CN103167265A - Video processing method and video processing system based on intelligent image identification - Google Patents

Abstract

The invention relates to a video processing method based on intelligent image identification. The video processing method based on the intelligent image identification comprises that video information which is collected in real time or recorded by a peripheral unit (PU) is input to an intelligent video unit (IVU) by a monitoring platform; the IVU carries out image analysis and identification of a monitoring object for the video information according to predetermined monitoring rules so as to form an identification result; the IVU sends the identification result to an intelligent video management (IVM) server; the IVM server sends the identification result back to the monitoring platform; and the monitoring platform sends the identification result to the PU so that the PU can overlay the identification result on video pictures corresponding to the video information in an additional intelligent information mode, and then the video pictures are numbered again and stored. The invention further relates to a video processing system based on the intelligent image identification. According to the video processing method and the video processing system, the intelligent processing capability of the IVU and the PU are combined, the identification result obtained through the analysis and the identification by the IVU is provided for the PU, and the PU is used for carrying out overlaying processing and storage of the intelligent information, and therefore a user can conveniently obtain the video information with the intelligent information from the PU.

Description

Method for processing video frequency and system based on intelligent image identification

Technical field

The present invention relates to the intelligent video monitoring applied technical field, relate in particular to a kind of method for processing video frequency and system based on intelligent image identification.

Background technology

Along with the development of Video Supervision Technique, Video Supervision Technique is widely used in every field, has produced the video resource of magnanimity thereupon.For these monitor video resources, most monitor supervision platforms in the past still adopt the mode of artificial treatment to process, and the mode of this artificial treatment is larger to the waste of manpower and materials, is difficult in the sizable application of commercialization product formation.

Intelligent monitoring technology has become the important development direction of video monitoring at present, under the implementation pattern of intelligent monitoring, by realizing certain image intelligent analysis ability by application module in the monitor supervision platform deploy, make the video monitoring automation and intelligentification, thereby overcome the many disadvantages of artificial treatment mode.

But remain in present intelligent monitoring scheme in following problem: the deployment of concentrating on monitor supervision platform at present image intelligent disposal ability and data storage capacities, although this mode is given in management and has been brought facility, but for the supervisor, if wish, the video content after Intelligent treatment is browsed, can only be by obtain the mode of data to platform, therefore tend to be subject to certain restriction because of the factor such as network, terminal, possible problem is as follows:

1, the video information that comprises the intelligent use data can only just can be obtained by platform, and if in time do not remove in a long time the platform gathering information, may remove this part record by platform;

2, when obtaining video content, the supervisor may directly obtain corresponding video from front end and carry out playback and inspection, but because front end does not possess the intellectual analysis function, therefore can't show intelligent content, makes troubles to the supervisor.

Summary of the invention

The objective of the invention is to propose a kind of method for processing video frequency and system based on intelligent image identification, can make the user in the past end unit obtain easily video information with intelligent information.

For achieving the above object, the invention provides a kind of method for processing video frequency based on intelligent image identification, comprising:

Monitor supervision platform is recorded to intelligent image recognition unit input front end unit or the video information of Real-time Collection;

Described intelligent image recognition unit carries out the identification of graphical analysis and monitored object according to default monitoring rules to described video information, form recognition result;

Described intelligent image recognition unit sends to the intelligent image management server with described recognition result;

Described intelligent image management server returns to described monitor supervision platform with described recognition result;

Described monitor supervision platform sends to described front end unit with described recognition result, so that described front end unit is superimposed upon the mode of described recognition result with additional intelligent information in video pictures corresponding to described video information, and stores after recompile.

Further, described intelligent image recognition unit is specially according to the operation that default monitoring rules carries out the identification of graphical analysis and monitored object to described video information:

Described intelligent image recognition unit carries out graphical analysis to described video information, forms corresponding image, semantic and describes;

Described intelligent image recognition unit is described according to described image, semantic, identifies the relevant information of monitored object according to default monitoring rules.

Further, record to intelligent image recognition unit input front end unit at described monitor supervision platform or the video information of Real-time Collection after, also comprise:

Described intelligent image recognition unit sends to described intelligent image management server with the primary statistics information of the movement of objects situation in described video information;

Described intelligent image management server returns to described monitor supervision platform with described primary statistics information;

Described monitor supervision platform sends to described front end unit with described primary statistics information, so that described front end unit is superimposed upon the mode of described primary statistics information with additional intelligent information in video pictures corresponding to described video information, and stores after recompile.

Further, also comprise: described front end unit receives the video request of being sent by client unit, and the video information that comprises intelligent information of storing is transmitted to described client unit via described monitor supervision platform.

For achieving the above object, the present invention also provides a kind of processing system for video based on intelligent image identification, comprises monitor supervision platform, intelligent image recognition unit, front end unit and intelligent image management server, wherein,

Monitor supervision platform, described front end unit is recorded or the video information of Real-time Collection for inputting to described intelligent image recognition unit, and when receiving recognition result, described recognition result is sent to described front end unit;

The intelligent image recognition unit is used for described video information being carried out the identification of graphical analysis and monitored object according to default monitoring rules, forms recognition result, and described recognition result is sent to described intelligent image management server;

The intelligent image management server is used for described recognition result is returned to described monitor supervision platform;

Front end unit is used for described recognition result is superimposed upon video pictures corresponding to described video information in the mode of additional intelligent information, and stores after recompile.

Further, described intelligent image recognition unit specifically comprises:

The intelligent image analytic unit is used for described video information is carried out graphical analysis, forms corresponding image, semantic and describes;

The monitored object recognition unit is used for describing according to described image, semantic, identifies the relevant information of monitored object according to default monitoring rules;

The first transmitting element is used for described recognition result is sent to described intelligent image management server.

Further, described intelligent image recognition unit also comprises:

The movement of objects statistic unit is used for the movement of objects situation of described video information is added up, and obtains primary statistics information;

The second transmitting element is used for described primary statistics information is returned to described monitor supervision platform.

Further, described front end unit specifically comprises:

The supercomposed coding unit is used for described recognition result and/or primary statistics information are superimposed upon video pictures corresponding to described video information in the mode of additional intelligent information, and recompile;

The video storage unit is used for storing the video information after described recompile.

Further, described front end unit also comprises:

The video request response unit is used for receiving the video request of being sent by client unit, and the video information that comprises intelligent information of storing is transmitted to described client unit via described monitor supervision platform.

Based on technique scheme, the present invention combines intelligent processing capacity and the front end unit of intelligent image recognition unit, the recognition result that the analysis of intelligent image recognition unit and identification are obtained offers front end unit, carried out the video superimpose of these intelligent information by front end unit and process and preserve, so make the user in the past end unit can obtain easily video information with intelligent information.

Description of drawings

Accompanying drawing described herein is used to provide a further understanding of the present invention, consists of the application's a part, and illustrative examples of the present invention and explanation thereof are used for explaining the present invention, do not consist of improper restriction of the present invention.In the accompanying drawings:

Fig. 1 is a kind of network architecture diagram of application scenarios that the present invention is based on the method for processing video frequency embodiment of intelligent image identification.

Fig. 2 is the schematic flow sheet of an embodiment that the present invention is based on the method for processing video frequency of intelligent image identification.

Fig. 3 is the signaling process schematic diagram of another embodiment that the present invention is based on the method for processing video frequency of intelligent image identification.

Fig. 4 is the structural representation of an embodiment that the present invention is based on the processing system for video of intelligent image identification.

Fig. 5 is the concrete structure schematic diagram that the present invention is based on intelligent image recognition unit in another embodiment of processing system for video of intelligent image identification.

Fig. 6 is the concrete structure schematic diagram that the present invention is based on front end unit in the another embodiment of processing system for video of intelligent image identification.

Embodiment

Below by drawings and Examples, technical scheme of the present invention is described in further detail.

As shown in Figure 1, be the network architecture diagram of a kind of application scenarios of the method for processing video frequency embodiment that the present invention is based on intelligent image identification.in this application scenarios, center management server 1 (Central Management Server, be called for short CMS) respectively with intelligent image management server 2 (Intelligent Video Management, be called for short IVM), intelligent image recognition unit 3 (Intelligent Video Unit, be called for short IVU), video turns Dispatching Unit 7 (Video Transfer and Distribute Unit, be called for short VTDU), network video recording unit 6 (Network Recode Unit, be called for short NRU), client unit 4 (Client Unit, be called for short CU) and front end unit 5 (Peripheral Unit, be called for short PU) be connected, other annexations are referring to Fig. 1.Monitor supervision platform can be made ofCMS 1, VTDU 7 and NRU 6 etc., but is not limited to several equipment shown in figure, allows minimizing equipment or increases new equipment.

Fig. 1 introduces the possible application scenarios of method for processing video frequency of the present invention, but those skilled in the art will appreciate that the application scenarios that provides in Fig. 1, also allows other possible application scenarioss and the corresponding network architecture.

As shown in Figure 2, be the idiographic flow schematic diagram of an embodiment of the method for processing video frequency that the present invention is based on intelligent image identification.In the present embodiment, the Video processing flow process based on intelligent image identification comprises the following steps:

Step 101, monitor supervision platform are recorded or the video information of Real-time Collection to IVU input PU;

Step 102, IVU carry out the identification of graphical analysis and monitored object according to default monitoring rules to video information, form recognition result;

Step 103, IVU send to IVM with recognition result, and IVM returns to monitor supervision platform with recognition result;

Step 104, monitor supervision platform send to PU with recognition result, so that PU is superimposed upon the mode of recognition result with additional intelligent information in video pictures corresponding to video information, and store after recompile.

In the present embodiment, monitor supervision platform may directly be initiated the video analysis request to IVU, can be also to answer the request of PU to come to initiate the video analysis request to IVU, can comprise video information or the video information of PU Real-time Collection, perhaps address information of these video informations etc. that PU has recorded in concrete video analysis request.

IVU can carry out the identification of graphical analysis and monitored object according to default monitoring rules to described video information, monitoring rules can be set by CU by the user, for example wishes certain type of vehicle of the A place process of 8 a.m. to 11 is analyzed.IVU can first carry out graphical analysis to video information, forms corresponding image, semantic and describes, and for example type is motorbus, and car plate is capital XXXXX etc., describes according to the image, semantic that obtains, and identifies the relevant information of monitored object according to default monitoring rules.

IVU sends to IVM with the recognition result that identifies, and via IVM, recognition result is returned to monitor supervision platform.This recognition result is as intelligent information, send to corresponding PU by monitor supervision platform again, this PU just can be superimposed upon the mode of this recognition result with additional intelligent information in video pictures corresponding to video information, preserve by recompile, thereby formed video information with intelligent information in PU.Synchronize with video pictures in the mode of similar captions from formal intelligent information and occur, facilitate the user to browse.

When the user utilizes CU to send video request, PU can be transmitted to CU via monitor supervision platform according to the video information that comprises intelligent information that this video request that receives will be stored, and utilize the function that presents of CU self to present with the video information of intelligent information, the process of consulting that this just has great convenience for the user.

After the IVU enter video information, IVU can also send to IVM with the primary statistics information of the movement of objects situation in video information (such as information such as the flow of the stream of people at crossing, road, wagon flow, flow velocitys) at monitor supervision platform.IVM can return to monitor supervision platform with this primary statistics information, monitor supervision platform sends to PU with primary statistics information, PU can be superimposed upon the mode of these primary statistics information with additional intelligent information in video pictures corresponding to video information, and stores after recompile.Further, PU can all be superimposed upon in the mode of additional intelligent information primary statistics information and the recognition result that obtains in video pictures corresponding to video information, thereby makes the video pictures amount of information larger, satisfies user's verification demand.

As shown in Figure 3, be the signaling process schematic diagram of another embodiment of the method for processing video frequency that the present invention is based on intelligent image identification.In the present embodiment, the Video processing flow process comprises:

Step 201, CU need the condition of the video superimpose class intelligent information of certain PU by monitor supervision platform setting, such as certain PU in the oversize vehicle of certain time period situation etc. of passing through at this crossing, PU place.

Step 202, monitor supervision platform send the intellectual analysis request to IVU, and this PU are recorded or the video information of Real-time Collection offers IVU.

Step 203, IVU carry out the identification of graphical analysis and monitored object according to default monitoring rules to described video information, form recognition result, also can add up the movement of objects situation in video information, obtain primary statistics information.

Step 204, IVU send to monitor supervision platform via IVM with recognition result (can be also primary statistics information).

Step 205, monitor supervision platform send to PU with recognition result (can be also primary statistics information).

Step 206, PU are superimposed upon the mode of recognition result (can be also primary statistics information) with additional intelligent information in video pictures corresponding to described video information, and store after recompile.

Step 207, CU can send video request to PU via monitor supervision platform, and the video information that comprises intelligent information that PU will store is transmitted to CU via monitor supervision platform.

One of ordinary skill in the art will appreciate that: all or part of step that realizes said method embodiment can be completed by the hardware that program command is correlated with, aforesaid program can be stored in a computer read/write memory medium, this program is carried out the step that comprises said method embodiment when carrying out; And aforesaid storage medium comprises: the various media that can be program code stored such as ROM, RAM, magnetic disc or CD.

As shown in Figure 4, be the structural representation of an embodiment of the processing system for video that the present invention is based on intelligent image identification.In the present embodiment, the processing system for video based on intelligent image identification comprises monitor supervision platform 8,IVU 3,PU 5 and IVM 2.Wherein monitor supervision platform 8 can be recorded or the video information of Real-time Collection toIVU 3input PU 5, and when receiving recognition result, described recognition result is sent to PU 5.The concrete structure of monitor supervision platform 8 also can be with reference to the structure that is made ofCMS 1, VTDU 7 andNRU 6 etc. shown in Figure 1.

IVU 3 is responsible for described video information being carried out the identification of graphical analysis and monitored object according to default monitoring rules, forms recognition result, and described recognition result is sent to IVM 2.IVM 2 is responsible for recognition result is returned to monitor supervision platform 8.PU 5 is responsible for the mode of described recognition result with additional intelligent information is superimposed upon in video pictures corresponding to described video information, and stores after recompile.

The present embodiment combines intelligent processing capacity and the PU of IVU, the recognition result that IVU analyzes and identification obtains is offered PU, carried out the video superimpose of these intelligent information by PU and process and preserve, and then make the user can obtain easily video information with intelligent information from PU

As shown in Figure 5, be the concrete structure schematic diagram of intelligent image recognition unit in another embodiment of the processing system for video that the present invention is based on intelligent image identification.In the present embodiment, IVU specifically comprises: intelligent image analytic unit 31, monitored object recognition unit 32 and the first transmitting element 33.Wherein, intelligent image analytic unit 31 is responsible for described video information is carried out graphical analysis, forms corresponding image, semantic and describes.Monitored object recognition unit 32 is responsible for describing according to described image, semantic, identifies the relevant information of monitored object according to default monitoring rules.The first transmitting element 33 is responsible for described recognition result is sent to IVM.

In another embodiment, IVU can further include: movement of objects statistic unit and the second transmitting element, wherein the movement of objects statistic unit is responsible for the movement of objects situation in described video information is added up, and obtains primary statistics information.The second transmitting element is responsible for primary statistics information is returned to described monitor supervision platform.

As shown in Figure 6, be the concrete structure schematic diagram of front end unit in the another embodiment of the processing system for video that the present invention is based on intelligent image identification.In the present embodiment,PU 5 specifically comprises: supercomposed coding unit 51 and video storage unit 52, wherein supercomposed coding unit 51 is responsible for described recognition result and/or the primary statistics information mode with additional intelligent information is superimposed upon in video pictures corresponding to described video information, and recompile.Video storage unit 52 is responsible for the video information after the described recompile of storage.

In another embodiment, PU can further include: the video request response unit, be used for receiving the video request of being sent by CU, and the video information that comprises intelligent information of storing is transmitted to CU via monitor supervision platform.

Should be noted that at last: above embodiment is only in order to illustrate that technical scheme of the present invention is not intended to limit; Although with reference to preferred embodiment, the present invention is had been described in detail, those of ordinary skill in the field are to be understood that: still can modify or the part technical characterictic is equal to replacement the specific embodiment of the present invention; And not breaking away from the spirit of technical solution of the present invention, it all should be encompassed in the middle of the technical scheme scope that the present invention asks for protection.

Claims (9)

1. method for processing video frequency based on intelligent image identification comprises:

Monitor supervision platform is recorded to intelligent image recognition unit input front end unit or the video information of Real-time Collection;

Described intelligent image recognition unit carries out the identification of graphical analysis and monitored object according to default monitoring rules to described video information, form recognition result;

Described intelligent image recognition unit sends to the intelligent image management server with described recognition result;

Described intelligent image management server returns to described monitor supervision platform with described recognition result;

Described monitor supervision platform sends to described front end unit with described recognition result, so that described front end unit is superimposed upon the mode of described recognition result with additional intelligent information in video pictures corresponding to described video information, and stores after recompile.

2. method according to claim 1, wherein said intelligent image recognition unit is specially according to the operation that default monitoring rules carries out the identification of graphical analysis and monitored object to described video information:

Described intelligent image recognition unit carries out graphical analysis to described video information, forms corresponding image, semantic and describes;

Described intelligent image recognition unit is described according to described image, semantic, identifies the relevant information of monitored object according to default monitoring rules.

3. method according to claim 2, wherein record to intelligent image recognition unit input front end unit at described monitor supervision platform or the video information of Real-time Collection after, also comprise:

Described intelligent image recognition unit sends to described intelligent image management server with the primary statistics information of the movement of objects situation in described video information;

Described intelligent image management server returns to described monitor supervision platform with described primary statistics information;

Described monitor supervision platform sends to described front end unit with described primary statistics information, so that described front end unit is superimposed upon the mode of described primary statistics information with additional intelligent information in video pictures corresponding to described video information, and stores after recompile.

4. according to claim 1～3 arbitrary described methods wherein also comprise:

Described front end unit receives the video request of being sent by client unit, and the video information that comprises intelligent information of storing is transmitted to described client unit via described monitor supervision platform.

5. the processing system for video based on intelligent image identification, comprise monitor supervision platform, intelligent image recognition unit, front end unit and intelligent image management server, wherein,

Monitor supervision platform, described front end unit is recorded or the video information of Real-time Collection for inputting to described intelligent image recognition unit, and when receiving recognition result, described recognition result is sent to described front end unit;

The intelligent image recognition unit is used for described video information being carried out the identification of graphical analysis and monitored object according to default monitoring rules, forms recognition result, and described recognition result is sent to described intelligent image management server;

The intelligent image management server is used for described recognition result is returned to described monitor supervision platform;

Front end unit is used for described recognition result is superimposed upon video pictures corresponding to described video information in the mode of additional intelligent information, and stores after recompile.

6. system according to claim 5, wherein said intelligent image recognition unit specifically comprises:

The intelligent image analytic unit is used for described video information is carried out graphical analysis, forms corresponding image, semantic and describes;

The monitored object recognition unit is used for describing according to described image, semantic, identifies the relevant information of monitored object according to default monitoring rules;

The first transmitting element is used for described recognition result is sent to described intelligent image management server.

7. system according to claim 6, wherein said intelligent image recognition unit also comprises:

The movement of objects statistic unit is used for the movement of objects situation of described video information is added up, and obtains primary statistics information;

The second transmitting element is used for described primary statistics information is returned to described monitor supervision platform.

8. system according to claim 7, wherein said front end unit specifically comprises:

The supercomposed coding unit is used for described recognition result and/or primary statistics information are superimposed upon video pictures corresponding to described video information in the mode of additional intelligent information, and recompile;

The video storage unit is used for storing the video information after described recompile.

9. according to claim 5-8 arbitrary described systems, wherein said front end unit also comprises:

The video request response unit is used for receiving the video request of being sent by client unit, and the video information that comprises intelligent information of storing is transmitted to described client unit via described monitor supervision platform.

Images