattern distribution map of the display equipment acquired by the sensor.
  2. 2. The method according to claim 1, wherein the data processing the image to be processed based on the target mapping relationship to obtain a backlight profile comprises: performing data processing on the image to be processed through the trained super-resolution convolutional neural network model to obtain a backlight pattern distribution map; And the trained super-resolution convolutional neural network model is used for representing the target mapping relation.
  3. 3. The method of claim 2, wherein the super-resolution convolutional neural network model comprises a feature extraction layer, a nonlinear mapping layer, and a reconstruction layer, connected in sequence; performing data processing on the image to be processed through a super-resolution convolutional neural network model to obtain a backlight profile, wherein the method comprises the following steps: performing data processing on the image to be processed through the feature extraction layer to obtain a plurality of first-layer feature images; Respectively carrying out nonlinear mapping processing on the plurality of first-layer feature images through the nonlinear mapping layer to obtain a plurality of second-layer feature images; And carrying out convolution processing on the plurality of second-layer feature maps through the reconstruction layer to obtain the backlight light pattern distribution map.
  4. 4. A method according to claim 3, wherein the data processing of the image to be processed by the feature extraction layer to obtain a plurality of first layer feature maps comprises: according to the backlight brightness range of the display equipment, carrying out normalization processing on the brightness of each pixel in the image to be processed to obtain a normalized input image; And carrying out convolution operation on the normalized input image through a plurality of convolution cores in the feature extraction layer to obtain a plurality of first-layer feature graphs.
  5. 5. The method of claim 2, further comprising a training process of a super-resolution convolutional neural network model; The training process of the super-resolution convolutional neural network model comprises the following steps: Acquiring sample picture information, wherein the sample picture information comprises a brightness value of each pixel in a sample display picture; performing backlight extraction processing on the sample picture information to obtain a sample backlight gray scale map, wherein the values of pixels in the backlight gray scale map are used for controlling the backlight brightness of backlight partitions of the display equipment; Performing interpolation processing on the sample backlight gray level image to obtain a sample to-be-processed image with a target size; acquiring a sample backlight light pattern distribution map, wherein the sample backlight light pattern distribution map is acquired by the sensor when a backlight module of the display device works according to the sample backlight gray scale map; and training the super-resolution convolutional neural network model by taking the sample to-be-processed image as a training sample and the sample backlight type distribution map as a training label.
  6. 6. The method of claim 5, wherein training the super-resolution convolutional neural network model with the sample to-be-processed image as a training sample and the sample backlight profile as a training label comprises: According to the backlight brightness range of the display equipment, carrying out normalization processing on the brightness of each pixel in the sample to-be-processed image to obtain a sample image to be input; Acquiring a brightness difference value between the maximum brightness and the minimum brightness of pixels in the sample backlight light distribution map; Based on the brightness difference value, carrying out normalization processing on the brightness of each pixel in the sample backlight light type distribution map so as to obtain a target image; processing the sample image to be input through the super-resolution convolutional neural network model to obtain a predicted image; And updating parameters of the super-resolution convolutional neural network model based on errors between the predicted image and the target image.
  7. 7. The method of any one of claims 1 to 6, wherein the sensor is a CMOS camera.
  8. 8. The method according to any one of claims 1 to 6, wherein the target size is an image size of display screen information; the interpolating the backlight gray scale image to obtain an image to be processed with a target size includes: And carrying out bilinear interpolation on the backlight gray scale image to obtain the target-sized image to be processed.
  9. 9. An image compensation method, wherein the method is applied to a display device, the method comprising: Acquiring display screen information, wherein the display screen information comprises a brightness value of each pixel in a display screen; Performing backlight extraction processing on the display picture information to obtain a backlight gray scale map, wherein the values of pixels in the backlight gray scale map are used for controlling the backlight brightness of backlight partitions of the display equipment; performing interpolation processing on the backlight gray level image to obtain an image to be processed with a target size; Performing data processing on the image to be processed based on a target mapping relationship to obtain a backlight pattern distribution map, wherein the target mapping relationship is a mapping relationship between the image to be processed with a target size and the backlight pattern distribution map of display equipment acquired by a sensor; And acquiring an adjustment coefficient of the pixels of the target area in the display picture according to the brightness value of the pixels of the target area in the backlight type distribution diagram so as to carry out brightness compensation on the pixels of the target area in the display picture.
  10. 10. The method of claim 9, wherein the obtaining the adjustment coefficient of each pixel according to the brightness value of each pixel in the backlight profile comprises: and obtaining an adjustment coefficient of the pixels of the target area according to the ratio of the backlight full brightness value of the display equipment to the brightness value of the pixels of the target area in the backlight light type distribution diagram.
  11. 11. A backlight profile acquisition apparatus, the apparatus being provided to a display device, the apparatus comprising: a first acquisition module for acquiring display screen information including a luminance value of each pixel in a display screen; the display device comprises a first extraction module, a first display module and a second extraction module, wherein the first extraction module is used for carrying out backlight extraction processing on the display picture information to obtain a backlight gray scale image; The first interpolation module is used for carrying out interpolation processing on the backlight gray level image to obtain an image to be processed with a target size; The first mapping module is used for processing the image to be processed based on a target mapping relation to generate a backlight light type distribution map, wherein the target mapping relation is a mapping relation between the image to be processed with a target size and the backlight light type distribution map of the display equipment acquired by the sensor.
  12. 12. An image compensation apparatus, the apparatus being provided to a display device, the apparatus comprising: A second acquisition module configured to acquire display screen information including a luminance value of each pixel in a display screen; The second extraction module is used for carrying out backlight extraction processing on the display picture information to obtain a backlight gray scale image, wherein the values of pixels in the backlight gray scale image are used for controlling the backlight brightness of the backlight subareas of the display equipment; the second difference module is used for carrying out interpolation processing on the backlight gray level image to obtain an image to be processed with a target size; the second mapping module is used for carrying out data processing on the image to be processed based on a target mapping relation to obtain a backlight light type distribution map, wherein the target mapping relation is a mapping relation between the image to be processed with a target size and the backlight light type distribution map of the display equipment acquired by the sensor; And the pixel compensation module is used for acquiring the adjustment coefficient of the pixel of the target area in the display picture according to the brightness value of the pixel of the target area in the backlight type distribution diagram so as to carry out brightness compensation on the pixel of the target area in the display picture.
  13. 13. A system on a chip, wherein the system on a chip is arranged in a display device, and the system on a chip comprises a central processing unit and a neural network processing unit; the central processing unit is used for acquiring display picture information, wherein the display picture information comprises a brightness value of each pixel in a display picture; The central processing unit is also used for carrying out backlight extraction processing on the display picture information to obtain a backlight gray scale image, wherein the values of pixels in the backlight gray scale image are used for controlling the backlight brightness of the backlight subareas of the display equipment; the central processing unit is also used for carrying out interpolation processing on the backlight gray level image to obtain an image to be processed with a target size; the neural network processing unit is used for carrying out data processing on the image to be processed based on a target mapping relation to obtain a backlight pattern distribution map, wherein the target mapping relation is a mapping relation between the image to be processed with a target size and the backlight pattern distribution map of the display equipment acquired by the sensor; The central processing unit is further configured to obtain an adjustment coefficient of the pixel of the target area in the display screen according to the brightness value of the pixel of the target area in the backlight profile, so as to perform brightness compensation on the pixel of the target area in the display screen.
  14. 14. A computer readable storage medium having stored thereon computer instructions for causing a display device to perform the method of any one of claims 1 to 8 or for causing a display device to perform the method of claim 9 or 10.

Description

Backlight type distribution map acquisition method, image compensation method, device and system Technical Field The invention relates to the technical field of image display, in particular to a backlight type distribution map acquisition method, an image compensation device and an image compensation system. Background Local Dimming is a backlight control technique used in displays, particularly liquid crystal displays. The purpose is to improve the contrast and visual effect of the image by dynamically adjusting the backlight brightness of different areas. In the Local Dimming technique, the backlight is divided into a plurality of independent areas (referred to as backlight partitions or areas). The brightness of each region can be independently controlled so that the brightness of the different regions can be individually adjusted when displaying the image. However, this technique may cause some areas to be non-uniform in brightness and therefore pixel compensation is required to ensure that the overall brightness and color of the image display remains substantially unchanged from the full backlight brightness. The machine with the Local dimming function acquires an input picture, acquires a corresponding backlight matrix through a backlight extraction method, and then estimates a backlight pattern distribution diagram after the backlight module is lightened according to the backlight matrix and diverged to the periphery in the backlight module through a BMA (fuzzy diffusion) method, so that the follow-up image pixel compensation is conveniently carried out according to the backlight pattern distribution diagram obtained through estimation. However, in the above solution, the accuracy of the backlight profile obtained by the BMA method is low, resulting in poor image pixel compensation effect. Disclosure of Invention In view of the above, the present invention provides a backlight profile acquisition method, an image compensation device and an image compensation system, so as to solve the problem that the image pixel compensation effect is poor due to low accuracy of the obtained backlight profile. In a first aspect, the invention provides a backlight pattern distribution map generation method, which is applied to display equipment, and the method comprises the steps of obtaining display picture information, carrying out backlight extraction processing on the display picture information to obtain a backlight gray map, wherein the values of pixels in the backlight gray map are used for controlling backlight brightness of backlight partitions of the display equipment, carrying out interpolation processing on the backlight gray map to obtain a target size to-be-processed image, and processing the to-be-processed image based on a target mapping relation to generate the backlight pattern distribution map, wherein the target mapping relation is a mapping relation between the target size to-be-processed image and the backlight pattern distribution map of the display equipment acquired through a sensor. According to the backlight type distribution map generation method provided by the embodiment, the display device can acquire the display picture information to be displayed firstly, the display picture information comprises the brightness value of each pixel in the display picture, at the moment, the display device can conduct backlight extraction processing on the display picture information according to the backlight partition of the display device, so that a backlight gray map for indicating the backlight brightness of the backlight partition of the display device is obtained, the display device interpolates the backlight gray map to obtain an image to be processed with a target size, the image to be processed is directly mapped into the backlight type distribution map through the target mapping relation, and at the moment, the backlight type distribution map is close to the backlight type distribution map of the display device which is directly acquired through the sensor. The backlight distribution map obtained by the invention is closer to the actual backlight display of a display picture, the accuracy of the obtained backlight distribution image is improved, and the effect of image pixel compensation through the backlight distribution image is further improved. In an alternative embodiment, the data processing is performed on the image to be processed based on the target mapping relationship to obtain a backlight profile, including: Performing data processing on the image to be processed through the trained super-resolution convolutional neural network model to obtain a backlight pattern distribution map; The trained super-resolution convolutional neural network model is used for representing the target mapping relation. In the embodiment, the mapping relation between the image to be processed and the backlight pattern distribution map of the display picture acquired by the sensor can be represented by tr