Person tracking method based on multiple monitoring videos

Abstract

The invention relates to the technical field of monitoring, and discloses a person tracking method based on multiple monitoring videos, which comprises the following steps: S1, acquiring a temperature feature, a height feature and a temperature distribution feature of a monitoring target through a video monitoring unit, forming a corresponding target feature set, and sending the target feature set to a central processing module; according to the person tracking method based on the multiple monitoring videos, the temperature feature, the height feature and the temperature distribution feature of the monitoring target, the initial displacement speed data of the monitoring target entering the video monitoring unit and the entering timestamp are obtained, non-monitoring persons can be discharged through preliminary judgment, the number of follow-up operation data is reduced, and the efficiency of tracking is improved. And comparing the obtained temperature features, height features and temperature distribution features of the person with the corresponding data of each item of the target feature set, and when each item of data is the same as each item of data in the target feature set, generating and determining a monitoring target signal, thereby reducing operation steps, simplifying an operation algorithm, reducing analysis duration, and improving the accuracy of monitoring. The response speed is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of monitoring, and specifically to a method for tracking people based on multiple monitoring videos. Background Art

[0002] Video tracking technology plays an extremely important role in the security field. Its application can significantly improve the efficiency, accuracy, and response speed of the monitoring system. Video tracking technology can identify and locate the target person in the monitoring video in real time. Even in complex backgrounds or multi-person scenarios, it can accurately track specific personnel. This enables security personnel to quickly lock in suspicious objects, reduce reliance on manual monitoring, and improve the response speed. For example, in crowded places such as airports, stations, and shopping malls, video tracking can help quickly locate lost children or suspicious persons;

[0003] Currently, when the monitoring target moves back and forth in the areas monitored by multiple video monitoring devices, a large amount of data will be generated. The data processing units of these video monitoring devices need to calculate all the generated data to analyze whether the task newly entering other video monitoring devices is the monitoring target. It requires complex operations, increasing the analysis time and reducing the response speed. Summary of the Invention

[0004] (1) Technical Problems to be Solved

[0005] Aiming at the deficiencies of the prior art, the present invention provides a method for tracking people based on multiple monitoring videos. It is capable of obtaining the temperature characteristics, height characteristics, temperature distribution characteristics, initial displacement speed data, and entry timestamp of the monitoring target when it enters the video monitoring unit, and obtaining the final displacement speed data and departure timestamp of the monitoring target when it leaves the video monitoring unit, and calculating the acceleration of the monitoring target. When the monitoring target leaves the monitoring area of the current video monitoring unit, other monitoring units obtain the entry timestamp T of the person entering the monitoring area, the initial displacement speed data S of the person entering the monitoring area, and the received departure timestamp Lt i and the final displacement speed data is denoted as Zv i for comparison. As a preliminary comparison result, it is determined whether the person entering the monitoring area is the monitoring target. Through preliminary determination, non-monitoring persons can be excluded, reducing the quantity of subsequent operation data. When it is preliminarily determined that the person is the monitoring target, the temperature characteristics, height characteristics, and temperature distribution characteristics of the person are obtained, and the obtained temperature characteristics, height characteristics, and temperature distribution characteristics of the person are compared with the corresponding data of the target feature set. When all the data is the same as the data in the target feature set, a signal for determining the monitoring target is generated, thus achieving the advantages of reducing operation steps, simplifying the operation algorithm, reducing the analysis time, and improving the response speed, and solving the above problems.

[0006] (2) Technical solution

[0007] To achieve the above object, the present invention provides the following technical solution: A method for tracking a person based on multiple surveillance videos, comprising the following steps:

[0008] S1. Obtain the temperature feature, height feature and temperature distribution feature of the monitoring target through the video monitoring unit, form a corresponding target feature set, and send the target feature set to the central processing module;

[0009] S2. The video monitoring unit obtains the initial displacement speed data and entry timestamp of the monitoring target entering the video monitoring unit, and obtains the final displacement speed data and departure timestamp of the monitoring target leaving the video monitoring unit, and sends these data to the central processing module;

[0010] S3. The central processing module calculates the growth rate of the displacement speed of the monitoring target according to the initial displacement speed data and the final displacement speed data of the monitoring target, and calculates the acceleration of the monitoring target according to the entry timestamp and the departure timestamp;

[0011] S4. The central processing module sends the target feature set, the final displacement speed data, the departure timestamp and the monitoring target acceleration of the monitoring target to other video monitoring units;

[0012] S5. Other video monitoring units obtain the video data of the monitoring area, and judge whether the monitoring target is within the monitoring area of the video monitoring unit according to the received target feature set, the final displacement speed data, the departure timestamp and the monitoring target acceleration. When the monitoring target leaves the monitoring area of the current video monitoring unit, repeat step one and step two.

[0013] Preferably, the target feature set is expressed as: {Y i , G i , W i , FB i}, where the subscript i represents the number of the monitoring target.

[0014] Preferably, there are several video monitoring units, and all of them are network-connected to the central processing module. Each of the multiple video monitoring units is provided with a unique digital number, and the multiple digital numbers are respectively expressed as: {SP 1 , SP 2 ,..., SP n}, the subscripts 1 to n are the digital numbers corresponding to the video monitoring units, and the digital numbers contain the coordinate data of each video monitoring unit.

[0015] Preferably, the initial displacement speed data is expressed as Cv i , and the entry timestamp is expressed as: Jti The final displacement velocity data is denoted as Zv i The departure timestamp is denoted as Lt i where the subscript i represents the number of the monitored target.

[0016] Preferably, the calculation expression for the growth rate of the displacement velocity of the monitored target is as follows:

[0017]

[0018] In the formula, MZS i represents the growth rate of the displacement velocity of the i-th monitored target, and Zv i - Cv i represents the difference between the final displacement velocity data and the initial displacement velocity data of the i-th monitored target. The quotient obtained by dividing this difference by the initial displacement velocity data of the i-th monitored target is the growth rate MZS of the displacement velocity of the monitored target i .

[0019] Preferably, the calculation expression for the acceleration of the monitored target is as follows:

[0020]

[0021] In the formula, MJS i represents the acceleration of the i-th monitored target, and Lt i - Jt i represents the difference between the departure timestamp and the entry timestamp of the i-th monitored target, that is, it reflects the time the i-th monitored target stays in the current video monitoring unit. The result obtained by dividing the growth rate of the displacement velocity of the i-th monitored target by the time the i-th monitored target stays in the current video monitoring unit is the acceleration MJS of the i-th monitored target i .

[0022] Preferably, after calculating the acceleration MJS of the i-th monitored target i the central processing module sends the target feature set of the i-th monitored target, the departure timestamp Lt i , the final displacement velocity data denoted as Zv i and the acceleration MJS of the i-th monitored target i to other video monitoring units.

[0023] Preferably, when other video monitoring units acquire video data of the monitoring area, they acquire the entry timestamp T when a person enters the monitoring area in the video data, the initial displacement velocity data S when the person enters the monitoring area, and the received departure timestamp Lt i , the final displacement velocity data denoted as Zv iCompare, as a preliminary comparison result, determine whether the person entering the monitoring area is the monitoring target;

[0024] The expression of the preliminary comparison result is:

[0025]

[0026] In the formula, DB 1 represents the preliminary comparison result, DB 1 = 1 indicates that the preliminary comparison result is that the person entering the monitoring area is the monitoring target, T = Lt i ±0.02, S = Zv i ±MJS i represents the condition for determining DB 1 = 1, DB 1 = 0 indicates that the preliminary comparison result is that the person entering the monitoring area is not the monitoring target, T ≠ Lt i ±0.02, S ≠ Zv i ±MJS i represents the condition for determining DB 1 = 0.

[0027] Preferably, when other video monitoring units determine through preliminary comparison that the person entering the monitoring area is the monitoring target, obtain the temperature characteristics, height characteristics, and temperature distribution characteristics of the person, and compare the obtained temperature characteristics, height characteristics, and temperature distribution characteristics of the person with the corresponding data in the target feature set. When the data are the same as the data in the target feature set, generate a signal for determining the monitoring target.

[0028] Preferably, when the signal for determining the monitoring target is generated, that is, the person entering the monitoring area is the monitoring target.

[0029] Compared with the prior art, the present invention provides a method for tracking a person based on multiple monitoring videos, having the following beneficial effects:

[0030] The present invention obtains the temperature characteristics, height characteristics, temperature distribution characteristics, initial displacement speed data of the monitoring target entering the video monitoring unit, and the entry timestamp, and obtains the final displacement speed data and the departure timestamp of the monitoring target leaving the video monitoring unit, and calculates the acceleration of the monitoring target. When the monitoring target leaves the monitoring area of the current video monitoring unit, other monitoring units obtain the entry timestamp T of the person entering the monitoring area, the initial displacement speed data S of the person entering the monitoring area, and the received departure timestamp Lt i , and the final displacement speed data is represented as Zv iCompare. As a preliminary comparison result, determine whether the person entering the monitoring area is the monitoring target. Through preliminary determination, non-monitoring persons can be excluded, reducing the amount of subsequent operation data. When it is preliminarily determined that the person is the monitoring target, obtain the temperature characteristics, height characteristics, and temperature distribution characteristics of the person, and compare the obtained temperature characteristics, height characteristics, and temperature distribution characteristics of the person with the corresponding data in the target feature set. When the data are the same as the data in the target feature set, generate a signal to determine the monitoring target, thereby reducing the number of operation steps, simplifying the operation algorithm, reducing the analysis time, and improving the response speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a schematic diagram of the method steps of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0032] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of the present invention.

[0033] Please refer to Figure 1 , a method for tracking a person based on multiple surveillance videos, including the following steps:

[0034] Step 1: Obtain the temperature characteristics, height characteristics, and temperature distribution characteristics of the monitoring target through the video monitoring unit, and form a corresponding target feature set, and send the target feature set to the central processing module;

[0035] Among them, the target feature set is expressed as: {Y i , G i , W i , FB i}, the subscript i in this dataset represents the number of the monitoring target, and there are several video monitoring units, and all are network-connected to the central processing module. Each of the multiple video monitoring units is set with a unique digital number, and the multiple digital numbers are respectively expressed as: {SP 1 , SP 2 ,..., SP n}, the subscripts 1 to n are the digital numbers corresponding to the video monitoring units, and the digital numbers contain the coordinate data of each video monitoring unit.

[0036] By assigning a unique digital number to each of the multiple video monitoring units, and the digital number contains the coordinate data of the corresponding video monitoring unit. Through the coordinate data, the specific coordinates of the monitoring area where the monitoring target is located can be quickly determined, and the current position of the monitoring target can be quickly grasped.

[0037] It should be noted that the video monitoring unit is an intelligent infrared thermal imaging camera, which can measure the temperature of an object with high precision, and can intuitively display the temperature distribution and changes of the target. This type of camera is an existing and mature technology, which will not be elaborated in this article. It can also be a thermal imaging camera or a combination of a camera and a thermal imaging device. Any monitoring device that can directly obtain the temperature distribution, height, and body temperature of the target can be used as the video monitoring unit.

[0038] Step 2: The video monitoring unit obtains the initial displacement speed data and entry timestamp of the monitoring target entering the video monitoring unit, and obtains the final displacement speed data and departure timestamp of the monitoring target leaving the video monitoring unit, and sends this data to the central processing module;

[0039] The initial displacement speed data is denoted as Cv i , and the entry timestamp is denoted as: Jt i , the final displacement speed data is denoted as Zv i , and the departure timestamp is denoted as Lt i , where the subscript i represents the number of the monitoring target;

[0040] Step 3: The central processing module calculates the growth rate of the displacement speed of the monitoring target based on the initial displacement speed data and the final displacement speed data of the monitoring target, and calculates the acceleration of the monitoring target based on the entry timestamp and the departure timestamp;

[0041] The calculation expression for the growth rate of the displacement speed of the monitoring target is as follows:

[0042]

[0043] In the formula, MZS i represents the growth rate of the displacement speed of the i-th monitoring target, Zv i - Cv i represents the difference between the final displacement speed data and the initial displacement speed data of the i-th monitoring target. The quotient of this difference divided by the initial displacement speed data of the i-th monitoring target is the growth rate MZS of the displacement speed of the monitoring target i ;

[0044] The calculation expression for the acceleration of the monitoring target is as follows:

[0045]

[0046] In the formula, MJS i represents the acceleration of the i-th monitoring target, Lt i - Jt iDenote the difference between the departure timestamp and the entry timestamp of the i-th monitored target, that is, it reflects the time the i-th monitored target stays in the current video monitoring unit. The result obtained by dividing the acceleration rate of the displacement speed of the i-th monitored target by the time the i-th monitored target stays in the current video monitoring unit is the acceleration MJS of the i-th monitored target i ;

[0047] By calculating the acceleration of the monitored target, the current motion trend of the monitored target can be analyzed, whether it is accelerating or decelerating, which can assist the monitoring personnel in grasping the motion trend of the monitored target so as to make reasonable responses;

[0048] Step Four: The central processing module sends the target feature set, the final displacement speed data, the departure timestamp, and the monitored target acceleration of the monitored target to other video monitoring units;

[0049] Specifically, when the acceleration MJS of the i-th monitored target is calculated i After that, the central processing module will send the target feature set, the departure timestamp Lt i of the i-th monitored target, and the final displacement speed data denoted as Zv i as well as the acceleration MJS of the i-th monitored target i to other video monitoring units;

[0050] Step Five: Other video monitoring units obtain the video data of the monitoring area, and judge whether the monitored target is within the monitoring area of this video monitoring unit according to the received target feature set, the final displacement speed data, the departure timestamp, and the monitored target acceleration. When the monitored target leaves the monitoring area of the current video monitoring unit, repeat Step One and Step Two;

[0051] Among them, the specific judgment steps are as follows:

[0052] (1). When other video monitoring units obtain the video data of the monitoring area, they obtain the entry timestamp T when the person enters the monitoring area in the video data, the initial displacement speed data S when the person enters the monitoring area, and the received departure timestamp Lt i , and the final displacement speed data denoted as Zv i for comparison. As a preliminary comparison result, judge whether the person entering the monitoring area is the monitored target;

[0053] The expression of the preliminary comparison result is:

[0054]

[0055] In the formula, DB 1 represents the preliminary comparison result, DB 1= 1 indicates that the preliminary comparison result is that the person entering the monitoring area is the monitoring target, T = Lt i ±0.02, S = Zv i ±MJS i Indicates the condition for determining DB 1 = 1, DB 1 = 0 indicates that the preliminary comparison result is that the person entering the monitoring area is not the monitoring target, T ≠ Lt i ±0.02, S ≠ Zv i ±MJS i Indicates the condition for determining DB 1 = 0;

[0056] (2) When other video monitoring units determine through preliminary comparison that the person entering the monitoring area is the monitoring target, obtain the temperature characteristics, height characteristics, and temperature distribution characteristics of the person, and compare the obtained temperature characteristics, height characteristics, and temperature distribution characteristics of the person with the corresponding data in the target feature set. When all the data are the same as the data in the target feature set, generate a signal to determine the monitoring target;

[0057] (3) When generating a signal to determine the monitoring target, that is, the person entering the monitoring area is the monitoring target.

[0058] By obtaining the temperature characteristics, height characteristics, temperature distribution characteristics, initial displacement speed data of the monitoring target when entering the video monitoring unit, and the entry timestamp, and obtaining the final displacement speed data and departure timestamp of the monitoring target when leaving the video monitoring unit, and calculating the acceleration of the monitoring target. When the monitoring target leaves the monitoring area of the current video monitoring unit, other monitoring units obtain the entry timestamp T of the person entering the monitoring area, the initial displacement speed data S of the person entering the monitoring area, and the received departure timestamp Lt i The final displacement speed data is denoted as Zv i Compare as the preliminary comparison result to determine whether the person entering the monitoring area is the monitoring target. Through preliminary determination, non - monitoring persons can be excluded, reducing the number of subsequent operation data. When it is preliminarily determined that the person is the monitoring target, obtain the temperature characteristics, height characteristics, and temperature distribution characteristics of the person, and compare the obtained temperature characteristics, height characteristics, and temperature distribution characteristics of the person with the corresponding data in the target feature set. When all the data are the same as the data in the target feature set, generate a signal to determine the monitoring target, thus reducing the number of operation steps, simplifying the operation algorithm, reducing the analysis duration, and improving the response speed.

[0059] Although embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A person tracking method based on multiple surveillance videos, characterized in that: The following steps are involved: S1. Obtain temperature characteristics, height characteristics and temperature distribution characteristics of the monitored target through the video monitoring unit, form a corresponding target feature set, and send the target feature set to the central processing module; S2, the video monitoring unit obtains the initial displacement speed data and entry timestamp of the monitored target entering the video monitoring unit, and obtains the final displacement speed data and exit timestamp of the monitored target leaving the video monitoring unit, and sends these data to the central processing module; S3, the central processing module calculates the growth rate of the displacement speed of the monitored target according to the initial displacement speed data and the final displacement speed data of the monitored target, and calculates the acceleration of the monitored target according to the entry timestamp and the exit timestamp; S4, the central processing module sends the target feature set, final displacement velocity data, departure timestamp and acceleration of the monitored target to other video monitoring units; S5. Other video surveillance units obtain video data of the surveillance area, and determine whether the surveillance target is within the surveillance area of ​​the video surveillance unit based on the received target feature set, final displacement velocity data, departure timestamp, and surveillance target acceleration. When the surveillance target leaves the surveillance area of ​​the current video surveillance unit, repeat steps 1 and 2.

2. The method for tracking a person based on multiple surveillance videos according to claim 1, characterized in that: The target feature set is expressed as: {Y i , G i , W i , FB i }, where the subscript i represents the number of the monitored target.

3. The method for tracking a person based on multiple surveillance videos according to claim 2, characterized in that: There are several video monitoring units, and all of them are connected to the central processing module network. The multiple video monitoring units are respectively provided with unique digital numbers, and the multiple digital numbers are respectively represented as: {SP1, SP2, ..., SP n }, subscripts 1 to n are digital numbers corresponding to the video surveillance units, and the digital numbers include the coordinate data of each video surveillance unit.

4. The method for tracking a person based on multiple surveillance videos according to claim 3 is characterized in that: The initial displacement velocity data is expressed as Cv i , the entry timestamp is expressed as: Jt i , the final displacement velocity data is expressed as Zv i , the departure timestamp is denoted as Lt i , where the subscript i represents the number of the monitored target.

5. The method for tracking a person based on multiple surveillance videos according to claim 4 is characterized in that: The calculation expression of the speed increase of the monitoring target displacement speed is as follows: In the formula, MZS i Indicates the speed increase of the displacement velocity of the ith monitored target, Zv i -Cv i It represents the difference between the final displacement speed data and the initial displacement speed data of the ith monitored target. The quotient of this difference divided by the initial displacement speed data of the ith monitored target is the speed increase MZS of the displacement speed of the controlled target. i .

6. The method for tracking a person based on multiple surveillance videos according to claim 5, characterized in that: The acceleration calculation expression of the monitoring target is as follows: In the formula, MJS i Expressed as the acceleration of the ith monitored target, Lt i -Jt i It represents the difference between the departure timestamp and the entry timestamp of the ith monitoring target, that is, the time that the ith monitoring target stays in the current video monitoring unit. The acceleration rate of the ith monitoring target divided by the time that the ith monitoring target stays in the current video monitoring unit is the acceleration MJS of the ith monitoring target. i .

7. The method for tracking a person based on multiple surveillance videos according to claim 6, characterized in that: When the acceleration MJS of the i-th monitoring target is calculated i After that, the central processing module will send the target feature set and the departure timestamp Lt of the i-th monitoring target i , the final displacement velocity data is expressed as Zv i And the acceleration MJS of the i-th monitored target i Send to other video surveillance units.

8. The method for tracking a person based on multiple surveillance videos according to claim 7, characterized in that: When acquiring video data of the monitoring area, the other video monitoring units acquire the entry timestamp T of the person entering the monitoring area, the initial displacement speed data S of the person entering the monitoring area, and the received departure timestamp Lt in the video data. i , the final displacement velocity data is expressed as Zv i Comparison, as a preliminary comparison result, determines whether the person entering the monitoring area is a monitoring target; The expression of the preliminary comparison result is: In the formula, DB1 represents the preliminary comparison result, DB1=1 represents the preliminary comparison result that the person entering the monitoring area is the monitoring target, T=Lt i ±0.02, S = Zv i ±MJS i Indicates the condition for determining DB1=1, DB1=0 indicates that the preliminary comparison result is that the person entering the monitoring area is not the monitoring target, T≠Lt i ±0.02, S≠Zv i ±MJS i This shows the condition for determining DB1=0.

9. The method for tracking a person based on multiple surveillance videos according to claim 8, characterized in that: When the other video surveillance units determine through preliminary comparison that a person entering the surveillance area is a surveillance target, the temperature characteristics, height characteristics and temperature distribution characteristics of the person are obtained, and the obtained temperature characteristics, height characteristics and temperature distribution characteristics of the person are compared with the corresponding data of the target feature set. When the data are the same as the data in the target feature set, a surveillance target determination signal is generated.

10. The method for tracking a person based on multiple surveillance videos according to claim 9, characterized in that: When the monitoring target determination signal is generated, the person entering the monitoring area is the monitoring target.

Images