A method, system and terminal for analyzing the movement of visitors in a venue

Abstract

The present invention provides a method, system and terminal for analyzing the movement of visitors in a venue, comprising the following steps: calibrating cameras in the venue; dividing the venue into regions according to the coordinate positions of the calibrated cameras; re-identifying pedestrians on the image information collected by the cameras to obtain the trajectory data of visitors in different regions; processing the trajectory data of the visitors; and analyzing the movement of visitors based on the processed trajectory data of the visitors. The method, system and terminal for analyzing the movement of visitors in a venue of the present invention have good universality, are easy to analyze and present the discrete trajectories of visitors, and will not disclose the privacy information of visitors.

Description

Technical Field

[0001] The present invention belongs to the field of computer vision and relates to a pedestrian movement line analysis method, and in particular to a venue visitor movement line analysis method, system and terminal. Background Art

[0002] With the development and implementation of computer vision technology, pedestrian re-identification technology has been widely used in the field of intelligent video analysis. Pedestrian re-identification refers to retrieving the target pedestrian object from images or videos taken at different times by cameras in different locations. Pedestrian re-identification belongs to a category of ID recognition methods. By collecting biological feature information through sensors or cameras, images in complex video environments can be processed and analyzed.

[0003] Traditional pedestrian movement analysis methods are mostly based on the algorithm principle of WIFI wireless positioning. This method requires tourists to carry WIFI signal transmission devices with them, which has poor portability and experience. In addition, the positioning accuracy of the router for WIFI devices is low, so it is impossible to accurately obtain the movement coordinates of tourists within the meter level. In addition, traditional pedestrian movement analysis methods collect discrete trajectory fragments of tourists, with large data volumes and poor regularity. At the same time, they are limited by factors such as camera resolution, shooting angle, and privacy policy, making the analysis and presentation of pedestrian movement more difficult and complex.

[0004] Therefore, how to further improve and optimize the performance of existing pedestrian movement analysis methods and provide users with a more convenient, easy-to-operate and more information-secure pedestrian movement analysis method is a technical problem that researchers in this field urgently need to solve. Summary of the invention

[0005] In view of the shortcomings of the prior art mentioned above, the purpose of the present invention is to provide a method, system and terminal for analyzing the movement of visitors in venues. Compared with the existing pedestrian movement analysis method, the present invention has better universality, is easier to analyze and present the discrete trajectory data of visitors, and pays more attention to the personal privacy of visitors.

[0006] To achieve the above-mentioned purpose and other related purposes, the present invention provides a method for analyzing tourist movement lines in a venue, comprising the following steps: calibrating cameras in the venue; dividing the venue into areas according to the coordinate positions of the calibrated cameras; re-identifying pedestrians on image information collected by the cameras to obtain tourist trajectory data in different areas; processing the tourist trajectory data; and analyzing tourist movement lines based on the processed tourist trajectory data.

[0007] In one embodiment of the present invention, calibrating the cameras in the venue includes the following steps:

[0008] Calibrate the internal parameters of the camera, wherein the internal parameters include the resolution of the image captured by the camera;

[0009] The external parameters of the camera are calibrated, wherein the external parameters include the coordinates and horizontal rotation angle of the camera in the CAD base map of the venue plane.

[0010] In one embodiment of the present invention, obtaining tourist trajectory data in different areas includes the following steps:

[0011] Get each tourist’s track ID from each camera;

[0012] Count the track IDs belonging to the same tourist in different cameras;

[0013] Record the appearance time and disappearance time of each visitor's track based on the track ID;

[0014] Obtaining the coordinates of the tourist projected in the camera perspective base map;

[0015] The track appearance time, track disappearance time and the coordinates are used as the tourist track data.

[0016] In one embodiment of the present invention, the data processing of the tourist trajectory data includes deduplication of invalid passenger flow, and the deduplication of invalid passenger flow includes the following steps:

[0017] Obtaining passenger flow information within the venue, the passenger flow information including the number of times tourists appear, the cumulative duration of tourists appearing, and the proportion of grids in the area covered by historical trajectories;

[0018] Based on the passenger flow information, the non-tourist probability score of the tourist is calculated using the formula:

[0019]

[0020] Where C represents the number of times tourists appear, T represents the cumulative duration of tourists’ appearance, and S represents the proportion of grid cells in the area covered by historical trajectories;

[0021] Invalid passenger flow is removed from the tourist trajectory data according to the non-tourist possibility score.

[0022] In one embodiment of the present invention, the data processing of the tourist trajectory data includes deduplication of trajectories within a region, and the deduplication of trajectories within a region includes the following steps:

[0023] Transforming the coordinates of the tourists in the camera perspective base map collected by the camera into the coordinates in the venue plane CAD base map;

[0024] Mark the venue area where the coordinates after coordinate transformation are located;

[0025] Sorting the venue areas in chronological order;

[0026] Remove duplicates from adjacent and identical venue areas.

[0027] In one embodiment of the present invention, when the coordinates of the tourists in the camera perspective base map collected by the camera are transformed into the coordinates in the venue plane CAD base map, the formula is used:

[0028] X=X0+(w x *cosθ-w y *sinθ) / f

[0029] Y=Y0+(w y *cosθ+w x *sinθ) / f

[0030] Where (X0, Y0) is the coordinate of the camera in the CAD base map of the venue; (w x ,w y ) is the coordinate of the tourist in the camera perspective base map; θ is the horizontal rotation angle of the camera relative to the north direction; f is the scaling factor of the venue plane CAD base map compared to the real scene.

[0031] In one embodiment of the present invention, performing tourist movement line analysis based on the processed tourist trajectory data includes the following steps:

[0032] Based on a specific route screening mechanism, all tourist trajectory data are screened to obtain the same route trajectory;

[0033] Count and sort the same moving line trajectories;

[0034] The ranking from high to low will be used as the recommended ranking for tourists' walking routes.

[0035] In one embodiment of the present invention, the specific route screening mechanism needs to simultaneously meet the following two screening conditions:

[0036] Cover at least one of the venue entrance area and exit area;

[0037] Cover at least one half of the total number of areas within the venue excluding the entrance and exit areas.

[0038] Correspondingly, the present invention provides a venue visitor movement line analysis system, comprising:

[0039] Calibration module, used to calibrate the cameras in the venue;

[0040] A division module, used for dividing the venue into regions according to the calibrated coordinate positions of the cameras;

[0041] A recognition module, used to re-identify pedestrians based on the image information collected by the camera, and obtain the trajectory data of tourists in different areas;

[0042] A processing module, used for processing the tourist trajectory data;

[0043] The analysis module is used to analyze tourist movement routes based on the processed tourist trajectory data.

[0044] The present invention provides a venue visitor movement line analysis terminal, including a processor and a memory;

[0045] The memory is used to store computer programs;

[0046] The processor is used to execute the computer program stored in the memory so that the terminal executes the above-mentioned venue visitor movement line analysis method.

[0047] As described above, the venue visitor movement line analysis method, system and terminal of the present invention have the following beneficial effects:

[0048] (1) It can locate and analyze the movement trajectory of visitors in venues without the need for users to use any portable signal transmitting equipment, which has better universality;

[0049] (2) Taking into account the completeness of tourists’ activity trajectories in the entrance and exit areas and on-site areas, the discrete data is structured to make it easier to analyze and present;

[0050] (3) The time sequence of tourists passing through all venue areas is recorded without showing the actual walking routes of tourists, which effectively protects the privacy information of tourists. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 Shown is a flow chart of a method for analyzing the movement of visitors in a venue in one embodiment of the present invention.

[0052] Figure 2 Shown is a schematic diagram of the venue area division in an embodiment of the venue visitor movement line analysis method of the present invention.

[0053] Figure 3 Shown is a schematic structural diagram of a venue visitor movement line analysis system in one embodiment of the present invention.

[0054] Figure 4 Shown is a schematic structural diagram of a venue visitor movement line analysis terminal in one embodiment of the present invention.

[0055] Component number description

[0056] S1~S5 Method Steps

[0057] 31 Calibration module

[0058] 32 Divide Module

[0059] 33 Identification module

[0060] 34 Processing Module

[0061] 35 Analysis Module

[0062] 41 Processor

[0063] 42 Memory DETAILED DESCRIPTION

[0064] The following describes the embodiments of the present invention through specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention.

[0065] It should be noted that the illustrations provided in this embodiment are only used to illustrate the basic concept of the present invention in a schematic manner. Therefore, the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0066] The venue visitor movement line analysis method, system and terminal of the present invention screen and analyze the movement trajectories of tourists in the venue based on pedestrian re-identification and area division. Users do not need to use any portable signal transmitting equipment, have good universality, high visitor trajectory coverage, are easy to present and analyze tourists' discrete trajectory data, and will not disclose tourists' personal privacy.

[0067] like Figure 1 As shown, in one embodiment, the venue visitor movement line analysis method of the present invention includes the following steps:

[0068] Step S1: calibrate the cameras in the venue.

[0069] In one embodiment, when calibrating a camera in a venue, firstly, the internal parameters of the camera are calibrated, and the internal parameters include the resolution of the image taken by the camera; then, the external parameters of the camera are calibrated, and the external parameters include the coordinates and horizontal rotation angle of the camera in the CAD base map of the venue.

[0070] Specifically, when calibrating the resolution of the image taken by the camera, the fixed parameter can be confirmed according to the camera model, or it can be automatically identified through the camera supporting software; the coordinates of the camera in the venue plan CAD base map can be confirmed according to the actual installation point map and CAD software, and the horizontal rotation angle can be measured with the help of measurable devices such as mobile phones or rotation angle azimuths.

[0071] Step S2: dividing the venue into regions according to the calibrated coordinate positions of the cameras.

[0072] Specifically, when performing area division, users can adopt different area division methods according to the differences in actual venue scenarios, including fixed area division, range area division, and irregular area division.

[0073] For example, Figure 2 The range area division method with the camera as the center and a specific distance as the radius is adopted. The venue plane CAD base map is divided into C1 to C88 circular areas according to the coverage position of the camera. The origin of each area is the calibrated camera coordinate position, and the radius of each area is the same; each area is marked with a different number, and each area is independent of each other in the base map without overlap.

[0074] Step S3: re-identify pedestrians on the image information collected by the camera to obtain tourist trajectory data in different areas.

[0075] Specifically, a metric learning-based method is used to re-identify tourists at different times and in different areas: first, the full-body feature information of the tourist is obtained from the tourist trajectory ID; then, the full-body feature information of the tourist is compared with the feature information in the local database one by one, and the distance similarity between the two is measured; the similarity threshold is set to 80%, and if the similarity is greater than 80%, it is determined that the feature information belongs to the same tourist.

[0076] In one embodiment, obtaining tourist trajectory data in different areas includes the following steps:

[0077] 31) Get the trajectory ID of each tourist from each camera.

[0078] Specifically, the trajectory ID represents a single tourist image and is stored in the form of a binary feature vector.

[0079] 32) Count the track IDs belonging to the same tourist in different cameras.

[0080] Specifically, the trajectory IDs with the same binary feature vectors are identified as the trajectory IDs of the same tourist.

[0081] 33) Based on the track ID, record the track appearance time and track disappearance time of each tourist.

[0082] 34) Obtain the coordinates of the tourist projected in the camera perspective base map.

[0083] Specifically, the camera base map is mapped into a two-dimensional plane, and the coordinates of the tourists in the plane coordinate system are obtained by using CAD software, wherein the origin of the plane coordinate system is the intersection of the camera optical axis and the base map plane.

[0084] The above-mentioned track ID, track appearance time, track disappearance time and coordinates are collectively regarded as the tourist track data.

[0085] Step S4: Process the tourist trajectory data.

[0086] In one embodiment of the present invention, the data processing of the tourist trajectory data includes deduplication of invalid passenger flow, and the deduplication of invalid passenger flow includes the following steps:

[0087] (1) Obtaining passenger flow information within the venue, wherein the passenger flow information includes the number of times tourists appear, the cumulative duration of tourists appearing, and the proportion of grid cells in the area covered by historical trajectories.

[0088] Preferably, the number of times a tourist appears, the cumulative duration of the tourist's appearance and the proportion of the grid in the area covered by the historical trajectory are calculated based on the tourist trajectory data.

[0089] (2) Calculate the non-tourist probability score of the tourist based on the passenger flow information, using the formula:

[0090]

[0091] Where C represents the number of times tourists appear, T represents the cumulative duration of tourists’ appearance, and S represents the proportion of regional grids covered by historical trajectories.

[0092] In one embodiment, the formula is used to calculate the grid ratio of the historical trajectory coverage area:

[0093]

[0094] Where N represents the number of uniformly sized grids contained in the monitoring screen, S hit Indicates the number of grids hit by the tourist trajectory.

[0095] (3) Eliminate invalid passenger flow from the tourist trajectory data based on the non-tourist possibility score.

[0096] Preferably, 0.8 is used as the threshold of the non-tourist possibility score. If F is greater than the preset threshold, the independent individual is determined to be a non-tourist.

[0097] In one embodiment of the present invention, the data processing of the tourist trajectory data includes deduplication of trajectories within a region, and the deduplication of trajectories within a region includes the following steps:

[0098] (1) The coordinates of the tourists in the camera perspective base map collected by the camera are transformed into the coordinates in the venue plane CAD base map.

[0099] In one embodiment, when the coordinates of the tourist in the camera perspective base map collected by the camera are transformed into the coordinates in the venue plane CAD base map, the formula is used:

[0100] X=X0+(w x *cosθ-w y *sinθ) / f

[0101] Y=Y0+(w y *cosθ+w x *sinθ) / f

[0102] Where (X0, Y0) is the coordinate of the camera in the CAD base map of the venue; (w x ,w y ) is the coordinate of the tourist in the camera perspective base map; θ is the horizontal rotation angle of the camera; f is the scaling factor of the venue plane CAD base map compared to the real scene.

[0103] (2) Mark the venue area where the coordinates after coordinate transformation are located.

[0104] For example, the coordinates of a certain tourist in the camera perspective base map are transformed, and the transformed tourist coordinates are shown to be located in area C4 in the venue plan base map, then the venue area where the tourist appears in this time period is marked as C4.

[0105] (3) Sort the venue areas in chronological order.

[0106] For example, a tourist appears in venue areas C4, C7, C7, C6 and C5 at different time periods. By arranging the corresponding target areas in chronological order, the regional trajectory path of the tourist can be obtained: C4->C7->C7->C6->C5.

[0107] (4) De-duplication of adjacent and identical venue areas

[0108] For example, in the above example of venue area sorting, there are two adjacent and identical C7 areas. One of the C7 areas is removed to obtain a non-repeating area trajectory path for the tourist: C4->C7->C6->C5.

[0109] Step S5: Perform tourist movement line analysis based on the processed tourist trajectory data.

[0110] In one embodiment of the present invention, performing tourist movement line analysis based on the processed tourist trajectory data includes the following steps:

[0111] (1) Based on a specific route filtering mechanism, all tourist trajectory data are filtered to obtain the same route trajectory.

[0112] In one embodiment, the specific route screening mechanism requires satisfying the following two screening conditions at the same time:

[0113] Cover at least one of the venue entrance area and exit area;

[0114] Cover at least one half of the total number of areas within the venue excluding the entrance and exit areas.

[0115] (2) Count and sort the same moving line trajectories.

[0116] Preferably, the trajectories are sorted according to the number of times the same moving line trajectory occurs.

[0117] (3) The routes are ranked from high to low as the recommended order for tourists’ walking routes.

[0118] like Figure 3 As shown, in one embodiment, the venue visitor movement line analysis system of the present invention includes:

[0119] The calibration module 31 is used to calibrate the cameras in the venue.

[0120] In one embodiment, when calibrating a camera in a venue, firstly, the internal parameters of the camera are calibrated, and the internal parameters include the resolution of the image taken by the camera; then, the external parameters of the camera are calibrated, and the external parameters include the coordinates and horizontal rotation angle of the camera in the CAD base map of the venue.

[0121] Specifically, when calibrating the resolution of the image taken by the camera, the fixed parameter can be confirmed according to the camera model, or it can be automatically identified through the camera supporting software; the coordinates of the camera in the venue plan CAD base map can be confirmed according to the actual installation point map and CAD software, and the horizontal rotation angle can be measured with the help of measurable devices such as mobile phones or rotation angle azimuths.

[0122] The division module 32 is connected to the calibration module 31 , and is used to divide the venue into regions according to the coordinate positions of the calibrated cameras.

[0123] Specifically, when dividing areas, users can adopt different area division methods according to the actual venue scenario, including fixed area division, range area division, and irregular area division.

[0124] For example, Figure 2 The range area division method with the camera as the center and a specific distance as the radius is adopted. The venue plane CAD base map is divided into C1 to C88 circular areas according to the coverage position of the camera. The origin of each area is the calibrated camera coordinate position, and the radius of each area is the same; each area is marked with a different number, and each area is independent of each other in the base map without overlap.

[0125] The recognition module 33 is connected to the division module 32, and the recognition module 33 is used to re-identify pedestrians on the image information collected by the camera to obtain the trajectory data of tourists in different areas.

[0126] Specifically, a metric learning-based method is used to re-identify tourists at different times and in different areas: first, the full-body feature information of the tourist is obtained from the tourist trajectory ID; then, the full-body feature information of the tourist is compared with the feature information in the local database one by one, and the distance similarity between the two is measured; the similarity threshold is set to 80%, and if the similarity is greater than 80%, it is determined that the feature information belongs to the same tourist.

[0127] In one embodiment, obtaining tourist trajectory data in different areas includes the following steps:

[0128] 31) Get the trajectory ID of each tourist from each camera.

[0129] Specifically, the trajectory ID represents a single tourist image and is stored in the form of a binary feature vector.

[0130] 32) Count the track IDs belonging to the same tourist in different cameras.

[0131] Specifically, the trajectory IDs with the same binary feature vectors are identified as the trajectory IDs of the same tourist.

[0132] 33) Based on the track ID, record the track appearance time and track disappearance time of each tourist.

[0133] 34) Obtain the coordinates of the tourist projected in the camera perspective base map.

[0134] Specifically, the camera base map is mapped into a two-dimensional plane, and the coordinates of the tourists in the plane coordinate system are obtained by using CAD software, wherein the origin of the plane coordinate system is the intersection of the camera optical axis and the base map plane.

[0135] The above-mentioned track ID, track appearance time, track disappearance time and coordinates are collectively regarded as the tourist track data.

[0136] The processing module 34 is connected to the identification module 33, and the processing module 34 is used to process the tourist trajectory data.

[0137] In one embodiment of the present invention, the data processing of the tourist trajectory data includes deduplication of invalid passenger flow, and the deduplication of invalid passenger flow includes the following steps:

[0138] (1) Obtaining passenger flow information within the venue, wherein the passenger flow information includes the number of times tourists appear, the cumulative duration of tourists appearing, and the proportion of grid cells in the area covered by historical trajectories.

[0139] Preferably, the number of times a tourist appears, the cumulative duration of the tourist's appearance and the proportion of the grid in the area covered by the historical trajectory are calculated based on the tourist trajectory data.

[0140] (2) Calculate the non-tourist probability score of the tourist based on the passenger flow information, using the formula:

[0141]

[0142] Where C represents the number of times tourists appear, T represents the cumulative duration of tourists’ appearance, and S represents the proportion of regional grids covered by historical trajectories.

[0143] In one embodiment, the formula is used to calculate the grid ratio of the historical trajectory coverage area:

[0144]

[0145] Where N represents the number of uniformly sized grids contained in the monitoring screen, S hit Indicates the number of grids hit by the tourist trajectory.

[0146] (3) Eliminate invalid passenger flow from the tourist trajectory data based on the non-tourist possibility score.

[0147] Preferably, 0.8 is used as the threshold of the non-tourist possibility score. If F is greater than the preset threshold, the independent individual is determined to be a non-tourist.

[0148] In one embodiment of the present invention, the data processing of the tourist trajectory data includes deduplication of trajectories within a region, and the deduplication of trajectories within a region includes the following steps:

[0149] (1) The coordinates of the tourists in the camera perspective base map collected by the camera are transformed into the coordinates in the venue plane CAD base map.

[0150] In one embodiment, when the coordinates of the tourist in the camera perspective base map collected by the camera are transformed into the coordinates in the venue plane CAD base map, the formula is used:

[0151] X=X0+(w x *cosθ-w y *sinθ) / f

[0152] Y=Y0+(w y *cosθ+w x *sinθ) / f

[0153] Where (X0, Y0) is the coordinate of the camera in the CAD base map of the venue; (w x ,w y ) is the coordinate of the tourist in the camera perspective base map; θ is the horizontal rotation angle of the camera; f is the scaling factor of the venue plane CAD base map compared to the real scene.

[0154] (2) Mark the venue area where the coordinates after coordinate transformation are located.

[0155] For example, the coordinates of a certain tourist in the camera perspective base map are transformed as described above, and the transformed tourist coordinates are shown to be located in area C4 in the venue plan base map, then the venue area where the tourist appears in this time period is marked as C4.

[0156] (3) Sort the venue areas in chronological order.

[0157] For example, a tourist appears in venue areas C4, C7, C7, C6 and C5 at different time periods. By arranging the corresponding target areas in chronological order, the regional trajectory path of the tourist can be obtained: C4->C7->C7->C6->C5.

[0158] (4) De-duplication of adjacent and identical venue areas

[0159] For example, in the above example of venue area sorting, there are two adjacent and identical C7 areas. One of the C7 areas is removed to obtain a non-repeating area trajectory path for the tourist: C4->C7->C6->C5.

[0160] The analysis module 35 is connected to the recognition module 34, and the analysis module 35 is used to perform tourist movement line analysis based on the tourist trajectory data after data processing.

[0161] In one embodiment of the present invention, performing tourist movement line analysis based on the processed tourist trajectory data includes the following steps:

[0162] (1) Based on a specific route filtering mechanism, all tourist trajectory data are filtered to obtain the same route trajectory.

[0163] Preferably, the specific route screening mechanism needs to satisfy the following two screening conditions at the same time:

[0164] Cover at least one of the venue entrance area and exit area;

[0165] Cover at least one half of the total number of areas within the venue excluding the entrance and exit areas.

[0166] (2) Count and sort the same moving line trajectories.

[0167] Preferably, the same moving line trajectories are counted and sorted from high to low.

[0168] (3) The routes are ranked from high to low as the recommended order for tourists’ walking routes.

[0169] like Figure 4 As shown, in one embodiment of the present invention, the venue visitor route terminal of the present invention further includes a processor 41 and a memory 42 .

[0170] The memory 41 is used to store computer programs;

[0171] The memory 41 includes: ROM, RAM, disk, USB flash drive, memory card or optical disk and other media that can store program codes.

[0172] The processor 42 is used to execute the computer program stored in the memory, so that the terminal executes the above-mentioned venue visitor movement line analysis method. Specifically, after the movement line analysis is performed on the visitors of the current venue, the recommended ranking of the visitors' walking routes is stored in the memory.

[0173] Preferably, the processor 42 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0174] It should be noted that it should be understood that the division of the various modules of the above device is only a division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated. And these modules can all be implemented in the form of software called by processing elements; they can also be all implemented in the form of hardware; some modules can also be implemented in the form of software called by processing elements, and some modules can be implemented in the form of hardware. For example, the x module can be a separately established processing element, or it can be integrated in a chip of the above device. In addition, it can also be stored in the memory of the above device in the form of program code, and called and executed by a processing element of the above device. The implementation of other modules is similar. In addition, these modules can be fully or partially integrated together, or they can be implemented independently. The processing element described here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each module above can be completed by an integrated logic circuit of hardware in the processor element or instructions in the form of software.

[0175] For example, the above modules may be one or more integrated circuits configured to implement the above methods, such as one or more application specific integrated circuits (ASICs), or one or more digital signal processors (DSPs), or one or more field programmable gate arrays (FPGAs). For another example, when a module above is implemented in the form of a processing element scheduling program code, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processors that can call program codes. For another example, these modules may be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0176] In summary, the venue visitor movement line analysis method, system and terminal of the present invention locate and analyze the movement trajectory of tourists in the venue based on regional division and pedestrian re-identification. The method provided by the present invention has high universality and low risk of privacy leakage. The obtained ranking of tourists' walking routes can reflect the rationality of the venue's layout and the preferences of tourists to a certain extent, help improve the utilization rate of the venue space, and provide reference suggestions for the design and optimization of venue functions. Therefore, the present invention effectively overcomes the various shortcomings of the prior art and has a high industrial utilization value.

[0177] The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the present invention. Anyone familiar with the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by a person of ordinary skill in the art without departing from the spirit and technical concept disclosed by the present invention shall still be covered by the claims of the present invention.

Claims

1. A method for analyzing the movement of visitors in a venue, comprising the following steps: Calibrate the cameras in the venue; Dividing the venue into regions according to the calibrated coordinate positions of the cameras; Re-identify pedestrians on the image information collected by the camera to obtain tourist trajectory data in different areas; Performing data processing on the tourist trajectory data; Analyze tourist movement based on the processed tourist trajectory data; The data processing of the tourist trajectory data includes deduplication of trajectories within a region, and the deduplication of trajectories within a region includes the following steps: Transforming the coordinates of the tourists in the camera perspective base map collected by the camera into the coordinates in the venue plane CAD base map; Mark the venue area where the coordinates after coordinate transformation are located; Sorting the venue areas in chronological order; Remove duplicates from adjacent and identical venue areas.

2. The method for analyzing the movement of visitors in a venue according to claim 1, characterized in that: Calibration of cameras in the venue includes the following steps: Calibrate the internal parameters of the camera, wherein the internal parameters include the resolution of the image captured by the camera; The external parameters of the camera are calibrated, wherein the external parameters include the coordinates and horizontal rotation angle of the camera in the CAD base map of the venue plane.

3. The method for analyzing the movement of visitors in a venue according to claim 1, characterized in that: Obtaining tourist trajectory data in different areas includes the following steps: Get each tourist’s track ID from each camera; Count the track IDs belonging to the same tourist in different cameras; Record the appearance time and disappearance time of each visitor's track based on the track ID; Obtaining the coordinates of the tourist projected in the camera perspective base map; The track appearance time, track disappearance time and the coordinates are used as the tourist track data.

4. The method for analyzing the movement of visitors in a venue according to claim 1, characterized in that: The data processing of the tourist trajectory data includes deduplication of invalid passenger flow, and the deduplication of invalid passenger flow includes the following steps: Obtaining passenger flow information within the venue, the passenger flow information including the number of times tourists appear, the cumulative duration of tourists appearing, and the proportion of grids in the area covered by historical trajectories; Based on the passenger flow information, the non-tourist probability score of the tourist is calculated using the formula: Where C represents the number of times tourists appear, T represents the cumulative duration of tourists’ appearance, and S represents the proportion of grid cells in the area covered by historical trajectories; Invalid passenger flow is removed from the tourist trajectory data according to the non-tourist possibility score.

5. The method for analyzing the movement of visitors in a venue according to claim 1, characterized in that: When the coordinates of the tourists in the camera perspective base map collected by the camera are transformed into the coordinates in the venue plane CAD base map, the formula is used: X=X0+(w x *cosθ-w y *sinθ) / f Y=Y0+(w y *cosθ+w x *sinθ) / f Where (X0, Y0) is the coordinate of the camera in the CAD base map of the venue; (w x , w y ) is the coordinate of the tourist in the camera perspective base map; θ is the horizontal rotation angle of the camera relative to the north direction; f is the scaling factor of the venue plane CAD base map compared to the real scene.

6. The method for analyzing the movement of visitors in a venue according to claim 1, characterized in that: The tourist movement analysis based on the processed tourist trajectory data includes the following steps: Based on a specific route screening mechanism, all tourist trajectory data are screened to obtain the same route trajectory; Count and sort the same moving line trajectories; The ranking from high to low will be used as the recommended ranking for tourists' walking routes.

7. The method for analyzing the movement of visitors in a venue according to claim 6, characterized in that: The specific route screening mechanism requires the following two screening conditions to be met at the same time: Cover at least one of the venue entrance area and exit area; Cover at least one half of the total number of areas within the venue excluding the entrance and exit areas.

8. A venue visitor movement line analysis system, characterized in that: include: Calibration module, used to calibrate the cameras in the venue; A division module, dividing the venue into regions according to the calibrated coordinate positions of the cameras; A recognition module, used to re-identify pedestrians based on the image information collected by the camera, and obtain the trajectory data of tourists in different areas; A processing module, used for processing the tourist trajectory data; An analysis module is used to analyze tourist movement routes based on the processed tourist trajectory data; The data processing of the tourist trajectory data includes deduplication of trajectories within a region, and the deduplication of trajectories within a region includes the following steps: Transforming the coordinates of the tourists in the camera perspective base map collected by the camera into the coordinates in the venue plane CAD base map; Mark the venue area where the coordinates after coordinate transformation are located; Sorting the venue areas in chronological order; Remove duplicates from adjacent and identical venue areas.

9. A terminal for analyzing the movement of visitors in a venue, characterized in that: include: Processor and memory; The memory is used to store computer programs; The processor is used to execute the computer program stored in the memory so that the terminal executes the venue visitor movement line analysis method as described in any one of claims 1 to 7.

Images