WO2016070717A1

WO2016070717A1 - Method and device for processing video image

Info

Publication number: WO2016070717A1
Application number: PCT/CN2015/092575
Authority: WO
Inventors: 杨亮; 杨扬; 张强
Original assignee: 中兴通讯股份有限公司
Priority date: 2014-11-07
Filing date: 2015-10-22
Publication date: 2016-05-12
Also published as: CN105635634A

Abstract

Disclosed are a method and device for processing a video image. The method comprises: acquiring image information of a video image, and determining adjustable objects in the image information according to pre-stored image matching information; and adjusting each determined adjustable object according to a preset adjusting policy.

Description

Method and device for realizing video image processing

Technical field

Embodiments of the present invention relate to, but are not limited to, multimedia application processing technologies, and in particular, to a method and apparatus for implementing video image processing.

Background technique

With the development of multimedia application processing technology, image processing technology has also been developed in the process of image editing and processing. Through the image processing of artificial or single objects, the user experience of image processing in practical applications is greatly improved.

Video conferencing belongs to a special field of video images. The video conferencing system needs to collect information of multiple participants in multiple venues, and display the screen information obtained by the video capture device through the remote screen, for the conference site that appears in the video image. The scene, the configuration of the site, and the display image of the participant. The videoconferencing system of the related technology does not perform any processing. Therefore, the scene of the site may not be displayed clearly, or the scene may be too dark, resulting in the inability to clearly view the details of the site. The user experience of the video conferencing is not good, or the venue configuration is not in line with the video conference theme, or the video image of the venue has a mess or may affect the conference, etc., affecting the user experience of video conferencing for efficient communication; Video conferences that are in close contact, and undesired video image display, may affect the psychological assessment of participants, and even affect the original intention of the conference organizer to complete the conference tasks, reducing the efficiency of the implementation of conference content.

Summary of the invention

The following is an overview of the topics detailed in this document. This Summary is not intended to limit the scope of the claims.

Embodiments of the present invention provide a method and apparatus for implementing video image processing, which can adjust image information in a video image to improve display effect and user experience of the video image.

An embodiment of the present invention provides a method for implementing video image processing, including:

Obtaining image information of the video image, and determining the image letter according to the image matching information stored in advance Adjustable object in the interest;

Each of the determined adjustable objects is adjusted according to a preset adjustment strategy.

Optionally, the image matching information includes: matching condition information used for matching one or more objects: a video scene, a video person, and a video item;

Determining the adjustable objects in the image information includes:

The image information is matched with the image matching information stored in advance, and an object that matches the matching condition information in the image information is taken as an adjustable object.

Optionally, the preset adjustment policy includes:

When the adjustable object is a video scene, parameter adjustment is performed on the adjustable object, wherein the parameter includes one or more of the following: brightness, tone value, contrast value, saturation;

Demodulating or replacing one or more of the following features of the adjustable object when the adjustable object is a video person: facial features, skin color, hairstyle, glasses, clothing;

When the adjustable object is a video item, the adjustable object is overlaid or one or more of the following features of the adjustable object are adjusted: position, size, sharpness.

Optionally, the method further includes: pre-save image information of each video person, and when it is required to adjust image information of one or more video personnel, determine the image according to image information of a pre-saved video person. A video person in the video image; the image information of the video person in the video image is adjusted according to a preset fine adjustment policy of each video person.

An embodiment of the present invention further provides an apparatus for implementing video image processing, including: a determining unit and an adjusting unit; wherein

The determining unit is configured to acquire image information of the video image, and determine an adjustable object in the image information according to the image matching information stored in advance;

The adjusting unit is configured to separately adjust each of the determined adjustable objects according to a preset adjustment strategy.

Optionally, the determining unit is further configured to pre-store the image matching information, where the image matching information includes: matching condition information for matching one or more objects: a video scene, a video personnel, Video item

The determining unit is configured to acquire image information of the video image, match the image information with pre-stored image matching information, and use an object that matches the matching condition information in the image information as an adjustable object.

Optionally, the adjusting unit is configured to adjust each of the determined adjustable objects according to the following adjustment policies:

Optionally, the device further includes a fine adjustment unit configured to pre-save image information of each video person, and when the image information of one or more video personnel needs to be adjusted, according to the image information of the pre-saved video person, Determining a video person in the video image; adjusting image information of a video person in the video image according to a preset fine adjustment policy of each video person.

The embodiment of the invention further provides a computer readable storage medium storing program instructions, which can be implemented when the program instructions are executed.

Compared with the related art, the technical solution provided by the embodiment of the present invention includes: acquiring image information of a video image, determining an adjustable object in the image information according to pre-stored image matching information; and determining respectively according to a preset adjustment strategy. Each adjustable object is adjusted. In the embodiment of the present invention, after the image information of the video image is matched and analyzed, and the adjustable object is determined, the image information is adjusted according to the preset adjustment strategy, so that the part of the video image that affects the display effect of the adjustable object is adjusted and improved. The display effect of the video image improves the user experience of the video image application.

Other aspects will be apparent upon reading and understanding the drawings and detailed description.

BRIEF abstract

1 is a flowchart of a method for implementing video image processing according to an embodiment of the present invention;

2 is a schematic diagram of implementing video image processing in Application Example 1 of the present invention;

3 is a schematic diagram of implementing video image processing in Application Example 2 of the present invention;

FIG. 4 is a structural block diagram of an apparatus for implementing video image processing according to an embodiment of the present invention.

Embodiments of the invention

The embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be noted that, in the case of no conflict, the features in the embodiments and the embodiments of the present invention may be arbitrarily combined with each other.

1 is a flowchart of a method for implementing video image processing according to an embodiment of the present invention. As shown in FIG. 1, the method includes:

Step 100: Acquire image information of a video image, and determine an adjustable object in the image information according to the image matching information stored in advance;

In this step, the image matching information includes: matching condition information for matching one or more objects: a video scene, a video person, and a video item;

The determining the adjustable object in the image information comprises:

Since the image information needs to be adjusted, the acquired image information is generally stored in a device between the acquired image information and the displayed image information (such as some devices that buffer the video image), for example, the video conference saves the image information in In a multipoint control unit (MCU);

In this step, when the image matching information is the matching condition information of the video person, the matching object of the video image is determined to be the video person mainly through the matching method of the face recognition; and when the image matching information is the matching condition information of the video item, Then, the normal object attribute feature or the artificially input item attribute feature determines the attribute characteristics of a common item such as a water cup, a phone, a desk lamp, a notebook, etc. to determine that the adjustable object in the video image is a video item.

Optionally, during the video conference, if garbage or other excessive items appear in the actual image information, the video personnel's evaluation of the video activity participants will be affected, so the video items are still Including garbage, clothing, etc.

In this embodiment, a corresponding database may be established for image matching information such as a video scene, a video person, and a video item for storage.

Step 101: Adjust each of the determined adjustable objects according to a preset adjustment strategy.

In this step, the adjustment policy is set separately for each type of adjustable object described above, including:

When the adjustable object is a video scene, parameter adjustment is performed on the adjustable object, and the parameter includes one or more of the following: brightness, tone value, contrast value, saturation;

When the adjustable object is a video person, demodulate or replace the following one or more features of the adjustable object: facial features, skin color, hairstyle, glasses, clothing;

Taking a video conference as an example, for a site scene, the adjustment strategy includes preset brightness, tone value, contrast value, saturation, and other adjustment parameter information, and the parameter information of the image information in the video image is obtained, that is, the actual brightness and tone value of the site. , contrast value, saturation and other parameters, according to the set adjustment parameter information to adjust the image information to the corresponding value; for the participants, according to the preset adjustment strategy for skin color, hair, clothing, etc., the participants in the video The image information in the image is adjusted, for example, the participant's hair style personality can be replaced with a more mature image according to the adjustment strategy adjustment. When the facial features of the participant include noise, the sample coverage adjustment process can be performed; If an item, such as a wallet, a paper towel, or the like, affects the user experience of the video conference image display, the adjustment strategy may be set in advance to perform blurring, replacement, and even deletion processing.

The method of the embodiment of the present invention further includes: pre-storing image information of each video person, and when it is necessary to adjust image information of one or more video personnel, determining the video image according to image information of a pre-stored video person. The video personnel in the video; the image information of the video person in the video image is adjusted according to a preset fine adjustment policy of each video person.

It should be noted that the fine tuning policy refers to the corresponding adjustment policy set for each video personnel, and the adjusted policies can be set according to actual needs for different video personnel.

The method of the embodiment of the present invention is described below by using an application example. The application example is only for illustrating the embodiment of the present invention, and is not intended to limit the embodiment of the present invention. In the embodiment, the video conference is used as the actual application scenario. In the application environment of video recording, video live broadcast and network video, the same can be implemented. The device structure and processing process involved can be adjusted according to actual use.

Application example 1

FIG. 2 is a schematic diagram of implementing video image processing in this example, as shown in FIG. 2, including:

First, the image information of the video image is acquired, and the adjustable object in the image information is determined according to the image matching information stored in advance; the video image obtained here is sent to the central control device of the video conference system, and the image information is obtained through decoding, and the image is pre-stored. Match the information and determine that the scene is the site. According to the preset number, the number is 1;

Obtain the parameter information of the site scenario, and organize the parameter information through the list mode, including the following entries:

Object entry name: site

Number: 1;

Brightness value: 20;

Tone value: 30;

Contrast value: 40;

Saturation: 30.

Adjust the site scenario of site 1 according to the corresponding adjustment parameters and the adjustment policy corresponding to number 1. The adjustment strategy here is to adjust the parameters of the site.

For example, suppose the adjustment parameters corresponding to Site 1 are:

Object entry name: Site 1 adjustment parameters

Corresponding number: 1;

Adjust the brightness value to: 50;

Adjust the tonal value to: 40;

Adjust the contrast value to: 55;

Adjust the saturation value to: 45;

According to the adjustment parameter item corresponding to the site scenario, the parameter adjustment direction and the resizing are obtained by adjusting the parameter and the obtained actual parameter information, so as to adjust the image information to the preset adjustment parameter. That is, the brightness is increased by 30; the color tone is adjusted to be 10; the contrast is adjusted to 15; the saturation is increased by 15; after the adjustable image information is adjusted, the adjusted video image is sent for display, that is, sent to the video. Play on the display terminal of the conference.

Application example 2

FIG. 3 is a schematic diagram of implementing video image processing in this example, as shown in FIG. 3, including:

First, the image information of the video image is obtained, and the adjustable object in the image information is determined according to the image matching information stored in advance; and the pre-stored image matching information may determine that the image information includes two participants;

In this embodiment, the parameter information of the participant includes: the color of the participant, the hairstyle, whether the glasses are worn, and the clothing;

According to the preset adjustment strategy, the facial features, skin color, hairstyle, wearing glasses and clothing of the participants are adjusted, wherein the adjustment of the skin color and the hairstyle includes noise processing;

It is assumed that in the acquired image information, the parameter information of the participant and the facial feature information include noise, and the noise appearing on the face is covered according to a preset adjustment strategy, for example, by detecting normal image information near the noise. After sampling, adjust the coverage to the area where the noise is located for noise removal.

After the adjustable image information is adjusted, the video image after the image information is adjusted is displayed, that is, sent to the display terminal of the video conference for playback.

Application example 3

First, acquiring image information of the video image, determining an adjustable object in the image information according to the image matching information stored in advance; and determining, from the pre-stored image matching information, that the video item exists, including a large trash can;

In this embodiment, the parameter information of the trash can includes: the size, location, and definition of the trash can;

Adjust the item according to the preset adjustment strategy. If the trash can is small and fuzzy, you can reduce the sharpness of the image according to the preset adjustment strategy. If the trash can is large, the position is obvious. The image information of the displayed trash can is deleted according to a preset adjustment strategy, that is, the area where the trash can is removed is removed, and the surrounding image is filled for sampling;

The image after removing the object is filled by sampling the image of the periphery of the removed item.

4 is a structural block diagram of an apparatus for implementing video image processing according to an embodiment of the present invention. As shown in FIG. 4, the method includes: a determining unit 401 and an adjusting unit 402;

The determining unit 401 is configured to acquire image information of the video image, and determine an adjustable object in the image information according to the image matching information stored in advance;

Optionally, the determining unit 401 is further configured to pre-store the image matching information, including: matching condition information for matching one or more objects: a video scene, a video person, a video item;

The determining unit 401 is configured to acquire image information of the video image, match the image information with pre-stored image matching information, and use an object that matches the matching condition information in the image information as an adjustable object.

The adjusting unit 402 is configured to respectively adjust each of the determined adjustable objects according to a preset adjustment strategy.

Optionally, the adjusting unit 402 is configured to:

When the determined adjustable object includes a video scene, parameter adjustment is performed on the adjustable object, wherein the parameter includes one or more of the following: brightness, hue value, contrast value, saturation;

Determining or replacing one or more of the following features of the adjustable object, facial features, skin color, hairstyle, glasses, clothing when the determined adjustable object includes a video person;

When the determined adjustable object includes a video item, overwriting the adjustable object or Adjusting one or more of the following characteristics of the adjustable object: position, size, and definition;

Optionally, the apparatus of the embodiment of the present invention further includes a fine adjustment unit configured to pre-save image information of each video person, and when the image information of one or more video personnel needs to be adjusted, according to the pre-saved video personnel Image information, determining a video person in the video image; adjusting image information of a video person in the video image according to a preset fine-tuning policy of each video person.

One of ordinary skill in the art will appreciate that all or a portion of the above steps may be accomplished by a program that instructs the associated hardware, such as a read-only memory, a magnetic disk, or an optical disk. Alternatively, all or part of the steps of the above embodiments may also be implemented using one or more integrated circuits. Correspondingly, each module/unit in the foregoing embodiment may be implemented in the form of hardware or in the form of a software function module. Embodiments of the invention are not limited to any specific form of combination of hardware and software.

Industrial applicability

In the embodiment of the present invention, after the image information of the video image is matched and analyzed, and the adjustable object is determined, the image information is adjusted according to the preset adjustment strategy, so that the part of the video image that affects the display effect of the adjustable object is adjusted and improved. The display effect of the video image improves the user experience of the video image application.

Claims

A method for implementing video image processing, comprising:

Obtaining image information of the video image, and determining an adjustable object in the image information according to the image matching information stored in advance;

Each of the determined adjustable objects is adjusted according to a preset adjustment strategy.
The method of claim 1, wherein the image matching information comprises: matching condition information for matching one or more of the following objects: a video scene, a video person, a video item;

The determining the adjustable object in the image information comprises:

The image information is matched with the image matching information stored in advance, and an object that matches the matching condition information in the image information is taken as an adjustable object.
The method according to claim 1 or 2, wherein the preset adjustment strategy comprises:

When the adjustable object is a video scene, parameter adjustment is performed on the adjustable object, wherein the parameter includes one or more of the following: brightness, tone value, contrast value, saturation;

Demodulating or replacing one or more of the following features of the adjustable object when the adjustable object is a video person: facial features, skin color, hairstyle, glasses, clothing;

When the adjustable object is a video item, the adjustable object is overlaid or one or more of the following features of the adjustable object are adjusted: position, size, sharpness.
The method according to claim 1 or 2, further comprising pre-storing image information of each video person, when the image information of one or more video personnel needs to be adjusted, according to the pre-saved video personnel Image information, determining a video person in the video image; adjusting image information of a video person in the video image according to a preset fine-tuning policy of each video person.
An apparatus for implementing video image processing, comprising: a determining unit and an adjusting unit, wherein

The determining unit is configured to acquire image information of the video image, and determine an adjustable object in the image information according to the image matching information stored in advance;

The adjusting unit is configured to separately adjust each of the determined adjustable objects according to a preset adjustment strategy.
The apparatus according to claim 5, wherein said determining unit is further configured to pre-store said image matching information, said image matching information comprising: matching condition information for matching one or more of the following objects : video scenes, video personnel, video items;

The determining unit is configured to acquire image information of the video image, match the image information with pre-stored image matching information, and use an object that matches the matching condition information in the image information as an adjustable object.
The apparatus according to claim 5 or 6, wherein the adjustment unit is configured to adjust each of the determined adjustable objects according to the following adjustment strategies:

When the adjustable object is a video scene, parameter adjustment is performed on the adjustable object, wherein the parameter includes one or more of the following: brightness, tone value, contrast value, saturation;

Demodulating or replacing one or more of the following features of the adjustable object when the adjustable object is a video person: facial features, skin color, hairstyle, glasses, clothing;

When the adjustable object is a video item, the adjustable object is overlaid or one or more of the following features of the adjustable object are adjusted: position, size, sharpness.
The apparatus according to claim 5 or 6, further comprising a fine adjustment unit configured to pre-store image information of each video person, and when it is necessary to adjust image information of one or more video personnel, according to pre-preservation The image information of the video person determines the video person in the video image; and adjusts the image information of the video person in the video image according to a preset fine adjustment policy of each video person.
A computer readable storage medium storing program instructions that, when executed, can implement the method of any of claims 1-4.