Live broadcast content positioning method in video picture-in-picture architecture

Abstract

The invention discloses a live broadcast content positioning method in a video picture-in-picture architecture. The live broadcast content positioning method comprises the following steps: S1, intercepting a video picture and storing the video picture in a picture format; S2, selecting a plurality of pictures in the intercepted pictures, selecting the number of the pictures to be recorded as L, performing rectangular subtraction calculation on two pictures in the selected pictures as a group, taking an absolute value of a difference, and performing a plurality of groups of rectangular subtraction calculation to obtain a plurality of groups of results; S3, accumulating and summing the results in the step S2, and carrying out threshold operation; S4, performing filtering operation on the data obtained in the step S3; S5, carrying out row traversal and column traversal respectively, and obtaining a game window row range and a game window column range; and S9, obtaining correct game window matrix coordinates according to the result of the step S5, and segmenting the game window. According to the live broadcast content positioning method in the video picture-in-picture architecture provided by the invention, the live broadcast image area can be quickly positioned, and the influence caused by live broadcast image change is very small, and the occupied system resource is relatively low.

Description

technical field

[0001] The invention belongs to the technical field of network communication, relates to a video live broadcast method, in particular to a method for locating live content in a video picture-in-picture framework. Background technique

[0002] Compared with the live broadcast of mobile screen casting, in the live broadcast environment based on PC screen casting, the expression form of user-defined interface is added. In order to allow the live broadcast content and the host's characters to appear on the same screen at the same time, a picture-in-picture mode with multiple windows on the same screen is often used.

[0003] This mode increases the complexity of the display screen. The orientation and display ratio of the live content window in the screen are randomly defined by the user, which may cause deformation of the screen scale and bring challenges to the image-based live streaming content analysis system. , therefore, in this newly added mode, positioni...

Examples

Embodiment 1

[0110] see figure 1 , the present invention discloses a method for locating live content in a video picture-in-picture framework. Since there are many windows in the game video screen projected by the PC, the game windows are generally rectangular areas, and the motion vectors are relatively large compared to other windows. After subtracting multiple streaming images and accumulating the results, the gray value of the game window is large, while the gray value of other windows is small, and the edges of the game window are clear and rectangular. Traverse the image to extract the rectangular area of ​​the game window. The live content positioning method specifically includes the following steps:

[0111] Step S1, intercepting the video screen and saving it in a picture format;

[0112] Step S2, select several pictures in the intercepted picture, the number of pictures is recorded as L (such as 50), and every two pictures are a group of rectangular subtraction (such as 50 pict...

Embodiment 2

[0150] The difference between this embodiment and Embodiment 1 is that in this embodiment, the filtering method in step S4 may be other filtering methods than median filtering, and those skilled in the art may select other filtering methods according to needs.

Embodiment 3

[0152] The difference between this embodiment and Embodiment 1 is that in this embodiment, in step S2, every two pictures are used as a group to carry out rectangular subtraction, and all pictures can be divided into several groups, and the pictures in each group are compared with the pictures in other groups They may be completely different from each other, or they may be partly the same. When selecting pictures, the optimal solution is to select completely different pictures among the groups; as in the first embodiment, 50 pictures can be divided into 25 groups, and rectangular subtraction is performed on one group.

technical field

[0001] The invention belongs to the technical field of network communication, relates to a video live broadcast method, in particular to a method for locating live content in a video picture-in-picture framework. Background technique

[0002] Compared with the live broadcast of mobile screen casting, in the live broadcast environment based on PC screen casting, the expression form of user-defined interface is added. In order to allow the live broadcast content and the host's characters to appear on the same screen at the same time, a picture-in-picture mode with multiple windows on the same screen is often used.

[0003] This mode increases the complexity of the display screen. The orientation and display ratio of the live content window in the screen are randomly defined by the user, which may cause deformation of the screen scale and bring challenges to the image-based live streaming content analysis system. , therefore, in this newly added mode, positioni...

Examples

Embodiment 1

[0110] see figure 1 , the present invention discloses a method for locating live content in a video picture-in-picture framework. Since there are many windows in the game video screen projected by the PC, the game windows are generally rectangular areas, and the motion vectors are relatively large compared to other windows. After subtracting multiple streaming images and accumulating the results, the gray value of the game window is large, while the gray value of other windows is small, and the edges of the game window are clear and rectangular. Traverse the image to extract the rectangular area of ​​the game window. The live content positioning method specifically includes the following steps:

[0111] Step S1, intercepting the video screen and saving it in a picture format;

[0112] Step S2, select several pictures in the intercepted picture, the number of pictures is recorded as L (such as 50), and every two pictures are a group of rectangular subtraction (such as 50 pict...

Embodiment 2

[0150] The difference between this embodiment and Embodiment 1 is that in this embodiment, the filtering method in step S4 may be other filtering methods than median filtering, and those skilled in the art may select other filtering methods according to needs.

Embodiment 3

[0152] The difference between this embodiment and Embodiment 1 is that in this embodiment, in step S2, every two pictures are used as a group to carry out rectangular subtraction, and all pictures can be divided into several groups, and the pictures in each group are compared with the pictures in other groups They may be completely different from each other, or they may be partly the same. When selecting pictures, the optimal solution is to select completely different pictures among the groups; as in the first embodiment, 50 pictures can be divided into 25 groups, and rectangular subtraction is performed on one group.

Images