Method and device for detecting bad points in video and computer readable medium

Patent metadata

Field	Value
Publication number	US-12406462-B2
Application number	US-202117771941-A
Country	US
Kind code	B2
Filing date	Jul 5, 2021
Priority date	Aug 25, 2020
Publication date	Sep 2, 2025
Grant date	Sep 2, 2025

Abstract

Embodiments of the disclosure provide a method for detecting bad points in a video, including: performing extreme filtering respectively on first, second and third frames of images which are sequentially and continuously in the video to obtain first, second and third filtered images, respectively; wherein the extreme filtering is one of maximum filtering and minimum filtering; determining first and second difference images according to the first, second and third filtered images; determining a candidate image according to the first and second difference images; and determining that at least part of points in the second frame of image corresponding to the valid point in the candidate image are bad points. The embodiment of the disclosure also provides a device and a computer-readable medium for detecting bad points in the video.

First claim

The invention claimed is:

1. A method of detecting bad points in a video, comprising:
performing extreme filtering respectively on a first frame of image, a second frame of image and a third frame of image which are sequentially and continuously in the video, to obtain a first filtered image, a second filtered image and a third filtered image, respectively, wherein the extreme filtering is a maximum filtering or a minimum filtering;
determining a first difference image and a second difference image according to the first filtered image, the second filtered image and the third filtered image; wherein the first difference image represents a difference in brightness between the first filtered image and the second filtered image, and the second difference image represents a difference in brightness between the second filtered image and the third filtered image;
determining a candidate image according to the first difference image and the second difference image; a point in the candidate image whose corresponding point in the first difference image has a brightness exceeding a first threshold value and whose corresponding point in the second difference image has a brightness exceeding a second threshold value is a valid point, and other points in the candidate image are invalid points; and
determining that at least part of points in the second frame of image corresponding to the valid point in the candidate image are bad points,
wherein the determining a candidate image according to the first difference image and second difference image comprises:
determining a first binarized image according to the first difference image, and determining a second binarized image according to the second difference image; a point in the first binarized image whose corresponding point in the first difference image has a brightness exceeding the first threshold value is a first valid point, and a point in the second binarized image whose corresponding point in the second difference image has a brightness exceeding the second threshold value is a second valid point; and
determining the candidate image according to the first binarized image and the second binarized image; wherein a point in the candidate image whose corresponding point is the first valid point in the first binarized image and whose corresponding point is the second valid point in the second binarized image is a valid point, and other points in the candidate image are invalid points,
before the determining the first binarized image from the first difference image and determining the second binarized image from the second difference image, the method further comprises:
normalizing a brightness of all points in the first difference image into a number greater than or equal to 0 and less than or equal to 1 by taking a brightness of a point in the first difference image with a maximum brightness as 1 and a minimum brightness as 0; normalizing a brightness of all the points in the second difference image into a number greater than or equal to 0 and less than or equal to 1 by taking a brightness of a point in the second difference image with a maximum brightness as 1 and a minimum brightness as 0;
the first threshold value T 1 is calculated by following formula: T 1 =(1−minN 1 )*factor;
the second threshold value T 2 is calculated by following formula: T 2 =(1−minN 2 )*factor;
wherein the minN 1 is a normalized brightness of the point whose brightness is minimum in the first difference image, the minN 2 is a normalized brightness of the point whose brightness is minimum in the second difference image, and the factor is a preset coefficient larger than 0 and smaller than 1.

2. The method of claim 1 , wherein the performing the extreme filtering respectively on the first frame of image, the second frame of image and the third frame of image which are sequential and consecutive in the video comprises:
converting the first frame of image, the second frame of image and the third frame of image into gray scale images, respectively, wherein a brightness of each point of the gray scale images is represented by a gray scale value; and
performing the extreme filtering on the gray scale image of the first frame of image, the gray scale image of the second frame of image and the gray scale image of the third frame of image, respectively.

3. The method of claim 1 , wherein the performing the extreme filtering respectively on the first frame of image, the second frame of image and the third frame of image which are sequential and consecutive in the video comprises:
performing the extreme filtering multiple times, with a plurality of windows having sequentially reduced lengths, on each of the first frame of image, the second frame of image and the third frame of image.

4. The method of claim 1 , wherein,
the factor is greater than or equal to 0.1 and less than or equal to 0.5.

5. The method of claim 1 , wherein the determining that the at least part of the points in the second frame of image corresponding to the valid point in the candidate image are bad points comprises:
removing discrete ones of the valid points in the candidate image; and
determining points in the second frame of image corresponding to the remaining ones of the valid points in the candidate image as bad points.

6. The method of claim 5 , wherein the removing the discrete valid points in the candidate images comprises:
performing a corrosion processing on the candidate image by taking the valid points as a foreground and the invalid points as a background; and
performing an expansion processing on the candidate image by taking the valid point as a foreground and the invalid point as a background.

7. The method of claim 6 , wherein,
a structural element for the corrosion processing is a square whose side length is greater than or equal to 7 points and less than or equal to 15 points; and
a structural element for the expansion processing is a square whose side length is greater than or equal to 7 points and less than or equal to 15 points.

8. A device of detecting bad points in a video, comprising:
an input module configured to receive the video;
an output module configured to output the detected bad points in the video;
one or more processors; and
a memory having one or more programs stored thereon, that when executed by the one or more processors, cause the one or more processors to implement the method of detecting bad points in the video according to claim 1 ; and
one or more input/output interfaces connected between the one or more processors and the memory and configured to implement information interaction between the one or more processors and the memory.

9. A non-transitory computer-readable medium, on which a computer program is stored, and which performs the method of detecting bad points in the video according to claim 1 , when being executed by a processor.

10. The method of claim 1 , wherein the extreme filtering is the maximum filtering, a brightness of each point in the first difference image is a brightness of a corresponding point in the second filtered image minus a brightness of a corresponding point in the first filtered image, and a brightness of each point in the second difference image is a brightness of a corresponding point in the second filtered image minus a brightness of a corresponding point in the third filtered image.

11. The method of claim 1 , wherein the extreme filtering is the minimum filtering, a brightness of each point in the first difference image is a brightness of a corresponding point in the first filtered image minus a brightness of a corresponding point in the second filtered image, and a brightness of each point in the second differ

Assignees

• Boe Technology Group Co Ltd

Inventors

• Zhu Dan
• Duan Ran

Classifications

• G06T2207/20224

  Image subtraction · CPC title

• G06T5/50

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

• G06V10/28

  Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns · CPC title

• G06V20/46

  Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames · CPC title

• G06V10/443 Primary

  by matching or filtering · CPC title