WO2013107146A1

WO2013107146A1 - Video identification technology-based method and system for providing value-added service

Info

Publication number: WO2013107146A1
Application number: PCT/CN2012/076672
Authority: WO
Inventors: 邢晓江
Original assignee: 中兴通讯股份有限公司
Priority date: 2012-01-17
Filing date: 2012-06-08
Publication date: 2013-07-25
Also published as: CN102543125B; CN102543125A

Abstract

Disclosed is a video identification technology-based method for providing a value-added service, comprising: identifying video data; storing the identification result by means of a formatted markup text; and providing one or more services among the following on the basis of the identification result: a quick indexing service, a storage space saving service, and an information service. Correspondingly, also disclosed is a video identification technology-based system for providing the value-added service. The present invention allows for reduced storage space of the video data in a video surveillance system and for facilitated monitoring, indexing, and sharing of the video data.

Description

Method and system for providing value-added service based on video recognition technology

The present invention relates to video processing technologies, and in particular, to a method and system for providing value-added services based on video recognition technology. Background technique

Video surveillance is a common application in the Internet of Things system. It plays a very important role in people's lives and work. For example, the city security video surveillance system as a safe city construction, public security "scientific and technical police" and public security information An important part of the construction has been highly valued by the State Council, the Ministry of Public Security, local governments at all levels, and public security organs. In just a few years, the world’s largest and most widely covered video surveillance system has been built. It has made a good contribution to improving the security situation of the city, strengthening urban management, and creating a harmonious society.

As of the end of 2010, approximately 20 million cameras were deployed nationwide, of which more than 3 million were deployed by the public security department. There are also some problems with such a huge video surveillance system, mainly in:

1) Huge storage pressure. The resolution of the camera is about 1.2 million to 2 million pixels. Each camera generates about 100M of video data per hour. Ten thousand cameras produce 24TB of data every day. Although there are some intelligent compression technologies, the storage pressure is extremely high. .

2) A large number of people are required to watch manually, and the retrieval is difficult. These videos are currently monitored mainly by manual viewing. In a city in southern China, 500,000 cameras have been deployed, of which 230,000 are deployed by the public security department, and 1,800 people have to be hired to watch these cameras. The search is difficult, almost entirely through manual viewing.

3) Cross-domain sharing and joint coordination are difficult. Based on existing video surveillance methods, information sharing and joint coordination between systems and systems, between departments and departments, between cities and cities is difficult. 4) It is difficult to link with other systems. Based on the existing video surveillance system, it is difficult to achieve linkage with other systems in the city. Because of the difficulty in information sharing, the video surveillance system has become an "information island." Summary of the invention

In view of this, the main purpose of the embodiments of the present invention is to provide a method and system for providing value-added services based on video identification technology, which can reduce the storage space of video data in applications such as video surveillance, and facilitate video data monitoring, retrieval, and sharing.

In order to achieve the above object, the technical solution of the present invention is achieved as follows:

A method for providing value-added services based on video recognition technology, including:

Identify video data;

The recognition result is stored in a text formatted by the format ^;

One or more of the following services are provided based on the recognition result: a quick retrieval service, a storage space service, and an information service.

The recognition result is stored in a database, and the providing fast retrieval service is:

Searching in the database based on the entered query criteria;

The search results are provided to the user for viewing.

The provided storage space saving service is:

Delete or store only the recognition result of the video content that meets the preset condition.

The information providing service is:

The recognition result is provided to the preset service, and/or, when the recognition result satisfies the trigger condition, the corresponding operation is performed according to the preset trigger logic.

The method also includes: optimizing the identification of the video data.

The method further includes: controlling the recognition process according to a user operation or modifying the recognition result.

A system for providing value-added services based on video recognition technology, comprising: a pre-processing module and an application module; wherein The pre-processing module is configured to identify the video data, and store the recognition result in a formatted text manner;

The application module is configured to provide one or more of the following services based on the identification result stored by the pre-processing module: a fast retrieval service, a storage space saving service, and an information service.

The system also includes an optimization module.

The optimization module is configured to optimize the recognition result of the pre-processing module.

The system also includes a manual intervention module.

The manual intervention module is configured to implement interaction with the user, control the recognition process according to user operations, or modify the recognition result.

The system also includes a video capture module.

The video capture module is configured to collect video data for identification by the pre-processing module; the optimization module is further configured to optimize the video capture module;

The manual intervention module is further configured to control the video collection module according to a user operation. Embodiments of the present invention provide a method and system for providing value-added services based on video recognition technology, identify video by video recognition technology, mark various information identified by a formatting language, and then perform a database or other structured manner. Stored, structured tagged information that provides a variety of value-added services (such as video surveillance). The solution described in the embodiment of the present invention can reduce the storage space of video data in the video surveillance system, and is beneficial to video data monitoring, retrieval, and sharing. DRAWINGS

1 is a schematic flowchart of a method for providing a value-added service based on a video recognition technology according to an embodiment of the present invention;

2 is a schematic flowchart of providing a fast retrieval service according to an embodiment of the present invention;

FIG. 3 is a schematic flowchart of providing a storage space saving service according to an embodiment of the present invention; FIG. 4 is a schematic flowchart of providing an information service according to an embodiment of the present invention. detailed description

The basic idea of the embodiment of the present invention is: Identifying a video by a video recognition technology, marking the identified various information through a formatting language, and optimizing the recognition through an optimization and manual intervention module, and then based on the database Or structured information stored in other structured ways to provide a variety of value-added services.

It should be noted that the video recognition technology analyzes, extracts and features the individual video from the video by designing a certain computer algorithm to determine whether it meets certain rules, whether it belongs to some situations that need to be processed, and reminds.

At present, the data captured by the camera mainly includes: information on people, vehicles, traffic, traffic flow, crowd convergence, etc., which can be technically identified.

The main types of services currently available include:

Video recognition service: Identify specific objects, such as faces, license plates, and photos for comparison;

Video Analysis Services: Provide services such as perimeter intrusion detection, object motion monitoring, traffic flow, traffic statistics, PTZ (Pan/Tilt/Zoom) automatic tracking, etc.

Video Improvement Service: Provides night vision image enhancement processing, image stabilization system and other functions. The embodiment of the invention provides a method for providing value-added services based on video recognition technology. As shown in FIG. 1, the method includes:

Step 101: Identify video data;

Here, the identification of the video data may be further optimized, and/or the recognition process may be controlled according to the user operation or the recognition result may be modified to improve the recognition accuracy.

It should be noted that the video data may be real-time or offline video data, and the specific video format is not limited. When the video content is recognized, the car, the person, and the status information may be identified according to preset requirements. Information such as vehicle color, vehicle type, and license plate is recognized for the vehicle; information on bicycles, electric vehicles, and motorcycles is identified; people recognize information such as height, men, women, and clothing; The information identifies information such as crowd gathering, congestion, and car accidents.

Step 102: Store the recognition result in the formatted text.

For example, a format language can describe vehicle information as:

Time, camera, video location, location, vehicle model, license plate, vehicle color, speed, lane, violations...

The formatted language can describe the person information as:

Time, camera, video location, location, men and women, height, clothes situation, pants situation, hat condition, speed, hand held items...

The formatting information can be described as a formatted language:

Time, camera, video location, location, crowd gathering, vehicle congestion, car accident... It should be noted that the database can be selected to store the formatted tag information.

Step 103: Provide one or more of the following services based on the identification result: a fast retrieval service, a storage space saving service, and an information service.

Specifically, the providing the fast retrieval service may be: performing a retrieval in a database (for example, performing a SQL retrieval); and providing the retrieval result to the user for viewing.

As shown in FIG. 2, an embodiment of a process for providing a fast retrieval service includes:

Enter the query criteria;

Retrieve the database;

According to the search result, the related video is directly located, and the search result is provided to the user for viewing. For example, when public security personnel need to retrieve a certain type of car that escapes within a certain period of time, in order to locate the suspect, SQL retrieval is performed on the text of the formatted mark stored in the database. , you can easily locate specific video images, police officers can directly watch for confirmation and identification. At the same time, it is also convenient to implement continuous inquiry of the vehicle to determine its driving route, which is convenient for police officers to trace.

Specifically, the providing the storage space saving service may be: deleting or storing only the preset The recognition result of the conditional video content. For example, in the middle of the night, in the absence of people or vehicles, that is, without information, the video of the time period can be deleted to save space, and the video content of the key points can also be recorded as a photo. This also saves a lot of storage space.

As shown in FIG. 3, the process of providing a storage space saving service includes: saving a picture containing information according to the stored information;

After storing the picture, adding a retrieval record;

Delete the original video data.

Specifically, the information service mainly includes an information providing service and an information triggering service. In other words, the providing the information service may be: providing the identification result to the preset service, and/or, when the recognition result satisfies the trigger condition, according to the preset The trigger logic performs the corresponding action. For example, the identified information may be provided as a source of information to services such as: auto-navigation, vehicle grooming, etc., and trigger logic may be set. When the recognized information satisfies the trigger condition, the processing logic in the trigger logic is executed, such as : Warning, alarm, linkage and other services.

As shown in FIG. 4, an embodiment of a process for providing an information service includes:

Develop an information triggering strategy;

Load trigger policy;

The trigger condition is met, and the related operation is performed according to the trigger policy.

It should be noted that, in the embodiment of the present invention, the parameters related to video collection may be optimized, and/or the video collection process may be controlled according to user operations.

The embodiment of the present invention further provides a system for providing a value-added service based on a video recognition technology, where the system includes: a pre-processing module and an application module;

The pre-processing module is configured to identify the video data, and store the recognition result in a formatted text manner;

The application module is configured to provide the following one based on the recognition result stored by the preprocessing module Or a variety of services: fast retrieval services, storage space services, information services.

Optionally, the system further includes an optimization module,

The optimization module is configured to optimize the recognition result of the pre-processing module to improve the recognition accuracy. For example, in order to improve the image recognition accuracy, the image may be optimized according to the background, brightness, and weather conditions of the camera.

The system also includes a manual intervention module.

The manual intervention module is configured to implement interaction with the user, control the recognition process according to the user operation, or modify the recognition result to achieve the purpose of automatic re-optimization of the system, and improve the recognition accuracy rate. For example, when an operator finds that an object is not accurately identified, the effect of the recognition can be corrected.

The system also includes a video capture module.

The manual intervention module is further configured to control the video collection module according to a user operation. The markup language describes various content information recognized by the preprocessing module. For example, the pre-processing module can identify the car, person, and status information in the video content, identify the vehicle color, the model, the license plate, and the like, and identify the information of the bicycle, the electric vehicle, and the motorcycle for the vehicle; Information such as height, men and women, clothing, etc. are identified; in addition, status information such as crowd gathering, congestion, and car accidents can be identified. Here, the pre-processing module may store the identified information in a database manner to provide to the application module.

The main interfaces of the above systems include:

1) An interface between the pre-processing module and the video acquisition module to enable reading of video data.

2) optimizing the interface between the module and the pre-processing module, and the optimization module provides information such as background, shading, weather conditions, etc. of the camera to be optimized to the pre-processing module through the interface; When the image data does not meet the requirements of the preprocessing, the preprocessing module can provide optimization to the optimization module through the interface, and the result of the preprocessing can also be provided to the optimization module through the interface to support optimization.

3) The interface between the manual intervention module and the pre-processing module realizes the manual intervention of the recognition result or the process. For example, when the operator sees that the recognition result is incorrect, the identification result is forcibly modified through the interface, so that the pre-processing module is improved. Identifying the success rate, and allowing the pre-processing module to re-optimize according to the corrected result; in addition, when the image data does not meet the requirements of the pre-processing, the pre-processing module can provide manual intervention to the manual intervention module through the interface, pre-processing The results can also be provided to the manual intervention module through the interface to support manual intervention;

4) The interface between the pre-processing module and the application module, that is, the stored interface of the identified structured data to the application, which may be an open interface of the database and supports SQL.

5) optimizing the interface between the module and/or the manual intervention module and the video acquisition module, so that the optimization module and/or the manual intervention module can adjust the resolution of the camera, or the parameters of the pan/tilt, camera direction, and angle of view through the interface, Video data is easier to identify. Marking, and then providing various value-added services based on structured markup information stored in a database or other structured manner, can reduce the storage space of video data in applications such as video surveillance, and facilitate video data monitoring and retrieval. shared.

The above is only the preferred embodiment of the present invention and is not intended to limit the scope of the present invention.

Claims

Claim

A method for providing value-added services based on video recognition technology, wherein the method comprises: identifying video data;

The recognition result is stored in a text formatted by the format ^;

2. The method of claim 1, wherein the recognition result is stored in a database, the providing fast retrieval service is:

Searching in the database based on the entered query criteria;

The search results are provided to the user for viewing.

The method according to claim 1, wherein the providing a storage space saving service is: deleting or storing only the recognition result of the video content that satisfies the preset condition.

The method according to claim 1, wherein the providing the information service is: providing the recognition result to the preset service, and/or, when the recognition result satisfies the trigger condition, performing the corresponding operation according to the preset trigger logic .

5. The method according to any one of claims 1 to 4, wherein

The method also includes: optimizing the identification of the video data.

6. The method according to any one of claims 1 to 4, wherein

7. A system for providing value-added services based on video recognition technology, wherein the system comprises: a pre-processing module and an application module; wherein

8. The system according to claim 7, wherein the system further comprises an optimization module, the optimization module being configured to optimize the recognition result of the pre-processing module.

9. The system of claim 7 or 8, wherein the system further comprises a manual intervention module,

The system of claim 9, wherein the system further comprises a video capture module, the video capture module is configured to collect video data for recognition by the pre-processing module; and the optimization module is further configured as a video capture module. optimize;

The manual intervention module is further configured to control the video collection module according to a user operation.