CN111160381A - Multi-information fusion target track tracking system - Google Patents

Abstract

The invention provides a multi-information fusion target track tracking system which comprises a data acquisition layer, a data fusion layer and a data application layer. In the data acquisition layer, the data uploaded by the data acquisition equipment is received, forwarded and stored according to needs and service settings, and the acquired data is distributed to different data and fusion servers; the data fusion layer is used for cleaning, clustering and fusing data and providing diversified service interfaces such as addition, deletion, modification, check and the like for the upper application layer; and the data application layer provides a response of display and interface operation for the client, and realizes the display of data according to time, target, place and individual requirements. The method forms the behavior track of the specific target through comprehensive judgment of various data, and makes up the defect that the target track is tracked by only depending on videos by utilizing the diversity of the data.

Description

Multi-information fusion target track tracking system

Technical Field

The invention belongs to the technical field of security protection, relates to a video monitoring technology, and particularly relates to a target track tracking system for multi-information fusion in a monitoring video.

Background

In the current security field, under various service scenes, behavior tracks of people need to be checked, paths of suspicious people are analyzed, and people with abnormal behaviors are searched. When people identify images, the people only depend on a video image processing technology, so that the limitation is great, and the color and the texture of clothes are easy to cause target confusion; the human face has good distinguishability, but the requirement on the quality of the image is high; in a simple monitoring scene, due to the fact that the installation position of a camera cannot be completely covered, the angle of the camera has a blind area, the light of an image acquisition area is too bright or too dark, or due to factors such as distance and the like, the acquired video image is limited to a great extent when being used for face recognition and target tracking, and therefore data from various different sources need to be integrated to perform information clustering and fusion, and comprehensive judgment is performed to form a behavior track of a target.

Disclosure of Invention

With the maturity of various data acquisition technologies and the development of data fusion and data analysis technologies, the invention enables the integration of various means and the comprehensive application of various data to track the target track to be possible.

The technical scheme of the invention is as follows:

a multi-information fused target track tracking system is characterized by comprising:

the data acquisition layer mainly acquires data of different places and different types, and comprises the following steps: monitoring video data including monitoring video original files and snap pictures of communities, roads, checkpoints and key area positions; the travel data comprises effective identity information, riding time, types of riding tools, getting-off places and getting-off time which are acquired when the public transport vehicles are ridden; the hotel accommodations collect the information of the check-in people, the information of the co-workers, the check-in time and the leaving time; communication data, wherein specific positions of specific time periods are given through mobile phone positioning and positioning information given by a base station of an operator;

the data fusion layer is mainly used for gathering and integrating the data of the acquisition layer, cleaning, clustering the data and fusing the data according to actual needs and providing a service interface for the application layer; the data aggregation can be physical aggregation of data or logical aggregation of data, and is real-time or periodically aggregated to a server for processing data in an interface or platform mode; data cleaning is a process of rechecking and verifying data through a certain cleaning rule, and aims to delete repeated information, correct existing errors and provide data consistency; data clustering, namely clustering data through a clustering algorithm and merging the data belonging to a certain object; data fusion, namely combining data, integrating data, aggregating data and the like, wherein the deep interaction among the data is realized step by step from low to high;

and the data display layer is mainly used for performing visualization display on the fused data, including map display, video display, picture display and comprehensive display, vividly displaying the behavior track of the attention target and providing an auxiliary function for subsequent judgment and decision making.

The behavior track of the specific target is formed through comprehensive judgment of various data, the defect that the target track is tracked by only depending on videos is overcome by using the diversity of the data, the hidden clue information can be found more favorably through the comprehensive analysis by combining the various data, and more accurate and rich reference basis is provided for subsequent research and decision processing.

Drawings

FIG. 1 is a technical framework diagram of the present invention;

fig. 2 is a system configuration diagram of the present invention.

Detailed Description

As shown in FIG. 1, the system of the invention is divided into three layers, namely a data acquisition layer, a data fusion layer and a data display layer.

The data acquisition layer is mainly used for acquiring data of different types and different places, and comprises monitoring video data (monitoring video original files and snapshot pictures of communities, roads, checkpoints and key area positions); the travel data comprises effective identity information, riding time, types of riding tools, getting-off places, getting-off time and the like acquired when the public transport vehicles are ridden; the hotel accommodations can acquire the information of the persons living in, the information of the persons in the same row, the living-in time, the leaving time and the like; the communication data may be given a specific location of a specific time period by mobile phone positioning, positioning information given by a base station of an operator, and the like. And the data acquisition layer plays a role in getting through barriers for various data acquisition and collecting related data of different types.

And the data fusion layer is mainly used for converging and integrating the data of the acquisition layer, cleaning, clustering the data and fusing the data according to actual needs, and providing a service interface for the application layer. The data aggregation can be physical aggregation of data or logical aggregation of data, and is real-time or periodically aggregated to a server for processing data in an interface or platform mode; data cleansing is a process of rechecking and verifying data through a certain cleansing rule, and aims to delete duplicate information, correct existing errors, and provide data consistency. Data clustering, namely clustering data through a clustering algorithm and merging the data belonging to a certain object; and data fusion, namely three levels of data combination, data integration, data aggregation and the like, wherein the deep interaction between the data is gradually realized from low to high.

And the data display layer is mainly used for performing visualization display on the fused data, including map display, video display, picture display and comprehensive display, vividly displaying the behavior track of the attention target and providing an auxiliary function for subsequent judgment and decision making.

As shown in the system structure diagram of fig. 2, a data acquisition device is installed and deployed at the acquisition end, including a data acquisition device with the addition of a camera, a wifi probe, infrared detection, radar data acquisition, and the like.

And receiving the data uploaded by the data acquisition equipment at the data acquisition server, forwarding and storing the data according to the requirement and service setting, and distributing the acquired data to different data and fusing servers.

And the data fusion server is used for cleaning, clustering and fusing data and providing diversified service interfaces such as addition, deletion, modification, check and the like for the upper application layer.

The data application server deploys application services, provides display and interface operation response for the client, and realizes data display according to time, target, location and individual requirements.

Claims (1)

1. A multi-information fused target track tracking system is characterized by comprising:

the data acquisition layer mainly acquires data of different places and different types, and comprises the following steps: monitoring video data including monitoring video original files and snap pictures of communities, roads, checkpoints and key area positions; the travel data comprises effective identity information, riding time, types of riding tools, getting-off places and getting-off time which are acquired when the public transport vehicles are ridden; the hotel accommodations collect the information of the check-in people, the information of the co-workers, the check-in time and the leaving time; communication data, wherein specific positions of specific time periods are given through mobile phone positioning and positioning information given by a base station of an operator;

the data fusion layer is mainly used for gathering and integrating the data of the acquisition layer, cleaning, clustering the data and fusing the data according to actual needs and providing a service interface for the application layer; the data aggregation can be physical aggregation of data or logical aggregation of data, and is real-time or periodically aggregated to a server for processing data in an interface or platform mode; data cleaning is a process of rechecking and verifying data through a certain cleaning rule, and aims to delete repeated information, correct existing errors and provide data consistency; data clustering, namely clustering data through a clustering algorithm and merging the data belonging to a certain object; data fusion, namely combining data, integrating data, aggregating data and the like, wherein the deep interaction among the data is realized step by step from low to high;

and the data display layer is mainly used for performing visualization display on the fused data, including map display, video display, picture display and comprehensive display, vividly displaying the behavior track of the attention target and providing an auxiliary function for subsequent judgment and decision making.

Images