Method, apparatus and electronic device for detecting a display region of a video image in real time

Patent metadata

Field	Value
Publication number	US-2022044416-A1
Application number	US-202117444394-A
Country	US
Kind code	A1
Filing date	Aug 4, 2021
Priority date	Aug 5, 2020
Publication date	Feb 10, 2022
Grant date	—

Abstract

A method, apparatus and electronic device for detecting a display region of a video image in real time. The method includes: determining a bright pixel threshold line that distinguishes a display region from a black edge region of the current frame of image according to a pixel value of each pixel; determining a time-domain warning line according to time-domain motion statistic values of each row and each column of pixels of the current frame of image; determining a target optimal gradient line that distinguishes the display region from the black edge region of the current frame of image; and determining a boundary bright line between the display region and the black edge region of the current frame of image according to the bright pixel threshold line, the time-domain warning line and the target gradient line to determine the display region of the current frame of image.

First claim

1 . A method for detecting a display region of a video image in real time, comprising:
obtaining a pixel value of each pixel in a current frame of image, and determining a bright pixel threshold line that distinguishes a display region from a black edge region of the current frame of image according to the pixel value of each pixel; obtaining time-domain motion statistic values of pixels of the current frame of image relative to corresponding pixels of an immediately preceding frame of image, and determining a time-domain warning line that distinguishes the display region from the black edge region of the current frame of image according to the time-domain motion statistic values of each row and each column of pixels of the current frame of image; obtaining gradient information of each row and each column of pixels in the current frame of image, and determining a target gradient line that distinguishes the display region from the black edge region of the current frame of image according to the gradient information of each row and each column of pixels; determining a boundary bright line between the display region and the black edge region of the current frame of image according to positions of the bright pixel threshold line, the time-domain warning line and the target gradient line; and determining the display region of the current frame of image according to the boundary bright line.

2 . The method for detecting a display region of a video image in real time according to claim 1 , wherein said obtaining a pixel value of each pixel in a current frame of image, and determining a bright pixel threshold line that distinguishes a display region from a black edge region of the current frame of image according to the pixel value of each pixel, comprises:
reading a pixel value of each pixel in each row of the current frame of image sequentially from top to bottom, counting the number of bright pixels in each row of pixels, and determining a row in which the number of bright pixels exceeds a first preset number threshold as the bright pixel threshold line that distinguishes the display region from an upper black edge region of the current frame of image, wherein if the pixel value is greater than a bright pixel threshold, a corresponding pixel is determined as a bright pixel; reading a pixel value of each pixel in each row of pixels in the current frame of image downward from the bright pixel threshold line that distinguishes the display region from the upper black edge region of the current frame of image, and determining, when the number of bright pixels in a current row of pixels exceeds the first preset number threshold and the number of bright pixels between the current row and the bright pixel threshold line that distinguishes the display region from the upper black edge region of the current frame of image exceeds a first bright pixel number threshold, the current row as a bright pixel threshold line that distinguishes the display region from a lower black edge region of the current frame of image; reading a pixel value of each pixel in each column of pixels in the current frame of image sequentially from left to right, counting the number of bright pixels in each column of pixels, and determining a column in which the number of bright pixels exceeds a second preset number threshold as a bright pixel threshold line that distinguishes the display region from a left black edge region of the current frame of image, wherein if the pixel value is greater than the bright pixel threshold, a corresponding pixel is determined as a bright pixel; and reading a pixel value of each pixel in each column of pixels in the current frame of image towards the right from the bright pixel threshold line that distinguishes the display region from the left black edge region of the current frame of image, and determining, when the number of bright pixels of a current column of pixels exceeds the second preset number threshold and the number of bright pixels between the current column and the bright pixel threshold line that distinguishes the display region from the left black edge region of the current frame of image exceeds a second bright pixel number threshold, the current column as a bright pixel threshold line that distinguishes the display region from the right black edge region of the current frame of image.

3 . The method for detecting a display region of a video image in real time according to claim 2 , wherein said reading a pixel value of each pixel in each row of the current frame of image sequentially from top to bottom, counting the number of bright pixels in each row of pixels, and determining a row in which the number of bright pixels exceeds a first preset number threshold as the bright pixel threshold line that distinguishes the display region from an upper black edge region of the current frame of image, wherein if the pixel value being greater than a bright pixel threshold, a corresponding pixel being determined as a bright pixel, comprises:
reading a pixel value of each pixel in each row of pixels in the current frame of image sequentially from top to bottom; determining when the pixel value is greater than a first bright pixel sub-threshold, a corresponding pixel as a bright pixel, counting the number of bright pixels in each row of pixels, and determining the row in which the number of bright pixels exceeds the first preset number threshold, as a first bright pixel threshold line that distinguishes the display region from the upper black edge region of the current frame of image, or determining, when the pixel value is greater than a second bright pixel sub-threshold, a corresponding pixel as a bright pixel, counting the number of bright pixels in each row of pixels, and determining a row in which the number of bright pixels exceeds the second preset number threshold as a second bright pixel threshold line that distinguishes the display region from the upper black edge region of the current frame of image, wherein the first bright pixel sub-threshold value is greater than the second bright pixel sub-threshold value.

4 . The method for detecting a display region of a video image in real time according to claim 2 , wherein said reading a pixel value of each pixel in each row of pixels in the current frame of image downward from the bright pixel threshold line that distinguishes the display region from the upper black edge region of the current frame of image, and determining, when the number of bright pixels in a current row of pixels exceeds the first preset number threshold and the number of bright pixels between the current row and the bright pixel threshold line that distinguishes the display region from the upper black edge region of the current frame of image exceeds a first bright pixel number threshold, the current row as a bright pixel threshold line that distinguishes the display region from a lower black edge region of the current frame of image, comprises:
reading a pixel value of each pixel in each row of pixels in the current frame of image sequentially downwards from the bright pixel threshold line that distinguishes the display region from the upper black edge region of the current frame of image; determining, when the pixel value is greater than a first bright pixel sub-threshold, a corresponding pixel as a bright pixel, counting the number of bright pixels in each row of pixels, and determining the row in which the number of bright pixels exceeds the first preset number threshold as a first bright pixel threshold line that distinguishes the display region from the lower black edge region of the current frame of image; or determining, when the pixel value is greater than a second bright pixel sub-threshold, a corresponding pixel as a bright pixel, counting the number of bright pixels in each row of pixels, and determining a row in whic

Assignees

• Amlogic Shanghai Co Ltd

Inventors

• Ji Ping
• Bao Zheng
• Ji Tao
• Wang Chun
• Wang Dongjian
• Chen Xuyun

Classifications

• H04N21/44

  Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs · CPC title

• H04N5/262

  Studio circuits, e.g. for mixing, switching-over, change of character of image, other special effects {; Cameras specially adapted for the electronic generation of special effects} · CPC title

• H04N21/431 Primary

  Generation of visual interfaces {for content selection or interaction}; Content or additional data rendering · CPC title

• G06T7/12 Primary

  Edge-based segmentation · CPC title

• G06T2207/10016

  Video; Image sequence · CPC title