Image processing methods and apparatuses, electronic devices and storage media
An image processing method and apparatus, an electronic device and a storage medium are provided. The method includes: obtaining an original image, and for each of pixels in the original image, converting first color data of the pixel into second color data corresponding to a set color space (S 12 ); calculating a feature weighted value of the pixel according to the second color data and a preset rule (S 14 ); and determining a uniformity of the pixels based on the feature weighted values to mark the pixels (S 16 ).
1 . A method of processing an image, comprising:
obtaining an original image, and for each of pixels in the original image,
converting first color data of the pixel into second color data corresponding to a set color space;
calculating a feature weighted value of the pixel according to the second color data and a preset rule; and
determining a uniformity of the pixels based on the feature weighted values to mark the pixels;
wherein determining the uniformity of the pixels based on the feature weighted values to mark the pixels comprises:
for each of the pixels, determining a feature difference between the feature weighted value of the pixel and the feature weighted value of an adjacent pixel;
in response to determining that the feature difference is greater than a first threshold, marking the pixel as a first-type boundary pixel; and
in response to determining that the feature difference is not greater than the first threshold, marking the pixel as a non-boundary pixel.
2 . The method according to claim 1 , wherein the set color space is an HSV color space.
3 . The method according to claim 2 , wherein the second color data comprises hue data, saturation data and brightness data, and
calculating the feature weighted value of the pixel according to the second color data and the preset rule comprises:
obtaining third color data by preprocessing the second color data;
obtaining quantization values corresponding to the hue data, the saturation data and the brightness data by quantizing the third color data based on a color-based segmentation rule; and
determining the feature weighted value according to the quantization values corresponding to the hue data, the saturation data and the brightness data.
4 . The method according to claim 3 , wherein obtaining the third color data by preprocessing the second color data comprises:
in response to determining that the brightness data of the second color data is less than a first preset value, determining that the second color data is black data; and
in response to determining that the saturation data of the second color data is less than a second preset value and the brightness data of the second color data is greater than a third preset value, determining that the second color data is white data.
5 . The method according to claim 1 , wherein determining the uniformity of the pixels based on the feature weighted values to mark the pixels further comprises:
for each of the first-type boundary pixels,
determining a neighborhood of the first-type boundary pixel, and calculating a number of first-type boundary pixels in the neighborhood;
in response to determining that the number of first-type boundary pixels is greater than a second threshold, marking the first-type boundary pixel as a first-type non-uniform pixel;
in response to determining that the number of first-type boundary pixels is not greater than a second threshold, marking each of the first-type boundary pixel and the non-boundary pixel as a uniform pixel; and
generating a first uniformity distribution image based on the first-type non-uniform pixels and the uniform pixels.
6 . The method according to claim 5 , wherein determining the uniformity of the pixels based on the feature weighted values to mark the pixels further comprises:
obtaining second-type non-uniform pixels by performing expansion processing on the first uniformity distribution image, so as to generate a second uniformity distribution image;
in the second uniformity distribution image, marking the second-type non-uniform pixels whose adjacent pixels are the uniform pixels along preset directions as second-type boundary pixels; and
for each of the second-type boundary pixels, in a row and/or a column where the second-type boundary pixel is located, marking the second-type boundary pixel and a preset number of pixels adjacent to the second-type boundary pixel as transitional pixels; marking the second-type non-uniform pixels that are not the transitional pixels as target non-uniform pixels, and marking the uniform pixels that are neither the transitional pixels nor the second-type non-uniform pixels as target uniform pixels.
7 . The method according to claim 6 , wherein marking the second-type boundary pixel and the preset number of pixels adjacent to the second-type boundary pixel as transitional pixels comprises:
determining the preset number by randomly selecting a value within a preset range.
8 . The method according to claim 6 , further comprising:
determining respective uniformity compensation matrices according to the uniformity of the pixels; and
obtaining a display image by processing the pixels according to the respective uniformity compensation matrices.
9 . The method according to claim 8 , wherein determining the respective uniformity compensation matrices according to the uniformity of the pixels comprises:
respectively obtaining first color gamut conversion matrices corresponding to the target non-uniform pixels, second color gamut conversion matrices corresponding to the transitional pixels and third color gamut conversion matrices corresponding to the target uniform pixels, and obtaining display pixel conversion matrices of light-emitting elements in a display panel corresponding to the pixels, and
obtaining the display image by processing the pixels according to the respective uniformity compensation matrices comprises:
processing each of the target non-uniform pixels according to the corresponding first color gamut conversion matrix and the corresponding display pixel conversion matrix;
processing each of the transitional pixels according to the corresponding second color gamut conversion matrix and the corresponding display pixel conversion matrix; and
processing each of the target uniform pixels according to the corresponding third color gamut conversion matrix and the corresponding display pixel conversion matrix.
10 . The method according to claim 9 , wherein the first color gamut conversion matrices correspond to a first color gamut, the second color gamut conversion matrices correspond to a second color gamut, and the third color gamut conversion matrices correspond to a third color gamut; and
a range of the first color gamut is larger than a range of the second color gamut, and a range of the second color gamut is larger than a range of the third color gamut.
11 . The method according to claim 10 , further comprising:
determining the third color gamut according to a common color gamut range of the light-emitting elements;
determining the second color gamut according to the third color gamut and a first ratio; and
determining the first color gamut according to the third color gamut and a second ratio, wherein the second ratio is greater than the first ratio.
12 . The method according to claim 10 , wherein respectively obtaining the first color gamut conversion matrices corresponding to the target non-uniform pixels, the second color gamut conversion matrices corresponding to the transitional pixels and the third color gamut conversion matrices corresponding to the target uniform pixels comprises:
determining each of the first color gamut conversion matrices according to chromaticity coordinates and brightness corresponding to three primary colors in the first color gamut;
determining each of the second color gamut conversion matrices according to chromaticity coordinates and brightness corresponding to the three primary colors in the second color gamut; and
determining each of the third color gamut conversion matrices according to chromaticity coordinates and brightness corresponding to the three primary colors in the third color gamut.
13 . The method according to claim 9 , wherein obtaining the display pixel conversion matrices of the light-emitting elements in the display panel corresponding to the pixels comprises:
for each of the light-emitting elements in the display panel corresponding to the pixels, determining the corresponding display pixel conversion matrix according to RGB color data of the light-emitting element.
14 . An electronic device, comprising:
one or more processors, and a memory; and
one or more programs, wherein the one or more programs are stored in the memory and executed by the one or more processors, and the one or more programs comprise instructions for executing operates comprising:
obtaining an original image, and for each of pixels in the original image,
converting first color data of the pixel into second color data corresponding to a set color space;
calculating a feature weighted value of the pixel according to the second color data and a preset rule; and
determining a uniformity of the pixels based on the feature weighted values to mark the pixels;
wherein determining the uniformity of the pixels based on the feature weighted values to mark the pixels comprises:
for each of the pixels, determining a feature difference between the feature weighted value of the pixel and the feature weighted value of an adjacent pixel;
in response to determining that the feature difference is greater than a first threshold, marking the pixel as a first-type boundary pixel; and
in response to determining that the feature difference is not greater than the first threshold, marking the pixel as a non-boundary pixel.
15 . A nonvolatile computer-readable storage medium of a computer program, wherein when the computer program is executed by one or more processors, the processors are caused to execute the method according to claim 1 .
16 . The electronic device according to claim 14 , wherein the set color space is an HSV color space.
17 . The electronic device according to claim 16 , wherein the second color data comprises hue data, saturation data and brightness data, and
calculating the feature weighted value of the pixel according to the second color data and the preset rule comprises:
obtaining third color data by preprocessing the second color data;
obtaining quantization values corresponding to the hue data, the saturation data and the brightness data by quantizing the third color data based on a color-based segmentation rule; and
determining the feature weighted value according to the quantization values corresponding to the hue data, the saturation data and the brightness data.
18 . The electronic device according to claim 17 , wherein obtaining the third color data by preprocessing the second color data comprises:
in response to determining that the brightness data of the second color data is less than a first preset value, determining that the second color data is black data; and
in response to determining that the saturation data of the second color data is less than a second preset value and the brightness data of the second color data is greater than a third preset value, determining that the second color data is white data.
19 . A method of processing an image, comprising:
obtaining an original image, and for each of pixels in the original image,
converting first color data of the pixel into second color data corresponding to a set color space;
calculating a feature weighted value of the pixel according to the second color data and a preset rule; and
determining a uniformity of the pixels based on the feature weighted values to mark the pixels;
wherein the set color space is an HSV color space;
wherein the second color data comprises hue data, saturation data and brightness data, and
calculating the feature weighted value of the pixel according to the second color data and the preset rule comprises:
obtaining third color data by preprocessing the second color data;
obtaining quantization values corresponding to the hue data, the saturation data and the brightness data by quantizing the third color data based on a color-based segmentation rule; and
determining the feature weighted value according to the quantization values corresponding to the hue data, the saturation data and the brightness data.