Image processing method and device, storage medium and computer device

Patent metadata

Field	Value
Publication number	US-11210765-B2
Application number	US-201816300971-A
Country	US
Kind code	B2
Filing date	Apr 24, 2018
Priority date	Aug 28, 2017
Publication date	Dec 28, 2021
Grant date	Dec 28, 2021

Abstract

An image processing method and device, storage medium and computer device are provided. The method includes: generating an original gray scale image of an original image; performing a histogram equalization process on the original gray scale image to obtain an equalized gray scale image; generating decision factor distribution image, wherein the decision factor distribution image includes a first marked region including a region where pixels that are adjacent in position and have standard deviations smaller than set value in the original image are located, and second marked region; obtaining final gray scale image according to original gray scale image, equalized gray scale image and decision factor distribution image. Gray scale values of pixel corresponding to second marked region and first marked region in final gray scale image are respectively gray scale values of corresponding pixel in equalized gray scale image and original gray scale image; and restoring a processed image.

First claim

What is claimed is:

1. An image processing method, comprising:
generating an original gray scale image of an original image;
performing a histogram equalization process on the original gray scale image to obtain an equalized gray scale image;
generating a decision factor distribution image according to the original gray scale image, wherein the decision factor distribution image comprises a first marked region and a second marked region, the first marked region comprises a region where pixels that are adjacent in position and have standard deviations smaller than a set value in the original image are located, and the second marked region is a region other than the first marked region in the decision factor distribution image;
obtaining a final gray scale image according to the original gray scale image, the equalized gray scale image and the decision factor distribution image, wherein a gray scale value of a pixel corresponding to the second marked region in the final gray scale image is a gray scale value of a pixel corresponding to the second marked region in the equalized gray scale image, and a gray scale value of a pixel corresponding to the first marked region in the final gray scale image is a gray scale value of a pixel corresponding to the first marked region in the original gray scale image; and
restoring a processed image according to the final gray scale image.

2. The image processing method according to claim 1 , wherein the generating a decision factor distribution image according to the original gray scale image comprises:
calculating a standard deviation of each pixel in the original gray scale image to obtain a standard deviation distribution image according to the gray scale values of pixels within a certain region centered on each pixel in the original gray scale image, wherein the standard deviation distribution image comprises the standard deviation of each pixel in the original gray scale image; and
dividing a pixel having the standard deviation smaller than the set value in the standard deviation distribution image into the first marked region, and dividing a region other than the first marked region in the standard deviation distribution image as the second marked region to obtain the decision factor distribution image.

3. The image processing method according to claim 2 , wherein the calculating a standard deviation of each pixel in the original gray scale image to obtain a standard deviation distribution image according to the gray scale values of pixels within a certain region centered on each pixel in the original gray scale image comprises:
calculating a squared value of the gray scale value of each pixel in the original gray scale image to form a gray scale square image;
performing mean filtering on the original gray scale image to generate a first expectancy image, and performing mean filtering on the gray scale square image to generate a second expectancy image;
calculating a squared value of the gray scale value of each pixel in the first expectancy image to obtain a third expectancy image;
calculating a difference value between the gray scale values of pixels in corresponding positions of the second expectancy image and the third expectancy image to obtain a difference image; and
calculating a square root of the gray scale value of each pixel in the difference image to obtain the standard deviation distribution image.

4. The image processing method according to claim 3 , wherein the performing mean filtering on the original gray scale image comprises:
performing the mean filtering on the original gray scale image by means of a filtering template having a size of an m*m, m being 10 to 20, and the performing mean filtering on the gray scale square image comprises:
performing the mean filtering on the gray scale square image by means of a filtering template having a size of an m*m.

5. The image processing method according to claim 4 , wherein m is 15.

6. The image processing method according to claim 2 , wherein after the generating a decision factor distribution image according to the original gray scale image, the image processing method further comprises:
updating the decision factor distribution image to enable that in the updated decision factor distribution image, the second marked region comprises a region where pixels that are adjacent in position and have standard deviations greater than or equal to the set value in the original image are located, and a region with the number of the pixels therein smaller than a threshold value in the region where the pixels that are adjacent in position and have standard deviations smaller than the set value in the original image are located.

7. The image processing method according to claim 6 , wherein the threshold value is 8%-15% of the number of the pixels of the original image.

8. The image processing method according to claim 7 , wherein the threshold value is 10 % of the number of the pixels of the original image.

9. The image processing method according to claim 1 herein the set value ranges from 1 to 5.

10. An image processing device, comprising:
a generating circuit configured to generate an original gray scale image of an original image;
a histogram equalization circuit configured to perform a histogram equalization process on the original gray scale image to obtain an equalized gray scale image;
a first processing circuit configured to generate a decision factor distribution image according to the original gray scale image, wherein the decision factor distribution image comprises a first marked region and a second marked region, the first marked region comprises a region where pixels that are adjacent in position and have standard deviations smaller than a set value in the original image are located, and the second marked region is a region other than the first marked region in the decision factor distribution image;
a second processing circuit configured to obtain a final gray scale image according to the original gray scale image, the equalized gray scale image and the decision factor distribution image, wherein a gray scale value of a pixel corresponding to the second marked region in the final gray scale image is a gray scale value of a pixel corresponding to the second marked region in the equalized gray scale image, and a gray scale value of a pixel corresponding to the first marked region in the final gray scale image is a gray scale value of a pixel corresponding to the first marked region in the original gray scale image; and
a third processing circuit configured to restore a processed image according to the final gray scale image.

11. The image processing device according to claim 10 , wherein the first processing circuit is configured to: calculate a standard deviation of each pixel in the original gray scale image to obtain a standard deviation distribution image according to the gray scale values of pixels within a certain region centered on each pixel in the original gray scale image, wherein the standard deviation distribution image comprises the standard deviation of each pixel in the original gray scale image; and
divide a pixel having the standard deviation smaller than the set value in the standard deviation distribution image into the first marked region, and divide a region oilier than the first marked region in the standard deviation distribution image into the second marked region to obtain the decision factor distribution image.

12. The image processing device according to claim 11 , wherein the first processing circuit comprises:
a first calculating sub-circuit configured to calculate a squared value of the gray scale value of each pixel in the ori

Assignees

• Beijing Boe Display Tech Co
• Boe Technology Group Co Ltd

Inventors

• Guo Jun
• Shao Xibin
• Zhou Chengqi
• Chen Ming
• Li Shou
• Wang Jieqiong

Classifications

• G06T5/90 Primary

  Dynamic range modification of images or parts thereof · CPC title

• G06T5/50

  using two or more images, e.g. averaging or subtraction · CPC title

• G06T2207/10024

  Color image · CPC title

• G06T5/40

  using histogram techniques · CPC title

• G06T5/20

  using local operators · CPC title