Track information numbering	Main target	Time of taking a snapshot	Vehicle camera encoding
				1	Jing XXXXXXX	2020-01-01 01:30:00	S1
2	Jing XXXXXXX	2020-01-01 03:30:00	S2
				3	Jing XXXXXXX	2020-01-01 05:30:00	S3
4	Jing XXXXXXX	2020-01-01 07:30:00	S4
				5	Jing XXXXXX	2020-01-01 09:30:00	S5
6	Jing XXXXXXX	2020-01-01 11:30:00	S6
				7	Jing XXXXXXX	2020-01-01 13:30:00	S7
8	Jing XXXXXXX	2020-01-01 15:30:00	S8
				9	Jing XXXXXXX	2020-01-01 17:30:00	S9
10	Jing XXXXXXX	2020-01-01 19:30:00	S10
				11	Jing XXXXXXX	2020-01-01 21:30:00	S5
12	Jing XXXXXXX	2020-01-01 3:30:00	S3

3) Setting an accompanying parameter, and inquiring track data in a second data source (information acquired by an electronic fence of sensing source equipment of the IMSI code of the mobile phone) according to the set accompanying parameter to obtain a plurality of tracks of accompanying targets, namely accompanying tracks, wherein the specific method comprises the following steps:

3.1) Primary Trace of the Primary target obtained with the query in the analysis timeframe, i.e., the precise time that the Primary target passed each of the perceptual-source devices S1, S2 …, etc., as shown in Table 1.

3.2) setting an offset distance, and inquiring the spatial position based on the perception source equipment passed by the main track, wherein the method comprises the following steps: and inquiring other sensing source equipment (called accompanying sensing source equipment) in the range by taking the position of each sensing source equipment passed by the main track as a circle center and the offset distance as a radius, and grouping according to the type of the sensing source equipment to obtain the one-to-many virtual same-position relation between the sensing source equipment passed by all main targets and the other accompanying sensing source equipment in the offset distance range. Such as a set offset distance of 20 meters, approximately one intersection distance, the simulated relationship is shown in table 2.

TABLE 2 Master and companion trajectory perception source offset position look-up table

Primary track information numbering	Vehicle camera encoding	Electronic fence numbering
			1	S1	D11
2	S2	D21
			3	S3	D31
4	S4
			5	S5	D51
6	S6	D61、D62
			7	S7
8	S8	D81
			9	S9
10	S10	D101

The explanation of no correspondence in table 2 shows that no electronic fence device is installed in the vehicle camera offset range.

And 3.3) setting offset time, taking the accurate time of the main target passing through each perception source device obtained by inquiring in the step 3.1) as the center, adding or subtracting the offset time from front to back, and calculating the offset time range of each accompanying perception source device, thereby obtaining the time range of other types of accompanying perception source devices corresponding to the offset time range of the perception source device passing through the main track. For example, the offset time is set to be 60 seconds, and the resulting electronic fence accompanying the sensing-source device query time in the corresponding cheap range is shown in table 3.

3.4) inquiring all the accompanying target data passing through the accompanying perception source equipment in the second data source according to all the accompanying perception source equipment obtained in the step 3.2) and the offset time ranges of all the accompanying perception source equipment obtained in the step 3.3), aggregating all the data according to the accompanying targets, and then performing time sequencing on the data of each accompanying target to obtain the tracks of all the accompanying targets passing through the accompanying perception source equipment, namely a plurality of accompanying tracks.

The data for the first companion object 460016798008534 is shown in Table 4.

The data for the second companion target 460027754062396 is shown in table 5.

TABLE 3 Master and companion trajectory offset time lookup tables

TABLE 4 accompanying target 460016798008534 trajectory information

Number of accompanying track information	Companion target	Electronic fence numbering	With target acquisition time
				1	460016798008534	D11	2020-01-01 01:29:10
2	460016798008534	D21	2020-01-01 03:29:35
				3	460016798008534	D31	2020-01-01 05:29:56
4	460016798008534	D51	2020-01-01 09:30:10
				5	460016798008534	D61	2020-01-01 11:29:25
6	460016798008534	D62	2020-01-01 11:30:20
				7	460016798008534	D81	2020-01-01 15:29:47
8	460016798008534	D81	2020-01-01 15:30:45
				9	460016798008534	D101	2020-01-01 19:29:29
10	460016798008534	D51	2020-01-01 21:30:45
				11	460016798008534	D31	2020-01-01 23:29:29

TABLE 5 accompanying target 460027754062396 trajectory information

Also, trajectory information for a third companion target, a fourth companion target, or even more companion targets may be obtained.

4) And carrying out merging and normalization repeated operation on the accompanying tracks according to the accompanying parameters, and carrying out track comparison on the main track and the processed accompanying tracks to obtain the accompanying times and the number of accompanying perception source devices, thereby uniformly providing data support for accompanying and identity.

The specific method comprises the following steps:

4.1) according to the one-to-many virtual same-position relation between the main track perception source equipment and the accompanying perception source equipment, enabling the accompanying perception source equipment in the accompanying track to be grouped into corresponding main track perception source equipment;

the data accompanying the target 460016798008534 was subjected to trajectory processing in accordance with the relationship of Table 2, where D61, D62 were both normalized to S6, as shown in Table 6.

TABLE 6 track information after normalization with object 460016798008534

4.2) according to the offset time range of each main track perception source device, carrying out deduplication operation on the data of the accompanying tracks in the offset time range, wherein only one data exists in one offset time range, as shown in the table 7.

TABLE 7 De-duplication of track information with object 460016798008534

4.3): and (4) counting the accompanying tracks after the operation of the step 4.1) and the step 4.2), wherein the total number of records is the accompanying times of the main track.

When the data in table 6460016798008534 are counted, the number of times of accompanying is the total number of the accompanying track information numbers, that is, the offset distance is 20 m and the offset time is 60 seconds, the number of times of accompanying the mobile phone with the IMSI number of 460016798008534 and the license plate with the jing XXXXXX is 9.

4.4): and (4) counting the accompanying tracks after the operations of the step 4.1) and the step 4.2), wherein the number of different perception source devices passing through the accompanying tracks is the number of the accompanying perception source devices.

The data of 460016798008534 in table 6 are counted, and sensing source devices S5 and S3 that have passed twice in different time ranges are removed, so that the number of different sensing source devices that have passed is 7, that is, when the accompanying parameters are shifted by 20 meters and the shift time is 60 seconds, the number of mobile phones with IMSI codes of 460016798008534 and the number of license plates with the accompanying sensing source devices of jing XXXXXX is 7.

5) And (3) calculating each accompanying track in the step 3) for multiple times according to the step 4) to obtain all multiple accompanying targets of the main target Jing XXXXXXX license plate, and analyzing accompanying conditions according to the accompanying times of the accompanying targets and the number of accompanying perception source equipment.

The method can well find the accompanying relation for the sensing source data determined by the main key, such as the identity card number acquisition equipment, the license plate identification system, the IMSI, the MAC acquisition equipment and the like. However, with the development of artificial intelligence and AI technology, the key features of things structured by a face recognition system, a gait recognition system and a video, such as the height of people related to people, the color of clothes, whether to wear a hat or not, etc., the color, the appearance, the model, the brand, etc., of the car can be analyzed when the number plate of the car cannot be captured, the information captured by each sensing source device is compared, the information captured by each device is normalized and processed into virtual main key information, such as a portrait file, and if only the same file is used for the photos with the similarity close to 98% of the file, the file is encoded to obtain the virtual main key, and then the virtual main key is subjected to the accompanying analysis.

Based on the same inventive concept, another embodiment of the present invention provides a system for adjoint analysis between multi-source data by using the method of the present invention, which comprises:

The specific operation processes executed by each module are referred to the description of the method of the invention.

Based on the same inventive concept, another embodiment of the present invention provides an electronic device (computer, server, smartphone, etc.) comprising a memory storing a computer program configured to be executed by the processor, and a processor, the computer program comprising instructions for performing the steps of the inventive method.

Based on the same inventive concept, another embodiment of the present invention provides a computer-readable storage medium (e.g., ROM/RAM, magnetic disk, optical disk) storing a computer program, which when executed by a computer, performs the steps of the inventive method.

The particular embodiments of the present invention disclosed above are illustrative only and are not intended to be limiting, since various alternatives, modifications, and variations will be apparent to those skilled in the art without departing from the spirit and scope of the invention. The invention should not be limited to the disclosure of the embodiments in the present specification, but the scope of the invention is defined by the appended claims.

Claims

1. A method for adjoint analysis among multi-source data is characterized by comprising the following steps:

acquiring a main track of a main target passing through a perception source device of the main target within an accompanying analysis time range from a first data source;

setting an accompanying parameter, and performing track data query in a second data source according to the accompanying parameter to obtain a plurality of accompanying tracks;

and merging the accompanying tracks according to the accompanying parameters, and performing track comparison on the main track and the merged accompanying tracks to obtain the accompanying times and the number of the accompanying sensing source devices.

2. The method of claim 1, wherein unified latitude and longitude information is set for the aware-source devices in the first data source and the second data source.

3. The method of claim 1, wherein the companion parameter comprises: past the offset time of the aware-source devices, the offset distance between the aware-source devices.

4. The method of claim 1, wherein the setting of the accompanying parameters and the performing of the track data query in the second data source according to the accompanying parameters to obtain the plurality of accompanying tracks comprises:

setting an offset distance, and inquiring the spatial position based on the perception source equipment passed by the main track, wherein the method comprises the following steps: inquiring other sensing source equipment in the range by taking the position of each sensing source equipment passed by the main track as the center of a circle and the offset distance as the radius, and calling the other sensing source equipment as accompanying sensing source equipment, and grouping according to the type of the sensing source equipment to obtain one-to-many virtual same-position relation between the sensing source equipment passed by all main targets and the other accompanying sensing source equipment in the offset distance range; setting offset time, taking the accurate time of the main target passing through each perception source device as a center, adding or subtracting the offset time from front to back, calculating the offset time range passing through each accompanying perception source device, and obtaining the time range of other types of accompanying perception source devices corresponding to the offset time range of the perception source device passing through the main track;

according to all the accompanying perception source devices and the offset time ranges of all the accompanying perception source devices, data of all the accompanying targets passing through the accompanying perception source devices are inquired in the second data source, all the data are aggregated according to the accompanying targets, and then time sequencing is carried out on the data of each accompanying target, so that tracks of all the accompanying targets passing through the accompanying perception source devices, namely a plurality of accompanying tracks, are obtained.

5. The method of claim 1, wherein said merging the companion trajectory according to the companion parameters comprises:

according to the one-to-many virtual same-position relation between the perception source equipment of the main track and the accompanying perception source equipment, the accompanying perception source equipment in the accompanying track is grouped into corresponding main track perception source equipment;

and according to the offset time range of each perception source device of the main track, carrying out deduplication operation on the data of the accompanying track in the offset time range, wherein only one piece of data exists in one offset time range.

6. The method of claim 5, wherein the comparing the main track and the merged companion track to obtain the companion times and the number of companion sensing source devices comprises:

counting the accompanying tracks after the merging operation, wherein the total number is the accompanying times of the main track;

and counting the accompanying tracks after the merging operation, wherein the number of different sensing source devices passing through the accompanying tracks is the number of the accompanying sensing source devices.

7. The method of claim 1, wherein the primary target is an identification card number, a license plate number, an IMSI number, or an MAC code, and the sensing source device is an identification card recognition terminal, a license plate recognition device, or an IMSI/MAC acquisition device.

8. A system for companion analysis between multi-source data, comprising:

9. An electronic apparatus, comprising a memory and a processor, the memory storing a computer program configured to be executed by the processor, the computer program comprising instructions for performing the method of any of claims 1 to 7.

10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program which, when executed by a computer, implements the method of any one of claims 1 to 7.