CN113806459B - Method, system and storage medium for identifying staff on basis of Beidou grid - Google Patents

Abstract

The invention discloses a peer person identification method, a peer person identification system and a storage medium based on Beidou grids, relates to the technical field of peer person identification, and can quickly and accurately identify peer persons. The technical scheme of the invention comprises the following steps: and performing grid subdivision and coding on the monitored area by using an earth subdivision grid coding technology of the Beidou grid position code to obtain a grid map. And gridding the personnel track to obtain a track grid area, wherein the grid size of the track grid area is the same as that of the grid map. All the grids in each track grid region constitute a track grid set. Judging whether track grids in the track grid set of the two persons meet or not, and if the number of the meeting grids exceeds a set threshold, judging that the two persons are the same person.

Description

Method, system and storage medium for identifying staff on basis of Beidou grid

Technical Field

The invention relates to the technical field of peer identification, in particular to a peer identification method, a peer identification system and a storage medium based on Beidou grids.

Background

The technology related to the identification of the staff can be applied to various scenes, such as security and protection fields, internet social contact and the like.

At present, the common adoption of the technical scheme for identifying the appointed person is that whether any person and the appointed person continuously appear in one image is judged by utilizing the face recognition technology of a camera, and because the condition that the camera captures the face and accurately identifies in reality is harsh, the identification error and failure are caused because the angle, shielding and other reasons cannot acquire enough face data, thereby greatly reducing the identification accuracy of the peer.

Meanwhile, in the case of post judgment, the obtained data is always track data of all people in a period of time, and a rapid method is needed for extracting the co-workers from a large pile of track data.

Therefore, a scheme capable of rapidly and accurately identifying the staff is needed.

Disclosure of Invention

In view of the above, the invention provides a peer person identification method, a system and a storage medium based on Beidou grid, which can quickly and accurately identify peer persons.

In order to achieve the above purpose, the technical scheme of the invention comprises the following steps:

s1: and performing grid subdivision and coding on the monitored area by using an earth subdivision grid coding technology of the Beidou grid position code to obtain a grid map.

S2: and gridding the personnel track to obtain a track grid area, wherein the grid size of the track grid area is the same as that of the grid map.

All the grids in each track grid region constitute a track grid set.

S3: judging whether track grids in the track grid set of the two persons meet or not, and if the number of the meeting grids exceeds a set threshold, judging that the two persons are the same person.

Further, all grids in each track grid region form a track grid set, and specifically the following steps are adopted:

and constructing a state relation index table for each person corresponding to the track grid area.

The state relation index table comprises a personnel identity code ID, a time t and a personnel state p; the personnel state p is the track grid code of the personnel at the moment t.

And determining personnel states p at all discrete moments in the time period to be analyzed according to the constructed state relation index table to form a track grid set.

Further, judging whether track grids in the track grid set of two persons meet or not, specifically, adopting the following method:

the track grid sets of the two persons are L respectively ¹ And L ² 。

First person trajectory grid set L ¹ ＝{(p ¹ _n ，t ¹ _n ) N=1..n }; n is the total number of track grids of the first person; t is t ¹ _n The time corresponding to the nth track grid of the first person; p is p ¹ _n And the person state corresponding to the nth track grid of the first person.

Second person trajectory grid set L ² ＝{(p ² _m ，t ² _m ) M=1..m }; m is the total number of track grids of the second person; t is t ² _m The time corresponding to the mth track grid of the second person; p is p ² _m And the personnel state corresponding to the mth track grid of the second personnel.

Judgment of L ¹ And L ² If the meeting grids exist, judging that the two persons are the same person if the meeting grids exist and the number of the meeting grids exceeds a set threshold value.

The set threshold is set as: r×min (length (L) ¹ ),Length(L ² ) A) is provided; wherein length (L) ¹ ) Represents L ¹ Is a length of (2); length (L) ² ) Represents L ² Is a length of (2); r is a set scale.

Further, judge L ¹ And L ² Whether an encounter grid exists or not, the specific method is as follows:

for an nth track grid (p ¹ _n ，t ¹ _n ) And an mth track grid (p ² _m ，t ² _m ) If the following four conditions are satisfied at the same time, the two are the meeting grid:

condition a: distance (p) ¹ _n ,p ² _m )|<dist; wherein |distance (p) ¹ _n ,p ² _m ) Represents p ¹ _n ,p ² _m (a straight line distance between the center points of the two may be used); dist is a preset distanceA threshold value.

Condition b: t ¹ _n -t ² _m |<tim; wherein tim is a preset time threshold.

Condition c: p is p ² _m Is L ² Satisfies the condition b and the distance p ¹ _n The nearest track grid.

Condition d: if p of condition c is satisfied ² _m The number exceeds 1, and |t is selected from the above ¹ _n -t ² _m The grid with smallest is p ¹ _n Is used for the meeting grid of the (a).

Further, the preset time threshold tim is set to 15 seconds.

The invention further provides a peer identification system based on the Beidou grid, which comprises a monitoring area subdivision module, a grid module and a peer judgment module.

And the monitoring area subdivision module is used for inputting a monitored area map, performing grid subdivision and encoding on the monitored area by using an earth subdivision grid encoding technology of Beidou grid position codes, obtaining a grid map and outputting the grid map.

And the gridding module is used for inputting a human track, determining track grid areas according to the positions of the human track on the grid map, forming a track grid set by all grids in each track grid area, and outputting the track grid set.

The peer judging module acquires track grid sets of two persons to be judged, judges whether track grids in the track grid sets of the two persons meet, and judges that the two persons are peers if the number of the meeting grids exceeds a set threshold.

Further, the meshing module specifically adopts the following steps:

and constructing a state relation index table for each person corresponding to the track grid area.

The state relation index table comprises a personnel identity code ID, a time t and a personnel state p; the personnel state p is the track grid code of the personnel at the moment t.

And determining personnel states p at all discrete moments in the time period to be analyzed according to the constructed state relation index table to form a track grid set.

Further, the peer judgment module specifically adopts the following method:

the track grid sets of the two persons are L respectively ¹ And L ² ；

First person trajectory grid set L ¹ ＝{(p ¹ _n ，t ¹ _n ) N=1..n }; n is the total number of track grids of the first person; t is t ¹ _n The time corresponding to the nth track grid of the first person; p is p ¹ _n And the person state corresponding to the nth track grid of the first person.

Second person trajectory grid set L ² ＝{(p ² _m ，t ² _m ) M=1..m }; m is the total number of track grids of the second person; t is t ² _m The time corresponding to the mth track grid of the second person; p is p ² _m And the personnel state corresponding to the mth track grid of the second personnel.

Judgment of L ¹ And L ² If the meeting grids exist, judging that the two persons are the same person if the meeting grids exist and the number of the meeting grids exceeds a set threshold value.

The set threshold is set as: r×min (length (L) ¹ ),Length(L ² ) A) is provided; wherein length (L) ¹ ) Represents L ¹ Is a length of (2); length (L) ² ) Represents L ² Is a length of (2); r is a set scale.

For an nth track grid (p ¹ _n ，t ¹ _n ) And an mth track grid (p ² _m ，t ² _m ) If the following four conditions are satisfied at the same time, the two are the meeting grid:

condition a: distance (p) ¹ _n ,p ² _m )|<dist; wherein |distance (p) ¹ _n ,p ² _m ) Represents p ¹ _n ,p ² _m Is the distance between (2)Distance (linear distance between the two center points can be used); dist is a preset distance threshold;

condition b: t ¹ _n -t ² _m |<tim; wherein tim is a preset time threshold;

condition c: p is p ² _m Is L ² Satisfies the condition b and the distance p ¹ _n A nearest track grid;

condition d: if p of condition c is satisfied ² _m The number exceeds 1, and |t is selected from the above ¹ _n -t ² _m The grid with smallest is p ¹ _n Is used for the meeting grid of the (a).

Another embodiment of the present invention also provides a computer readable storage medium having stored thereon computer instructions which, when executed by a processor, implement the steps of the beidou grid based peer identification method of any one of claims 1 to 6.

The beneficial effects are that:

the invention provides a peer identification scheme based on Beidou grids, which comprises the steps of firstly, carrying out grid division and grid coding on a monitoring area; then establishing a state relation index of the personnel and the gridding area to form a personnel movement track grid set; then searching the track length which is overlapped with other target objects in a buffer zone (dist) of the personnel track, and if the overlapping ratio exceeds a certain value, identifying two people as the same person; and repeating the steps until all the peer combinations are found. Compared with the defects of low recognition rate and easy error of the same person in the prior art, the invention provides a measurable and computer programmable solution, and can greatly improve the accuracy and efficiency of the recognition of the same person.

Drawings

Fig. 1 is a flow chart of a peer identification method based on Beidou grid.

Detailed Description

The invention will now be described in detail by way of example with reference to the accompanying drawings.

The invention provides a peer person identification method based on Beidou grids, which is shown in a flow chart in fig. 1 and comprises the following steps:

s1: performing grid subdivision and coding on the monitored area by using an earth subdivision grid coding technology of Beidou grid position codes to obtain a grid map; the mesh can be split according to the requirement, and the mesh is as small as centimeter level.

S2: and gridding the personnel track to obtain a track grid area, wherein the grid size of the track grid area is the same as that of the grid map.

All the grids in each track grid region constitute a track grid set.

Constructing a state relation index table aiming at the corresponding track grid area of each person; the state relation index table comprises a personnel identity code ID, a time t and a personnel state p; the personnel state p is the track grid code of the personnel at the moment t; and determining personnel states p at all discrete moments in the time period to be analyzed according to the constructed state relation index table to form a track grid set.

S3: judging whether track grids in the track grid set of the two persons meet or not, and if the number of the meeting grids exceeds a set threshold, judging that the two persons are the same person.

Specifically, the following method is adopted:

the track grid sets of the two persons are L respectively ¹ And L ² ；

First person trajectory grid set L ¹ ＝{(p ¹ _n ，t ¹ _n ) N=1..n }; n is the total number of track grids of the first person; t is t ¹ _n The time corresponding to the nth track grid of the first person; p is p ¹ _n The person state corresponding to the nth track grid of the first person;

second person trajectory grid set L ² ＝{(p ² _m ，t ² _m ) M=1..m }; m is the total number of track grids of the second person; t is t ² _m The time corresponding to the mth track grid of the second person; p is p ² _m The personnel state corresponding to the mth track grid of the second personnel;

judgment of L ¹ And L ² If the meeting grids exist, judging that the two persons are the same person if the meeting grids exist and the number of the meeting grids exceeds a set threshold value;

the set threshold is set as: r×min (length (L) ¹ ),Length(L ² ) A) is provided; wherein length (L) ¹ ) Represents L ¹ (the total number of grids N in the set can be used as a length scale); length (L) ² ) Represents L ² Is a length of (2); r is the set scale (e.g., r is set to 80%).

Judgment of L ¹ And L ² Whether an encounter grid exists or not, the specific method is as follows:

for an nth track grid (p ¹ _n ，t ¹ _n ) And an mth track grid (p ² _m ，t ² _m ) If the following four conditions are satisfied at the same time, the two are the meeting grid:

condition a: distance (p) ¹ _n ,p ² _m )|<dist; wherein |distance (p) ¹ _n ,p ² _m ) Represents p ¹ _n ,p ² _m (a straight line distance between the center points of the two may be used); dist is a preset distance threshold;

condition b: t ¹ _n -t ² _m |<tim; wherein tim is a preset time threshold;

condition c: p is p ² _m Is L ² Satisfies the condition b and the distance p ¹ _n A nearest track grid;

condition d: if p of condition c is satisfied ² _m The number exceeds 1, and |t is selected from the above ¹ _n -t ² _m The grid with smallest is p ¹ _n Is used for the meeting grid of the (a).

Wherein the preset distance threshold Dist is determined according to the size of the track grid, and the Dist is set to be not more than 2 times of the size of the track grid. The preset time threshold tim is set to 15 seconds.

The invention also provides a peer identification system based on the Beidou grid, which comprises a monitoring area subdivision module, a grid module and a peer judgment module.

And the monitoring area subdivision module is used for inputting a monitored area map, performing grid subdivision and encoding on the monitored area by using an earth subdivision grid encoding technology of Beidou grid position codes, obtaining a grid map and outputting the grid map.

And the gridding module is used for inputting a human track, determining track grid areas according to the positions of the human track on the grid map, forming a track grid set by all grids in each track grid area, and outputting the track grid set.

The peer judging module acquires track grid sets of two persons to be judged, judges whether track grids in the track grid sets of the two persons meet, and judges that the two persons are peers if the number of the meeting grids exceeds a set threshold.

Wherein, the gridding module specifically adopts the following steps:

and constructing a state relation index table for each person corresponding to the track grid area.

The state relation index table comprises a personnel identity code ID, a time t and a personnel state p; the personnel state p is the track grid code of the personnel at the moment t.

And determining personnel states p at all discrete moments in the time period to be analyzed according to the constructed state relation index table to form a track grid set.

The peer judgment module specifically adopts the following method:

the track grid sets of the two persons are L respectively ¹ And L ² 。

First person trajectory grid set L ¹ ＝{(p ¹ _n ，t ¹ _n ) N=1..n }; n is the total number of track grids of the first person; t is t ¹ _n The time corresponding to the nth track grid of the first person; p is p ¹ _n And the person state corresponding to the nth track grid of the first person.

Second person trajectory grid set L ² ＝{(p ² _m ，t ² _m ) M=1..m }; m is the total number of track grids of the second person; t is t ² _m The time corresponding to the mth track grid of the second person; p is p ² _m And the personnel state corresponding to the mth track grid of the second personnel.

Judgment of L ¹ And L ² If the meeting grids exist, judging that the two persons are the same person if the meeting grids exist and the number of the meeting grids exceeds a set threshold value.

The set threshold is set as: r×min (length (L) ¹ ),Length(L ² ) A) is provided; wherein length (L) ¹ ) Represents L ¹ (the total number of grids N in the set can be used as a length scale); length (L) ² ) Represents L ² Is a length of (2); r is the set scale (e.g., r is set to 80%).

For an nth track grid (p ¹ _n ，t ¹ _n ) And an mth track grid (p ² _m ，t ² _m ) If the following four conditions are satisfied at the same time, the two are the meeting grid:

condition a: distance (p) ¹ _n ,p ² _m )|<dist; wherein |distance (p) ¹ _n ,p ² _m ) Represents p ¹ _n ,p ² _m (a straight line distance between the center points of the two may be used); dist is a preset distance threshold.

Condition b: t ¹ _n -t ² _m |<tim; wherein tim is a preset time threshold.

Condition c: p is p ² _m Is L ² Satisfies the condition b and the distance p ¹ _n The nearest track grid.

Condition d: if p of condition c is satisfied ² _m The number exceeds 1, and |t is selected from the above ¹ _n -t ² _m The grid with smallest is p ¹ _n Is used for the meeting grid of the (a).

The embodiment of the invention also provides a computer readable storage medium, wherein the computer readable storage medium is stored with computer instructions, and the computer instructions can realize steps in the peer personnel identification method based on the Beidou grid when being executed by a processor.

The embodiment of the invention provides an operation example:

(1) The tracks of the candidate persons are meshed, so that the state of each person in the mesh along with time (such as the mesh l occupied by the person A at the moment t) can be indexed;

(2) Selecting person name (A), time period to be used as analysis base track of the same person

(3) Setting a space limit dist grid number, setting a time limit tim and setting an r value

(4) Confirming, entering into analysis of the same person

The calculation flow is as follows:

1. obtaining a basic personnel grid track A and calculating the length grid number W of the grid track _a ；

2. Making a buffer area with a dist value according to the track;

3. searching and recording other personnel tracks falling into the buffer zone, setting the tracks as an X track set, and calculating the grid number WX_i of each track in the buffer zone;

4. track starting point grid labels and end point grid labels in X are searched one by one and recorded as X_i (X ₁ ，x _n )；

5. The nearest grid in track A is searched by the X1 grid in X_i and recorded as X_i/a ₁ ；

6. The nearest grid in track A is searched by the X2 grid in X_i and recorded as X_i/a ₂ ；

7. Retrieving the nearest grid in the track A by using the xn grid in the X_i, and recording as X_i/a _n ；

8. Comparison a ₁ And a _n If a) _n Is earlier than a ₁ If the track X_i is the reverse track of the track A, deleting all the reverse tracks to obtain a reduced homodromous track set X_i (X1, xn) and simultaneously usingX_i(a ₁ ,a _n ) Positioning the start-stop position relation between the track in the X and the track A;

9. for tracks X_i in the X set, from X ₁ Correspond to a ₁ Beginning grid-by-grid comparison time difference |Xt with the grid in track A _i –At _i |<tim, and distance (X _i ,A _i )|<The dist distance difference records grids on the X_i track meeting the requirement until the grid of the terminal point xn is stopped, and gathers the grid numbers MX_i meeting dist and tim;

10. calculating r value, wherein r=MX_i/WX_i, and when r is more than or equal to 80%, the person X_i is considered as the peer of the person A;

11. and displaying the tracks of the pedestrians on the main page and outputting a list of pedestrians.

In summary, the above embodiments are only preferred embodiments of the present invention, and are not intended to limit the scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (6)

1. The peer personnel identification method based on the Beidou grid is characterized by comprising the following steps of:

s1: performing grid subdivision and coding on the monitored area by using an earth subdivision grid coding technology of Beidou grid position codes to obtain a grid map;

s2: gridding the personnel track to obtain a track grid area, wherein the grid size of the track grid area is the same as that of the grid map;

all grids in each track grid region form a track grid set;

s3: judging whether track grids in the track grid set of the two persons meet or not, and if the number of the meeting grids exceeds a set threshold, judging that the two persons are the same person;

judging whether track grids in the track grid set of two persons meet or not, and specifically, adopting the following method:

the track grid sets of the two persons are L respectively ¹ And L ² ；

First person trajectory grid set L ¹ ＝{(p ¹ _n ，t ¹ _n ) N=1..n }; n is the total number of track grids of the first person; t is t ¹ _n The time corresponding to the nth track grid of the first person; p is p ¹ _n The person state corresponding to the nth track grid of the first person;

second person trajectory grid set L ² ＝{(p ² _m ，t ² _m ) M=1..m }; m is the total number of track grids of the second person; t is t ² _m The time corresponding to the mth track grid of the second person; p is p ² _m The personnel state corresponding to the mth track grid of the second personnel;

judgment of L ¹ And L ² If the meeting grids exist, judging that the two persons are the same person if the meeting grids exist and the number of the meeting grids exceeds a set threshold value;

the set threshold is set as: r×min (length (L) ¹ ),Length(L ² ) A) is provided; wherein length (L) ¹ ) Represents L ¹ Is a length of (2); length (L) ² ) Represents L ² Is a length of (2); r is a set scale;

the judgment L ¹ And L ² Whether an encounter grid exists or not, the specific method is as follows:

for an nth track grid (p ¹ _n ，t ¹ _n ) And an mth track grid (p ² _m ，t ² _m ) If the following four conditions are satisfied at the same time, the two are the meeting grid:

condition a: distance (p) ¹ _n ,p ² _m )|<dist; wherein |distance (p) ¹ _n ,p ² _m ) Represents p ¹ _n ,p ² _m (using the straight line distance between the two center points); dist is a preset distance threshold;

condition b: t ¹ _n -t ² _m |<tim; wherein the method comprises the steps oftim is a preset time threshold;

condition c: p is p ² _m Is L ² Satisfies the condition b and the distance p ¹ _n A nearest track grid;

condition d: if p of condition c is satisfied ² _m The number exceeds 1, and |t is selected from the above ¹ _n -t ² _m The grid with smallest is p ¹ _n Is used for the meeting grid of the (a).

2. The method of claim 1, wherein all grids in each track grid region form a track grid set, specifically comprising the steps of:

constructing a state relation index table aiming at the corresponding track grid area of each person;

the state relation index table comprises a personnel identity code ID, a time t and a personnel state p; the personnel state p is the track grid code of the personnel at the moment t;

and determining personnel states p at all discrete moments in the time period to be analyzed according to the constructed state relation index table to form a track grid set.

3. The method according to claim 2, wherein the preset time threshold tim is set to 15 seconds.

4. The peer identification system based on the Beidou grid is characterized by comprising a monitoring area subdivision module, a gridding module and a peer judgment module;

the monitoring area subdivision module is used for inputting a monitored area map, performing grid subdivision and coding on the monitored area by using an earth subdivision grid coding technology of Beidou grid position codes, obtaining a grid map and outputting the grid map;

the gridding module inputs the human track, determines track grid areas according to the positions of the human track on the grid map, all grids in each track grid area form a track grid set, and outputs the track grid set;

the peer judgment module acquires track grid sets of two persons to be judged, judges whether track grids in the track grid sets of the two persons meet, and judges that the two persons are peers if the number of the meeting grids exceeds a set threshold;

the peer judgment module specifically adopts the following method:

the track grid sets of the two persons are L respectively ¹ And L ² ；

First person trajectory grid set L ¹ ＝{(p ¹ _n ，t ¹ _n ) N=1..n }; n is the total number of track grids of the first person; t is t ¹ _n The time corresponding to the nth track grid of the first person; p is p ¹ _n The person state corresponding to the nth track grid of the first person;

second person trajectory grid set L ² ＝{(p ² _m ，t ² _m ) M=1..m }; m is the total number of track grids of the second person; t is t ² _m The time corresponding to the mth track grid of the second person; p is p ² _m The personnel state corresponding to the mth track grid of the second personnel;

judgment of L ¹ And L ² If the meeting grids exist, judging that the two persons are the same person if the meeting grids exist and the number of the meeting grids exceeds a set threshold value;

the set threshold is set as: r×min (length (L) ¹ ),Length(L ² ) A) is provided; wherein length (L) ¹ ) Represents L ¹ Is a length of (2); length (L) ² ) Represents L ² Is a length of (2); r is a set scale;

for an nth track grid (p ¹ _n ，t ¹ _n ) And an mth track grid (p ² _m ，t ² _m ) If the following four conditions are satisfied at the same time, the two are the meeting grid:

condition a: distance (p) ¹ _n ,p ² _m )|<dist; wherein |distance (p) ¹ _n ,p ² _m ) Represents p ¹ _n ,p ² _m (using the straight line distance between the two center points); dist is a preset distance threshold;

condition b: t ¹ _n -t ² _m |<tim; wherein tim is a preset time threshold;

condition c: p is p ² _m Is L ² Satisfies the condition b and the distance p ¹ _n A nearest track grid;

condition d: if p of condition c is satisfied ² _m The number exceeds 1, and |t is selected from the above ¹ _n -t ² _m The grid with smallest is p ¹ _n Is used for the meeting grid of the (a).

5. The system of claim 4, wherein the gridding module comprises the steps of:

constructing a state relation index table aiming at the corresponding track grid area of each person;

the state relation index table comprises a personnel identity code ID, a time t and a personnel state p; the personnel state p is the track grid code of the personnel at the moment t;

and determining personnel states p at all discrete moments in the time period to be analyzed according to the constructed state relation index table to form a track grid set.

6. A computer readable storage medium having stored thereon computer instructions, which when executed by a processor, implement the steps of the beidou grid based peer identification method of any of claims 1 to 3.