WO2023029574A1 - 获取客流信息的方法、装置、计算机设备以及存储介质 - Google Patents

获取客流信息的方法、装置、计算机设备以及存储介质 Download PDF

Info

Publication number
WO2023029574A1
WO2023029574A1 PCT/CN2022/092492 CN2022092492W WO2023029574A1 WO 2023029574 A1 WO2023029574 A1 WO 2023029574A1 CN 2022092492 W CN2022092492 W CN 2022092492W WO 2023029574 A1 WO2023029574 A1 WO 2023029574A1
Authority
WO
WIPO (PCT)
Prior art keywords
passenger flow
target area
target
target object
information
Prior art date
Application number
PCT/CN2022/092492
Other languages
English (en)
French (fr)
Inventor
范畅
Original Assignee
上海商汤智能科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 上海商汤智能科技有限公司 filed Critical 上海商汤智能科技有限公司
Publication of WO2023029574A1 publication Critical patent/WO2023029574A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30196—Human being; Person
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30242—Counting objects in image

Definitions

  • the present disclosure relates to the technical field of computer vision, in particular, to a method, device, computer equipment and storage medium for acquiring passenger flow information.
  • Passenger flow statistics can give information such as the passenger flow of a certain place within a certain period of time.
  • passenger flow identification and statistics are important research directions in the field of computer vision.
  • passenger flow statistics based on computer vision are mostly positive passenger flow statistics, that is, the camera needs to be able to directly capture the places where passenger flow statistics are required, which is difficult to meet some requirements. The requirement that cameras cannot be installed inside, for example, high-end shopping malls cannot install cameras in brand stores.
  • Embodiments of the present disclosure at least provide a method, device, computer equipment, and storage medium for acquiring passenger flow information.
  • the embodiment of the present disclosure provides a method for obtaining passenger flow information, including:
  • Passenger flow information in the target area is determined based on the running track of the target object and the passenger flow determination flag.
  • the running trajectory of the target object generated in the preset area adjacent to the entrance and exit of the target area and not in the target area, and the passenger flow judgment sign corresponding to the target area it is possible to predict more accurately the relative The specific behavior of the target area, such as predicting that the target object enters or leaves the target area, and then the passenger flow information in the target area can be obtained when the target device does not capture the target area.
  • the determining passenger flow information in the target area based on the running track of the target object and the passenger flow determination flag includes:
  • passenger flow information in the target area is determined.
  • the target object has the behavior of entering or leaving the target area.
  • the specific behavior of the target object such as entering or leaving the target area, can more accurately obtain the passenger flow information in the target area that the target device cannot capture, and realize the passenger flow statistics of the target area.
  • the passenger flow information includes effective passenger flow in the target area
  • the determining the passenger flow information in the target area based on the moving direction includes:
  • the moving direction includes a first direction indicating that the target object enters the target area and a second direction indicating that the target object leaves the target area, the corresponding target object is used as the target Effective passenger flow objects in the area;
  • the quantity of the effective passenger flow objects is taken as the effective passenger flow in the target area.
  • the moving direction includes the first direction
  • the moving direction includes the second direction
  • the target object is an effective passenger flow object in the target area. Therefore, the above-mentioned embodiment can accurately count the effective passenger flow of the target area by judging whether the moving direction includes the first direction and the second direction. Passenger flow, and then targeted management target areas.
  • the effective passenger flow of brand stores can be counted to assist shopping mall operators in evaluating the store's passenger flow; or provide business reference for the store Data, such as a large number of customers entering the store but low sales, one possible reason is that the goods do not meet customer needs.
  • the management personnel can make timely adjustments to the goods based on the business reference data.
  • the passenger flow information further includes the length of stay of the valid passenger flow object in the target area
  • the determining the passenger flow information in the target area based on the moving direction further includes:
  • the time difference between the first time and the second time can be used to determine the effective passenger flow object in the target area.
  • the longer the stay time the more targeted management target areas can be achieved.
  • the length of stay of at least some (for example, each) customers in the store can be counted. If the stay time is short, it means that the goods in the store do not meet the needs of the customer. At this time, the store can Targeted design of goods that meet the requirements of customers with a short stay.
  • the passenger flow information further includes the average length of stay of the valid passenger flow objects in the target area
  • the determining the passenger flow information in the target area based on the moving direction further includes:
  • an average length of stay of the effective passenger flow objects in the target area is determined.
  • the average length of stay of valid passenger flow objects in the target area can be accurately determined by using the length of stay of at least some (for example, each) target objects in the target area. Furthermore, through the determined average length of stay of the effective passenger flow objects in the target area, the target area can be managed in a targeted manner. For example, in the application scenario of high-end shopping malls, the average length of stay of customers who visit the store within a day can be counted. If the average length of stay is longer, it means that the goods in the store are more in line with customer needs, and the goods in this period can be focused on promotional goods .
  • the passenger flow information further includes at least one of the passenger flow entering and leaving the target area
  • the determining the passenger flow information in the target area based on the moving direction includes at least one of the following:
  • the moving direction includes a first direction indicating that the target object enters the target area
  • the corresponding target object is used as an incoming passenger flow object in the target area; the number of the incoming passenger flow objects, as incoming foot traffic within said target area;
  • the moving direction includes a second direction indicating that the target object leaves the target area
  • the corresponding target object is used as a passenger flow object leaving the target area; the number of the passenger flow objects leaving the target area is, as departure traffic within the target area.
  • the moving direction includes the first direction
  • the number of incoming passenger flow objects entering the target area can be accurately counted
  • the moving direction includes the second direction the number of departing passenger flow objects leaving the target area can be accurately counted number of objects.
  • the identifying the target object in the video stream and determining the running track of the target object includes:
  • identifying a target object in the video stream determining a track point of the target object and timing information of the track point
  • the running track of the target object is determined.
  • the track points and the timing information of each track point can be used to accurately restore the running track of the target object and prepare for predicting the behavior of the target object.
  • the determining the passenger flow determination flag corresponding to the target area includes:
  • the passenger flow determination flag corresponding to the target area is determined based on the relative position information and the position information of the preset area.
  • the exact position of the passenger flow judgment sign can be determined through the above-mentioned relative position information and the position information of the preset area, and the behavior of the target object can be predicted by using the passenger flow judgment sign at this position, which can improve the accuracy of predicting the behavior of the target object .
  • the method further includes:
  • the passenger flow statistics report includes the passenger flow information, or the passenger flow information and at least one of the following: report identification information, video stream display device ID, target device ID, and target area ID.
  • the passenger flow statistical report can accurately and comprehensively reflect the relevant information of the passenger flow in the target area, and further, the target area can be managed in a targeted manner by using the passenger flow statistical report.
  • the embodiment of the present disclosure also provides a device for obtaining passenger flow information, including:
  • a video stream acquisition module configured to acquire a video stream obtained by at least one target device shooting a preset area; the preset area is adjacent to the entrance and exit of the target area, and is not within the target area;
  • a sign determination module configured to determine a passenger flow determination sign corresponding to the target area
  • a running track determining module configured to identify the target object in the video stream, and determine the running track of the target object
  • the passenger flow information determining module is configured to determine passenger flow information in the target area based on the running trajectory of the target object and the passenger flow determination flag.
  • the passenger flow information determining module is configured to determine the moving direction of the running trajectory relative to the passenger flow judgment sign when it is determined that the running trajectory intersects with the passenger flow judgment sign ; Based on the moving direction, determine passenger flow information in the target area.
  • the passenger flow information includes effective passenger flow in the target area
  • the passenger flow information determining module is configured to, when the moving direction includes a first direction indicating that the target object enters the target area and a second direction indicating that the target object leaves the target area, corresponding
  • the target object is used as the effective passenger flow object in the target area; the number of the effective passenger flow objects is used as the effective passenger flow in the target area.
  • the passenger flow information further includes the length of stay of the valid passenger flow object in the target area
  • the passenger flow information determination module is further configured to respectively determine that the running trajectory of the effective passenger flow object intersects with the passenger flow judgment mark in the first direction based on the running track of the effective passenger flow object and the passenger flow judgment mark and determine the second time when the running track of the effective passenger flow object intersects with the passenger flow judgment mark in the second direction; based on the first time and the second time, determine the The length of stay of valid passenger flow objects in the target area.
  • the passenger flow information further includes the average length of stay of the valid passenger flow objects in the target area
  • the passenger flow information determination module is further configured to determine the average stay of the effective passenger flow objects in the target area based on the length of stay of at least some of the effective passenger flow objects in the target area and the number of the effective passenger flow objects duration.
  • the passenger flow information further includes at least one of the passenger flow entering and leaving the target area
  • the passenger flow information determining module is further configured to use the corresponding target object as a passenger flow object entering the target area when the moving direction includes a first direction indicating that the target object enters the target area ; using the quantity of the incoming passenger flow object as the incoming passenger flow in the target area; when the moving direction includes a second direction indicating that the target object leaves the target area, the corresponding The target object is used as the departing passenger flow object in the target area; the number of the departing passenger flow objects is taken as the departing passenger flow in the target area.
  • the running track determining module is configured to identify a target object in the video stream, determine the track point of the target object and the timing information of the track point; based on the target object The track points and the timing information of the track points are used to determine the running track of the target object.
  • the sign determination module is configured to determine relative position information of the preset area and the entrance and exit of the target area; based on the relative position information and the position information of the preset area , determining the passenger flow determination flag corresponding to the target area.
  • the device for obtaining passenger flow information further includes a report generation module, configured to determine the passenger flow information in the target area based on the running trajectory of the target object and the passenger flow determination flag , generating and uploading a passenger flow statistics report based on the passenger flow information;
  • the passenger flow statistics report includes the passenger flow information, or includes the passenger flow information and at least one of the following: report identification information, video stream display device ID, target device ID, and target area ID.
  • an optional implementation manner of the present disclosure further provides a computer device, a processor, and a memory, the memory stores machine-readable instructions executable by the processor, and the processor is used to execute the instructions stored in the memory. machine-readable instructions, when the machine-readable instructions are executed by the processor, when the machine-readable instructions are executed by the processor, the above-mentioned first aspect is executed, or any possible implementation of the first aspect steps in the method.
  • an optional implementation manner of the present disclosure further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed, the above-mentioned first aspect, or any one of the first aspects in the first aspect, may be executed. Steps in one possible implementation.
  • an optional implementation manner of the present disclosure further provides a computer program product, including computer-readable codes, or a computer-readable storage medium bearing computer-readable codes, when the computer-readable codes are processed in an electronic device
  • the processor in the electronic device executes the above first aspect, or the steps in any possible implementation manner of the first aspect.
  • FIG. 1 shows a flow chart of a method for obtaining passenger flow information provided by an embodiment of the present disclosure
  • FIG. 2 shows a schematic layout of a preset area and a target area provided by an embodiment of the present disclosure
  • Fig. 3a shows a schematic diagram of a scenario in which the preset area is adjacent to the entrance and exit rules of the target area provided by the embodiment of the present disclosure
  • Fig. 3b shows a schematic diagram of a scenario in which the preset area and the entrance and exit of the target area are irregularly adjacent to each other provided by an embodiment of the present disclosure
  • Fig. 4 shows a flow chart of determining passenger flow information in a target area provided by an embodiment of the present disclosure
  • Fig. 5 shows a schematic diagram of the running track of the target object in the preset area provided by the embodiment of the present disclosure
  • Fig. 6 shows a schematic diagram of a device for obtaining passenger flow information provided by an embodiment of the present disclosure
  • Fig. 7 shows a schematic structural diagram of a computer device provided by an embodiment of the present disclosure.
  • passenger flow statistics is a method based on computer vision technology to count relevant task data.
  • identification and statistics of passenger flow are important research directions in the field of computer vision.
  • passenger flow statistics based on computer vision are mostly positive passenger flow statistics, that is, the camera needs to be able to shine on the places where passenger flow statistics are required, and it is difficult to meet some unsatisfactory requirements.
  • the need to install camera areas inside, for example, high-end shopping malls cannot install cameras in brand stores.
  • the present disclosure provides a method for obtaining passenger flow information, through the running track of the target object generated in the preset area adjacent to the entrance and exit of the target area and not in the target area, and the passenger flow corresponding to the target area Judgment flags can accurately predict the specific behavior of the target object relative to the target area, such as predicting that the target object enters or leaves the target area, and then can obtain the information in the target area when the target device does not capture the target area. Passenger flow information.
  • the method for obtaining passenger flow information provided by the embodiments of the present disclosure is generally executed by a computer with certain computing power device, in some possible implementation manners, the method for acquiring passenger flow information may be implemented by calling a computer-readable instruction stored in a memory by a processor.
  • the method for obtaining passenger flow information provided by the embodiments of the present disclosure will be described below by taking a computer device as an example for executing the subject.
  • FIG. 1 is a flow chart of a method for obtaining passenger flow information provided by an embodiment of the present disclosure
  • the method includes steps S101 to S104, wherein:
  • S101 Obtain a video stream obtained by at least one target device shooting a preset area; the preset area is adjacent to an entrance and exit of the target area, and is not within the target area.
  • the target device may be a camera or other shooting device used for image/video collection.
  • the shooting range of the target device may be a preset area, or a partial area within the preset area.
  • the preset area may be an area photographed by at least one target device; or, may be a region of interest (Region Of Interest, ROI) in the pre-set area photographed by at least one target device.
  • the preset areas are all adjacent to the entrance and exit of the target area.
  • the target area can be an area where passenger flow statistics need to be counted.
  • the user can upload a computer-aided design (Computer Aided Design, CAD) map of the area where passenger flow statistics need to be counted to the system, and draw the target area on the CAD map.
  • CAD computer-aided design
  • the target area Areas are areas that cannot be captured by the target device or areas that are difficult to capture completely, such as brand stores and other areas where cameras cannot be installed.
  • the target area may be associated with the target device.
  • One or more target devices may be installed outside a target area, and the installation location may be set according to actual application scenarios, which are not specifically limited in the embodiments of the present disclosure.
  • the location of the preset area can be selected according to the entrances and exits of the target area.
  • the area where the preset area is located can cover all target objects entering or leaving the target area.
  • the purpose of setting the preset area adjacent to the entrance and exit of the target area is to use the target device to capture the running track of the target object in the preset area that is about to enter or has left the target area.
  • FIG. 2 is a schematic layout diagram of a preset area and a target area. Wherein, the dotted line part indicates the entrance and exit of the target area.
  • multiple target devices may be set to shoot the preset area, so as to obtain a complete video stream obtained by shooting the preset area.
  • the target area includes two entrances and exits as an example.
  • the number of entrances and exits included in the target area may be one or more, which is not limited here.
  • a single entrance can allow people to enter or exit, or only allow people to enter or leave, which can be determined in combination with the number of entrances and exits included in the target area and preset rules, and is not limited here.
  • the entrance and exit have a single function, that is, the entrance and exit only support entry or exit, then there must be two entrances and exits with different functions among the multiple entrances and exits, that is, for people to enter and leave respectively, then the two entrances and exits can be respectively corresponding.
  • the data of the set area is aggregated, for example, the effective passenger flow is determined by combining the incoming passenger flow and the outgoing passenger flow respectively obtained by the two entrances and exits corresponding to the preset areas, which are not limited here.
  • At least one target device associated with the target area is first queried, and the identification information of the target device is determined, such as the identity document (Identity Document, ID) of the camera , after that, at least one target device can be used to directly obtain the video stream obtained by shooting the preset area.
  • the video stream shot by the target device can include but not limited to real time streaming protocol (Real Time Streaming Protocol, RTSP) The transmitted video stream.
  • RTSP Real Time Streaming Protocol
  • the video stream transmitted by RTSP is input to the deep neural network model for decoding processing to obtain the decoded image of each frame, and each frame of image records the target object appearing in the preset area.
  • the detection of target objects in irrelevant areas can be reduced, which is beneficial to improve detection efficiency and save computing resources.
  • the passenger flow judgment signs may include but not limited to marking lines or areas.
  • the marking line or the area can include multiple forms, specifically, the marking line can include regular marking lines (such as straight lines, etc.), or irregular marking lines (such as curves, etc.); the area can include regular areas (such as rectangular areas, circular area, elliptical area, etc.), or irregular area (such as irregular polygon, etc.).
  • Passenger flow judgment marks can be based on the position of the target area and the position of the preset area, the marking line and/or area drawn on the image corresponding to the preset area (can be manually drawn, automatically drawn, or manually drawn based on the automatically drawn content) /obtained by automatic correction), the position of the passenger flow judgment sign can be in the position adjacent to the preset area and the entrance and exit of the target area.
  • FIG. 3 b which is a schematic diagram showing a scene where the preset area and the entrance and exit of the target area are irregularly adjacent, wherein the passenger flow determination mark is a curve.
  • the drawing of the passenger flow determination sign in the embodiment of the present disclosure is not limited to the above two application scenarios, and the shape of the passenger flow determination sign can be drawn arbitrarily, which is not limited in the embodiment of the present disclosure.
  • the judgment direction of the passenger flow judgment mark can be set according to the relative position information between the preset area and the target area.
  • the determination direction of the passenger flow determination sign may include a first direction from the preset area to the target area and a second direction from the target area to the preset area.
  • the schematic diagram showing the first direction and the second direction can be referred to as shown in FIG. Can be used to indicate the direction the target object is leaving from the target area.
  • the system can also automatically draw passenger flow judgment marks, specifically determine the relative position information of the entrance and exit of the preset area and the target area; determine the passenger flow judgment marks based on the relative position information and the position information of the preset area.
  • the relative position information of the entrance and exit of the preset area and the target area is used.
  • the relative position information is that the preset area is adjacent to the entrance and exit of the target area.
  • S103 Identify the target object in the video stream, and determine the running track of the target object.
  • the running trajectory of the target object may be formed by connecting the trajectory points of the target object according to the sequence of time nodes generated by each trajectory point.
  • the target object in the video stream can be identified according to the following steps:
  • the feature points of at least one preset part of the target object may be acquired; based on the feature points of the at least one preset part, the target object in the video stream is determined. It should be noted that these feature points can be stored in real time after being obtained based on the video stream.
  • the real-time storage process can be to store the feature points of at least one preset part of each target object extracted in the video stream into the system in real time. , to use the stored information to match the same target object at different locations in the preset area at different times.
  • the preset parts may include the face, the front of the human body, the back of the human body (such as the back view) and the like.
  • the feature points of at least one preset part may include face feature points, human front feature points, and human back feature points.
  • the facial features, frontal features and back features of the target object in the image can be extracted first, and the facial features are associated with the frontal features and back features of the human body, After that, whether the system recognizes the front or back features of the human body, it can match with the associated facial features to determine the target object.
  • the running trajectory of the target object is determined.
  • the trajectory points of the target object and the timing information of the trajectory points can be determined; then, based on the trajectory points of the target object and the timing information of the trajectory points, the running trajectory of the target object can be determined .
  • the track point of the target object may include coordinate information of the current location of the target object.
  • the timing information of the track points may include the time nodes at which the track points are collected, and sequence information among multiple track points. Based on the coordinate information and timing information corresponding to the track point, the running track of the target object can be determined.
  • the system tracks the target object in real time and locates the target object to determine the track point (including coordinate information) of the target object in the preset area.
  • each track point is generated corresponding to a
  • the time node connects several track points of the target object in sequence according to the time node, and the running track of the target object can be generated.
  • S104 Determine passenger flow information in the target area based on the running trajectory of the target object and the passenger flow determination flag.
  • the passenger flow information may include but not limited to at least one of the following: valid passenger flow, the length of stay of valid passenger flow objects in the target area, the average length of stay of valid passenger flow objects in the target area, the number of passengers entering the target area, and Departure traffic in the target area, etc.
  • effective passenger flow objects are target objects with the behavior of entering and leaving the target area.
  • the valid passenger flow is the number of valid passenger flow objects.
  • the above S101-S104 can more accurately predict the target object’s relative Specific behaviors in the target area, such as predicting that the target object enters or leaves the target area, and then can obtain passenger flow information in the target area when the target device does not capture the target area.
  • FIG. 4 it is a flow chart for determining the passenger flow information in the target area, including steps S401-S403, specifically:
  • the running track of the target object is formed by connecting several track points of the target object according to the order of time nodes, the running track of the target object has a direction, that is, from the track point at the previous moment to the track point at the next moment.
  • FIG. 5 is a schematic diagram of the running track of the target object in the preset area.
  • the behavior of the target object can be predicted by judging whether the running track intersects with the passenger flow judgment mark.
  • the moving direction may include a first direction indicating that the target object enters the target area and/or a second direction indicating that the target object leaves the target area.
  • first direction indicating that the target object enters the target area
  • second direction indicating that the target object leaves the target area.
  • the display effect of the first direction and the second direction can be shown in FIG. 3a or 3b. Repeated parts will not be repeated.
  • S403 Determine passenger flow information in the target area based on the moving direction.
  • the dotted circle box indicates that the running track of the target object intersects with the passenger flow judgment mark, wherein, the running track 1 intersects with the passenger flow judgment mark, and it can be determined that the moving direction of the running track relative to the passenger flow judgment mark is the first One direction, and then it can be predicted that the target object has the behavior of entering the target area; the running track 2 intersects with the passenger flow judgment mark, and the moving direction of the running track relative to the passenger flow judgment mark can be determined as the second direction, and then it can be predicted that the target object has the behavior of leaving the target area the behavior of.
  • the passenger flow counting task when the passenger flow counting task is known as counting the passenger flow entering the target area, it is first judged whether the running track of the target object intersects with the passenger flow judgment mark, and if the intersection is confirmed, the running track is further judged The moving direction of the sign is judged relative to the passenger flow. Specifically, firstly, the first direction indicating that the target object enters the target area can be screened out of the moving direction of the target object's running track relative to the passenger flow determination sign, and it can be predicted that the target object has the behavior of entering the target area. At this time, the The corresponding target object is used as the incoming passenger flow object in the target area; after that, the number of incoming passenger flow objects is used as the incoming passenger flow in the target area. Exemplarily, the passenger flow entering the target area can be counted within one cycle.
  • the passenger flow counting task when the passenger flow counting task is known as counting the passenger flow leaving the target area, it is first judged whether the running trajectory of the target object intersects with the passenger flow judgment mark, and if the intersection is determined, further judgment is made on the running track.
  • the trajectory is relative to the moving direction of the passenger flow judgment sign.
  • the second direction indicating that the target object leaves the target area can be filtered out from the running track of the target object and the moving direction of the passenger flow determination sign, and it can be predicted that the target object has a behavior of leaving the target area.
  • the The corresponding target object is used as the departing passenger flow object in the target area; after that, the number of departing passenger flow objects is taken as the departing passenger flow in the target area.
  • the passenger flow leaving the target area can be counted within a period.
  • the footfall information includes effective footfall within the target area.
  • the known passenger flow statistics task is to count the effective passenger flow in the target area, first judge whether the running trajectory of the target object intersects with the passenger flow judgment mark, and if the intersection is confirmed, further judge the running trajectory relative to the passenger flow judgment mark direction of movement.
  • the moving direction includes a first direction indicating that the target object enters the target area and a second direction indicating that the target object leaves the target area
  • the corresponding target object is used as an effective passenger flow object in the target area, that is, it may include entering passenger flow Objects and departing passenger flow objects; the number of valid passenger flow objects is taken as the effective passenger flow in the target area.
  • the effective customer flow of brand stores can be counted.
  • Known brand stores cannot install cameras. If the store managers want to count the effective customer flow of this brand store, they can install the camera in the channel at the door of the store.
  • the customer information of entering and leaving the brand store is obtained to assist the store operator to evaluate the passenger flow of the brand store.
  • the specific assessment content such as judging whether the products introduced by the store meet customer needs, the age group information of visiting customers, And gender information of visiting customers, etc.
  • a passenger flow statistics report may also be generated and uploaded based on the passenger flow information.
  • the passenger flow statistics report may include the passenger flow information.
  • the customer flow statistics report can be used to provide business reference data for brand stores. For example, the number of customers entering the store is large but the sales are low. The goods do not meet the needs of customers. At this time, the management personnel can make adjustments to the goods in time according to the business reference data.
  • the passenger flow information also includes the length of stay of valid passenger flow objects in the target area.
  • the known passenger flow counting task as counting the length of stay of valid passenger flow objects in the target area, first determine the intersection time of the running trajectory of the effective passenger flow object and the passenger flow judgment mark (i.e. the following first time and second time ), and further determine the length of stay based on the time difference.
  • the running track of the effective passenger flow object and the passenger flow judgment mark determine the first time when the running track of the effective passenger flow object intersects with the passenger flow judgment mark in the first direction; based on the running track of the effective passenger flow object and the passenger flow judgment mark, determine The second time at which the trajectory of the effective passenger flow object intersects the passenger flow determination mark in the second direction; based on the first time and the second time, determine the length of stay of the effective passenger flow object in the target area.
  • the running trajectory of an effective passenger flow object is composed of several trajectory points, first determine the trajectory point A that intersects with the passenger flow judgment mark in the first direction, and determine the intersection of the trajectory point A and the passenger flow judgment mark.
  • the length of stay of the effective passenger flow object in the target area can be determined by the time difference between the first time and the second time , and then manage the target area in a targeted manner.
  • the passenger flow information in the passenger flow statistics report includes the length of stay of each customer in the store. If the stay time is short, it means that the goods in the store do not meet the needs of the customer. At this time, the store Products that meet the requirements of customers with shorter stays can be designed in a targeted manner.
  • the passenger flow information also includes the average length of stay of valid passenger flow objects in the target area.
  • the task of passenger flow statistics is to count the average length of stay of valid passenger flow objects in the target area.
  • first determine the intersection time of the intersection of the running track of the effective passenger flow object and the passenger flow judgment mark, that is, the first time and the second time Two time further determine the length of stay according to the time difference, after that, based on the length of stay and the number of effective passenger flow objects in the target area of at least some (for example) effective passenger flow objects, determine the average length of stay of the effective passenger flow objects in the target area
  • T i represents the length of stay of the i-th valid passenger flow object, i is less than N, and both i and N are positive integers.
  • N represents the number of valid passenger flow objects counted within the cycle range, and N takes a positive integer.
  • the cycle can be set according to the needs of the statistical task, such as one day or one week, and is not specifically limited here.
  • the target area can be managed in a targeted manner by determining the average length of stay of the effective passenger flow objects in the target area.
  • the customer flow information in the customer flow statistics report includes the average stay time of customers visiting the store within a day. If the average stay time is longer, it means that the goods in the store are more in line with customer needs. The focus of the products is as a promotional product.
  • the above passenger flow statistics report may also include but not limited to at least one of the following: report identification information, video stream display device ID, target device ID, and target area ID .
  • the report identification information may include information such as the name and number of the passenger flow statistics report.
  • the video stream display device is, for example, a light-emitting diode (Light-Emitting Diode, LED) display screen, a display screen of a computer, or a screen of a mobile device such as a mobile phone.
  • the user before starting the passenger flow counting task, the user needs to configure parameters first, for example, create a new page on the local page of the edge hardware (such as a computer, mobile phone and other mobile devices) to store the video captured by the camera and the passenger flow counting task Files of relevant information, configure the camera name, video stream address, add video tasks, etc. Afterwards, the user uploads the CAD drawing of the site to be monitored to the system, draws the target area on the CAD drawing, and configures the name of the target area, such as an ID.
  • the above information can be correspondingly generated in the passenger flow statistics report.
  • the system can track the running trajectory of the target object in real time, and output the completed trajectory message, which includes the entry target generated by each target object at different time points obtained by shooting the preset area by different target devices The behavior of the area and the message corresponding to the behavior of leaving the target area, and restore the track at the business layer.
  • the restoration process is to integrate the track points generated by each target object outside the target area to determine the running track of the target object. Afterwards, it is judged whether the running track intersects with the passenger flow judgment mark, and the moving direction of the running track relative to the passenger flow judgment mark, so as to complete the judgment of the behavior of entering or leaving the target area.
  • a target object ID person id
  • the target object is counted as a valid passenger flow object of the target area.
  • the message of the behavior of entering the target area or the behavior of leaving the target area includes: message generation timestamp (timestamp), edge hardware ID (deviceID), camera ID (cameraID), behavior of entering the target area (in), leaving Behavior of the target area (out), target area ID (areaID), etc.
  • the writing order of each step does not mean a strict execution order and constitutes any limitation on the implementation process.
  • the specific execution order of each step should be based on its function and possible
  • the inner logic is OK.
  • the embodiment of the present disclosure also provides a device for obtaining passenger flow information corresponding to the method for obtaining passenger flow information. Since the problem-solving principle of the device in the embodiment of the present disclosure is the same as the above-mentioned method for obtaining passenger flow information in the embodiment of the present disclosure Similar, therefore, the implementation of the device can refer to the implementation of the method, and repeated descriptions will not be repeated.
  • FIG. 6 it is a schematic diagram of a device for obtaining passenger flow information provided by an embodiment of the present disclosure.
  • the device includes: a video stream obtaining module 601 , a sign determining module 602 , a running track determining module 603 and a passenger flow information determining module 604 ;in,
  • a video stream acquiring module 601 configured to acquire a video stream obtained by at least one target device shooting a preset area; the preset area is adjacent to the entrance and exit of the target area, and is not within the target area;
  • a sign determination module 602 configured to determine a passenger flow determination sign corresponding to the target area
  • a running track determining module 603, configured to identify the target object in the video stream, and determine the running track of the target object
  • a passenger flow information determining module 604 configured to determine passenger flow information in the target area based on the running trajectory of the target object and the passenger flow determination flag.
  • the passenger flow information determining module 604 is configured to determine the movement of the running trajectory relative to the passenger flow judging sign when it is determined that the running trajectory intersects with the passenger flow judging sign direction: determining passenger flow information in the target area based on the moving direction.
  • the passenger flow information includes effective passenger flow in the target area
  • the passenger flow information determining module 604 is configured to: if the moving direction includes a first direction indicating that the target object enters the target area and a second direction indicating that the target object leaves the target area, The corresponding target object is used as an effective passenger flow object in the target area; the number of the effective passenger flow objects is used as the effective passenger flow in the target area.
  • the passenger flow information further includes the length of stay of the valid passenger flow object in the target area
  • the passenger flow information determining module 604 is further configured to respectively determine that the running trajectory of the valid passenger flow object in the first direction is consistent with the passenger flow determination mark based on the running trajectory of the effective passenger flow object and the passenger flow determination mark. Intersect the first time, and determine the second time when the running track of the effective passenger flow object intersects the passenger flow judgment mark in the second direction; based on the first time and the second time, determine the The length of stay of the effective passenger flow objects in the target area.
  • the passenger flow information further includes the average length of stay of the valid passenger flow objects in the target area
  • the passenger flow information determination module 604 is further configured to determine the average value of the effective passenger flow objects in the target area based on the length of stay of at least some of the effective passenger flow objects in the target area and the number of the effective passenger flow objects. Length of stay.
  • the passenger flow information further includes at least one of the passenger flow entering and leaving the target area
  • the passenger flow information determining module 604 is further configured to use the corresponding target object as the passenger flow entering the target area when the moving direction includes a first direction indicating that the target object enters the target area object; use the quantity of the incoming passenger flow objects as the incoming passenger flow in the target area; if the moving direction includes a second direction indicating that the target object leaves the target area, use the corresponding The target object is used as the departing passenger flow object in the target area; the number of the departing passenger flow objects is used as the departing passenger flow in the target area.
  • the running trajectory determination module 603 is configured to identify the target object in the video stream, determine the trajectory point of the target object and the timing information of the trajectory point; based on the target The track points of the object and the timing information of the track points determine the running track of the target object.
  • the sign determination module 602 is configured to determine relative position information of the preset area and the entrance and exit of the target area; based on the relative position information and the position of the preset area information to determine the passenger flow determination flag corresponding to the target area.
  • the device for obtaining passenger flow information further includes a report generating module 605, configured to determine the passenger flow information in the target area based on the running trajectory of the target object and the passenger flow determination flag Afterwards, based on the passenger flow information, a passenger flow statistics report is generated and uploaded;
  • the passenger flow statistics report includes the passenger flow information, or includes the passenger flow information and at least one of the following: report identification information, video stream display device ID, target device ID, and target area ID.
  • FIG. 7 is a schematic structural diagram of a computer device provided by an embodiment of the present disclosure, including:
  • Processor 71 and memory 72 stores machine-readable instructions executable by the processor 71, the processor 71 is used to execute the machine-readable instructions stored in the memory 72, and the machine-readable instructions are executed by the processor 71 During execution, the processor 71 performs the following steps:
  • memory 72 comprises memory 721 and external memory 722;
  • Memory 721 here is also called internal memory, is used for temporarily storing computing data in processor 71, and the data exchanged with external memory 722 such as hard disk, processor 71 communicates with memory 721 through memory 721.
  • the external memory 722 performs data exchange.
  • An embodiment of the present disclosure also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is run by a processor, the steps of the method for obtaining passenger flow information described in the above-mentioned method embodiments are executed.
  • the storage medium may be a volatile or non-volatile computer-readable storage medium.
  • the computer program product of the method for obtaining passenger flow information provided by the embodiments of the present disclosure includes computer-readable codes, or a computer-readable storage medium storing program codes, when the computer-readable codes are run in a processor of an electronic device
  • the instructions included in the program code can be used to execute the steps of the method for obtaining passenger flow information described in the above method embodiments.
  • the computer program product can be specifically realized by means of hardware, software or a combination thereof.
  • the computer program product is embodied as a computer storage medium, and in another optional embodiment, the computer program product is embodied as a software product, such as a software development kit (Software Development Kit, SDK) etc. wait.
  • a software development kit Software Development Kit, SDK
  • the disclosed devices and methods may be implemented in other ways.
  • the device embodiments described above are only illustrative.
  • the division of the units is only a logical function division.
  • multiple units or components can be combined or Some features may be ignored, or not implemented.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be through some communication interfaces, and the indirect coupling or communication connection of devices or units may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.
  • each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit.
  • the functions are realized in the form of software function units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium executable by a processor.
  • the technical solution of the present disclosure is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including Several instructions are used to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in various embodiments of the present disclosure.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disc and other media that can store program codes. .

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Image Analysis (AREA)

Abstract

一种获取客流信息的方法、装置、计算机设备以及存储介质，其中，该方法包括：获取至少一个目标设备对预设区域进行拍摄得到的视频流；所述预设区域与目标区域的出入口相邻，并且不在所述目标区域内（S101）；确定与所述目标区域对应的客流判定标志（S102）；识别所述视频流中的目标对象，并确定所述目标对象的运行轨迹（S103）；基于所述目标对象的运行轨迹和所述客流判定标志，确定所述目标区域内的客流信息（S104）。

Description

获取客流信息的方法、装置、计算机设备以及存储介质
本申请要求在2021年8月30日提交中国专利局、申请号为202111007027.X、申请名称为“获取客流信息的方法、装置、计算机设备以及存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本公开涉及计算机视觉技术领域，具体而言，涉及一种获取客流信息的方法、装置、计算机设备以及存储介质。
背景技术
客流统计可以给出某一特定的场所在一定时间内的客流量等信息。现今，客流的识别和统计是计算机视觉领域的重要研究方向，但传统方案中基于计算机视觉的客流统计多为正向客流统计，即需要摄像头能够直接拍摄到需要进行客流统计的场所，难以满足一些不能在内部安装摄像头区域的需求，比如，高端商场无法将摄像头装入品牌店铺内。
因此，针对不能在内部安装摄像头的场所，难以进行客流信息的识别和统计。
发明内容
本公开实施例至少提供一种获取客流信息的方法、装置、计算机设备以及存储介质。
第一方面，本公开实施例提供了一种获取客流信息的方法，包括：
获取至少一个目标设备对预设区域进行拍摄得到的视频流；所述预设区域与目标区域的出入口相邻，并且不在所述目标区域内；
确定与所述目标区域对应的客流判定标志；
识别所述视频流中的目标对象，并确定所述目标对象的运行轨迹；
基于所述目标对象的运行轨迹和所述客流判定标志，确定所述目标区域内的客流信息。
该方面，通过与目标区域的出入口相邻，且不在目标区域内的预设区域中产生的目标对象的运行轨迹、以及与目标区域对应的客流判定标志，能够较为准确地预测出目标对象相对于目标区域的具体的行为，比如预测目标对象进入或离开目标区域，进而能够在目标设备未对目标区域进行拍摄的情况下，获得目标区域内的客流信息。
一种可选的实施方式中，所述基于所述目标对象的运行轨迹和所述客流判定标志，确定所述目标区域内的客流信息，包括：
在确定所述运行轨迹与所述客流判定标志相交的情况下，确定所述运行轨迹相对于所述客流判定标志的移动方向；
基于所述移动方向，确定所述目标区域内的客流信息。
该实施方式，在运行轨迹与客流判定标志相交的情况下，说明目标对象有进入或离开目标区域的行为，之后，通过目标对象的运行轨迹相对于客流判定标志的移动方向，能够进一步的预测出目标对象的具体行为，比如进入目标区域或离开目标区域，进而能够较为准确地获取到目标设备无法拍摄到的目标区域内的客流信息，实现对该目标区域的客流统计。
一种可选的实施方式中，所述客流信息包括所述目标区域内的有效客流量；
所述基于所述移动方向，确定所述目标区域内的客流信息，包括：
在所述移动方向包括指示所述目标对象进入所述目标区域的第一方向和指示所述 目标对象离开所述目标区域的第二方向的情况下，将对应的所述目标对象作为所述目标区域的有效客流对象；
将所述有效客流对象的数量，作为所述目标区域内的有效客流量。
该实施方式，如果移动方向包括第一方向，则确定目标对象进入目标区域，如果移动方向包括第二方向，则确定该目标对象离开目标区域，在某一个目标对象既存在进入目标区域的行为，也存在离开目标区域的行为时，则该目标对象为目标区域的一个有效客流对象，因此，上述实施方式通过判断移动方向是否包括第一方向和第二方向，能够精确地统计出目标区域的有效客流量，进而具有针对性的管理目标区域，示例性的，在高端商场应用场景，可以统计品牌店铺的有效客流量，辅助商场运营人员对该店铺客流的评估；或者，为该店铺提供业务参考数据，比如进店顾客人数多但销售额低，则一种可能的原因是货品不符合顾客需求，此时，管理人员可以根据该业务参考数据，及时对货品做出调整。
一种可选的实施方式中，所述客流信息还包括所述有效客流对象在所述目标区域内的停留时长；
所述基于所述移动方向，确定所述目标区域内的客流信息，还包括：
基于所述有效客流对象的运行轨迹和所述客流判定标志，分别确定所述有效客流对象的运行轨迹在所述第一方向上与所述客流判定标志相交的第一时间，以及确定所述有效客流对象的运行轨迹在所述第二方向上与所述客流判定标志相交的第二时间；
基于所述第一时间和所述第二时间，确定所述有效客流对象在所述目标区域内的停留时长。
该实施方式，由于第一方向为进入目标区域的方向，第二方向为离开目标区域的方向，因此，通过第一时间与第二时间的时间差，就能够确定出有效客流对象在目标区域内的停留时长，进而能够实现有针对性的管理目标区域。示例性的，在高端商场应用场景，可以统计至少部分(例如每个)顾客在店铺内的停留时长，如果停留时长较短，则说明该店铺货品不满足该顾客需求，此时，该店铺可以具有针对性的为停留时长较短的顾客设计满足其要求的货品。
一种可选的实施方式中，所述客流信息还包括所述有效客流对象在所述目标区域内的平均停留时长；
所述基于所述移动方向，确定所述目标区域内的客流信息，还包括：
基于至少部分所述有效客流对象在所述目标区域的停留时长和所述有效客流对象的数量，确定所述有效客流对象在所述目标区域内的平均停留时长。
该实施方式，利用至少部分(例如每个)目标对象在目标区域内的停留时长，能够准确地确定有效客流对象在目标区域内的平均停留时长。进而，通过确定出的有效客流对象在目标区域内的平均停留时长，能够有针对性的管理目标区域。示例性的，在高端商场应用场景，可以统计一天内光顾店铺的顾客的平均停留时长，如果平均停留时长较长，则说明该店铺货品比较符合顾客需求，可以将该时期的货品重点作为推广货品。
一种可选的实施方式中，所述客流信息还包括所述目标区域的进入客流量和离开客流量中的至少一项；
所述基于所述移动方向，确定所述目标区域内的客流信息，包括如下至少一项：
在所述移动方向包括指示所述目标对象进入所述目标区域的第一方向的情况下，将 对应的所述目标对象作为所述目标区域的进入客流对象；将所述进入客流对象的数量，作为所述目标区域内的进入客流量；
在所述移动方向包括指示所述目标对象离开所述目标区域的第二方向的情况下，将对应的所述目标对象作为所述目标区域的离开客流对象；将所述离开客流对象的数量，作为所述目标区域内的离开客流量。
该实施方式，通过判断移动方向是否包括第一方向，能够精确地统计出进入目标区域的进入客流对象的数量，通过判断移动方向是否包括第二方向，能够精确地统计出离开目标区域的离开客流对象的数量。
一种可选的实施方式中，所述识别所述视频流中的目标对象，并确定所述目标对象的运行轨迹，包括：
识别所述视频流中的目标对象，确定所述目标对象的轨迹点以及所述轨迹点的时序信息；
基于所述目标对象的轨迹点以及所述轨迹点的时序信息，确定所述目标对象的运行轨迹。
该实施方式，可以通过轨迹点与每一轨迹点的时序信息，能够准确地还原目标对象的运行轨迹，为预测目标对象的行为做准备。
一种可选的实施方式中，所述确定与所述目标区域对应的客流判定标志，包括：
确定所述预设区域与所述目标区域的出入口的相对位置信息；
基于所述相对位置信息和所述预设区域的位置信息，确定与所述目标区域对应的所述客流判定标志。
该实施方式，通过上述相对位置信息和预设区域的位置信息，能够确定出客流判定标志的准确位置，利用该位置的客流判定标志预测目标对象的行为，能够提高预测目标对象的行为的准确率。
一种可选的实施方式中，在基于所述目标对象的运行轨迹和所述客流判定标志，确定所述目标区域内的客流信息之后，还包括：
基于所述客流信息，生成并上传客流统计报告；
所述客流统计报告包括所述客流信息，或者，所述客流信息以及以下至少一项：报告标识信息、视频流展示设备身份标识ID、目标设备身份标识ID、目标区域身份标识ID。
该实施方式，利用客流统计报告能够较为准确和全面的反应目标区域内的客流的相关信息，进而，利用该客流统计报告，能够实现有针对性地对目标区域进行管理。
第二方面，本公开实施例还提供一种获取客流信息的装置，包括：
视频流获取模块，用于获取至少一个目标设备对预设区域进行拍摄得到的视频流；所述预设区域与目标区域的出入口相邻，并且不在所述目标区域内；
标志确定模块，用于确定与所述目标区域对应的客流判定标志；
运行轨迹确定模块，用于识别所述视频流中的目标对象，并确定所述目标对象的运行轨迹；
客流信息确定模块，用于基于所述目标对象的运行轨迹和所述客流判定标志，确定所述目标区域内的客流信息。
一种可选的实施方式中，所述客流信息确定模块，用于在确定所述运行轨迹与所述 客流判定标志相交的情况下，确定所述运行轨迹相对于所述客流判定标志的移动方向；基于所述移动方向，确定所述目标区域内的客流信息。
一种可选的实施方式中，所述客流信息包括所述目标区域内的有效客流量；
所述客流信息确定模块，用于在所述移动方向包括指示所述目标对象进入所述目标区域的第一方向和指示所述目标对象离开所述目标区域的第二方向的情况下，将对应的所述目标对象作为所述目标区域的有效客流对象；将所述有效客流对象的数量，作为所述目标区域内的有效客流量。
一种可选的实施方式中，所述客流信息还包括所述有效客流对象在所述目标区域内的停留时长；
所述客流信息确定模块，还用于基于所述有效客流对象的运行轨迹和所述客流判定标志，分别确定所述有效客流对象的运行轨迹在所述第一方向上与所述客流判定标志相交的第一时间，以及确定所述有效客流对象的运行轨迹在所述第二方向上与所述客流判定标志相交的第二时间；基于所述第一时间和所述第二时间，确定所述有效客流对象在所述目标区域内的停留时长。
一种可选的实施方式中，所述客流信息还包括所述有效客流对象在所述目标区域内的平均停留时长；
所述客流信息确定模块，还用于基于至少部分所述有效客流对象在所述目标区域的停留时长和所述有效客流对象的数量，确定所述有效客流对象在所述目标区域内的平均停留时长。
一种可选的实施方式中，所述客流信息还包括所述目标区域的进入客流量和离开客流量中的至少一项；
所述客流信息确定模块，还用于在所述移动方向包括指示所述目标对象进入所述目标区域的第一方向的情况下，将对应的所述目标对象作为所述目标区域的进入客流对象；将所述进入客流对象的数量，作为所述目标区域内的进入客流量；在所述移动方向包括指示所述目标对象离开所述目标区域的第二方向的情况下，将对应的所述目标对象作为所述目标区域的离开客流对象；将所述离开客流对象的数量，作为所述目标区域内的离开客流量。
一种可选的实施方式中，所述运行轨迹确定模块，用于识别所述视频流中的目标对象，确定所述目标对象的轨迹点以及所述轨迹点的时序信息；基于所述目标对象的轨迹点以及所述轨迹点的时序信息，确定所述目标对象的运行轨迹。
一种可选的实施方式中，所述标志确定模块，用于确定所述预设区域与所述目标区域的出入口的相对位置信息；基于所述相对位置信息和所述预设区域的位置信息，确定与所述目标区域对应的所述客流判定标志。
一种可选的实施方式中，所述获取客流信息的装置还包括报告生成模块，用于在基于所述目标对象的运行轨迹和所述客流判定标志，确定所述目标区域内的客流信息之后，基于所述客流信息，生成并上传客流统计报告；
所述客流统计报告包括所述客流信息，或者，包括所述客流信息以及以下至少一项：报告标识信息、视频流展示设备身份标识ID、目标设备身份标识ID、目标区域身份标识ID。
第三方面，本公开可选实现方式还提供一种计算机设备，处理器、存储器，所述存 储器存储有所述处理器可执行的机器可读指令，所述处理器用于执行所述存储器中存储的机器可读指令，所述机器可读指令被所述处理器执行时，所述机器可读指令被所述处理器执行时执行上述第一方面，或第一方面中任一种可能的实施方式中的步骤。
第四方面，本公开可选实现方式还提供一种计算机可读存储介质，该计算机可读存储介质上存储有计算机程序，该计算机程序被运行时执行上述第一方面，或第一方面中任一种可能的实施方式中的步骤。
第五方面，本公开可选实现方式还提供一种计算机程序产品，包括计算机可读代码，或者承载有计算机可读代码的计算机可读存储介质，当所述计算机可读代码在电子设备的处理器中运行时，所述电子设备中的处理器执行上述第一方面，或第一方面中任一种可能的实施方式中的步骤。
关于上述获取客流信息的装置、计算机设备、及计算机可读存储介质的效果描述参见上述获取客流信息的方法的说明，这里不再赘述。
为使本公开的上述目的、特征和优点能更明显易懂，下文特举较佳实施例，并配合所附附图，作详细说明如下。
附图说明
为了更清楚地说明本公开实施例的技术方案，下面将对实施例中所需要使用的附图作简单地介绍，此处的附图被并入说明书中并构成本说明书中的一部分，这些附图示出了符合本公开的实施例，并与说明书一起用于说明本公开的技术方案。应当理解，以下附图仅示出了本公开的某些实施例，因此不应被看作是对范围的限定，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他相关的附图。
图1示出了本公开实施例所提供的一种获取客流信息的方法的流程图；
图2示出了本公开实施例所提供的一种预设区域与目标区域的布局示意图；
图3a示出了本公开实施例所提供的预设区域与目标区域出入口规则相邻场景下的展示示意图；
图3b示出了本公开实施例所提供的预设区域与目标区域出入口不规则相邻场景下的展示示意图；
图4示出了本公开实施例所提供的确定目标区域内的客流信息的流程图；
图5示出了本公开实施例所提供的目标对象在预设区域内的运行轨迹的示意图；
图6示出了本公开实施例所提供的一种获取客流信息的装置的示意图；
图7示出了本公开实施例所提供的一种计算机设备结构示意图。
具体实施方式
为使本公开实施例的目的、技术方案和优点更加清楚，下面将结合本公开实施例中附图，对本公开实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本公开一部分实施例，而不是全部的实施例。通常在此处描述和示出的本公开实施例的组件可以以各种不同的配置来布置和设计。因此，以下对本公开的实施例的详细描述并非旨在限制要求保护的本公开的范围，而是仅仅表示本公开的选定实施例。基于本公开的实施例，本领域技术人员在没有做出创造性劳动的前提下所获得的所有其他实施例，都属于本公开保护的范围。
另外，本公开实施例中的说明书和权利要求书及上述附图中的术语“第一”、“第二” 等是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换，以便这里描述的实施例能够以除了在这里图示或描述的内容以外的顺序实施。
在本文中提及的“多个或者若干个”是指两个或两个以上。“和/或”，描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况。字符“/”一般表示前后关联对象是一种“或”的关系。
经研究发现，客流统计是一种基于计算机视觉技术，来统计相关任务数据的方法。现今，客流的识别和统计是计算机视觉领域的重要研究方向，但传统方案中基于计算机视觉的客流统计多为正向客流统计，即需要摄像头能够照射到需要进行客流统计的场所，难以满足一些不能在内部安装摄像头区域的需求，比如，高端商场无法将摄像头装入品牌店铺内。
基于上述研究，本公开提供了一种获取客流信息的方法，通过与目标区域的出入口相邻，且不在目标区域内的预设区域中产生的目标对象的运行轨迹、以及与目标区域对应的客流判定标志，能够较为准确地预测出目标对象相对于目标区域的具体的行为，比如预测目标对象进入或离开目标区域，进而能够在目标设备未对目标区域进行拍摄的情况下，获得目标区域内的客流信息。
应注意到：相似的标号和字母在下面的附图中表示类似项，因此，一旦某一项在一个附图中被定义，则在随后的附图中不需要对其进行进一步定义和解释。
为便于对本实施例进行理解，首先对本公开实施例所公开的一种获取客流信息的方法进行详细介绍，本公开实施例所提供的获取客流信息的方法的执行主体一般为具有一定计算能力的计算机设备，在一些可能的实现方式中，该获取客流信息的方法可以通过处理器调用存储器中存储的计算机可读指令的方式来实现。
下面以执行主体为计算机设备为例对本公开实施例提供的获取客流信息的方法加以说明。
参见图1所示，其为本公开实施例提供的一种获取客流信息的方法的流程图，所述方法包括步骤S101～S104，其中：
S101：获取至少一个目标设备对预设区域进行拍摄得到的视频流；预设区域与目标区域的出入口相邻，并且不在目标区域内。
本步骤中，目标设备可以为摄像头等用于图像/视频采集的拍摄设备。目标设备的拍摄范围可以为预设区域，或预设区域内的部分区域。预设区域可以为至少一个目标设备拍摄的区域；或者，可以为预先设置的至少一个目标设备拍摄的区域中的感兴趣区域(Region Of Interest，ROI)。其中，预设区域都与目标区域的出入口相邻。目标区域可以为需要统计客流的区域，具体地，用户可以将需要统计客流的区域的计算机辅助设计(Computer Aided Design，CAD)图上传至***，并在CAD图上画出目标区域，这里，目标区域为目标设备不能拍摄到的区域或是难以拍摄完整的区域，比如品牌店铺等一些不能安装摄像头的区域。
示例性的，可以将目标区域与目标设备关联。一个目标区域外可以对应安装有一个或是多个目标设备，安装位置可以根据实际应用场景进行设定，本公开实施例不具体限定。
这里，预设区域的位置的选择可以根据目标区域的出入口进行设置，示例性的，预设区域所在区域能够覆盖所有进入或离开目标区域的目标对象。设置预设区域与目标区域的出入口相邻，是为了利用目标设备拍摄到在预设区域内目标对象即将进入或已经离开目标区域的运行轨迹。可以参见图2所示，其为一种预设区域与目标区域的布局示意图。其中，虚线条部分表示目标区域的出入口。
示例性的，在预设区域面积较大的情况下，可以设置多个目标设备拍摄预设区域，以获取完整的对预设区域进行拍摄得到的视频流。
需要说明的是，图2中以目标区域包括两个出入口为例进行说明，在实际应用中目标区域所包括的出入口数量可以为一个或是多个，在此不予限定。在一种可能的实现方式中，单个出入口可允许人员进出，或是只允许人员进入或离开，具体可结合目标区域包括的出入口数量以及预先设置的规则来确定，在此不予限定。对于出入口具备单一职能的情况下，即出入口仅支持进入或离开，那么多个出入口中必然存在两个出入口具备不同的职能，即分别供人员进入和离开，那么可将两个出入口分别对应的预设区域的数据进行汇总处理，比如，结合两个出入口对应预设区域分别得到的进入客流量和离开客流量来确定有效客流量等，在此不予限定。
具体实施时，在确定要获取目标区域的客流信息的情况下，先查询与目标区域相关联的至少一个目标设备，并确定目标设备的标识信息，比如摄像头的身份标识号(Identity Document，ID)，之后，利用至少一个目标设备，可以直接获取到对预设区域进行拍摄得到的视频流，这里，目标设备拍摄的视频流可以包括但不仅限于通过实时流传输协议(Real Time Streaming Protocol，RTSP)传送的视频流。之后，将RTSP传送的视频流输入到深度神经网络模型进行解码处理，以获取解码后的每一帧的图像，每一帧图像中记录有该预设区域出现的目标对象。
在一些实施例中，通过设置预设区域为至少一个目标设备拍摄的区域中的ROI，能够减少对不相关区域内目标对象的检测，有利于提高检测效率，节约计算资源。
S102：确定与目标区域对应的客流判定标志。
本步骤中，客流判定标志可包括但不仅限于标线或区域。其中，标线或区域可以包括多种形态，具体的，标线可以包括规则标线(比如直线等)，或者，不规则标线(比如曲线等)；区域可以包括规则区域(比如矩形区域、圆形区域和椭圆形区域等)，或者，不规则区域(比如不规则多边形等)。
客流判定标志可以是基于目标区域的位置和预设区域的位置，在预设区域对应的图像上绘制的标线和/或区域(可手动绘制，自动绘制，或是基于自动绘制的内容进行手动/自动校正而得到)，客流判定标志的位置可以在预设区域与目标区域出入口相邻的位置，具体地，可以参见图3a所示，其为预设区域与目标区域出入口规则相邻场景下的展示示意图，其中，客流判定标志为直线。或者，可以参见图3b所示，其为预设区域与目标区域出入口不规则相邻场景下的展示示意图，其中，客流判定标志为曲线。本公开实施例中客流判定标志的绘制情况不仅限于上述两种应用场景，客流判定标志的形状可以任意绘制，本公开实施例不进行限定。
在绘制完客流判定标志之后，可以根据预设区域与目标区域的相对位置信息，设置客流判定标志的判定方向。具体地，客流判定标志的判定方向可以包括从预设区域指向目标区域的第一方向和从目标区域指向预设区域的第二方向。这里，第一方向和第二方 向的展示示意图可以参见图3a或图3b所示，示例性的，第一方向31可以用于指示目标对象从预设区域进入目标区域的方向；第二方向32可以用于指示目标对象从目标区域离开的方向。
需要说明的是，在不同应用场景中客流判定标志的绘制和判定方向的设置可以根据实际应用场景进行设定，本公开实施例不进行具体限定。
在一些实施例中，***还可以自动绘制客流判定标志，具体地确定预设区域与目标区域的出入口的相对位置信息；基于相对位置信息和预设区域的位置信息，确定客流判定标志。这里，利用预设区域与目标区域的出入口的相对位置信息，比如，相对位置信息为预设区域与目标区域的出入口相邻，那么按照需求，在预设区域与目标区域相邻的出入口的位置处绘制客流判定标志(可以参见图3a或图3b所示)。
S103：识别视频流中的目标对象，并确定目标对象的运行轨迹。
本步骤中，目标对象的运行轨迹可以为目标对象的轨迹点按照每个轨迹点生成的时间节点的顺序连接而成。
具体实施时，可以按照下述步骤识别视频流中的目标对象：
S1031、对视频流解码后的图像进行特征点提取，得到预设区域图像的多个特征点。
S1032、将多个特征点与实时存储的目标对象包含的多个特征点分别进行比对和关联，确定视频流中的目标对象。
具体实施时，可以获取目标对象的至少一个预设部位的特征点；基于至少一个预设部位的特征点，确定视频流中的目标对象。需要说明的是，这些特征点可以在基于视频流得到后进行实时存储，这里实时存储过程可以为，将视频流中提取的每一个目标对象的至少一个预设部位的特征点实时存储到***中，以利用存储信息匹配不同时刻在预设区域不同位置的同一目标对象。这里，预设部位可以包括人脸、人体正面和人体背面(比如背影)等。至少一个预设部位的特征点，可以包括人脸特征点、人体正面特征点和人体背面特征点等。
示例性的，基于S101中解码后的每一帧图像，可以先提取图像中目标对象的人脸特征、人体正面特征和人体背面特征，并将人脸特征与人体正面特征和背面特征相关联，之后，***无论识别到的人体正面特征还是人体背面特征，都可以和其相关联的人脸特征进行匹配，确定目标对象。
之后，确定目标对象的运行轨迹，具体实施时，首先，可以确定目标对象的轨迹点以及轨迹点的时序信息；之后，基于目标对象的轨迹点以及轨迹点的时序信息，确定目标对象的运行轨迹。
这里，目标对象的轨迹点可以包括当前目标对象所在位置的坐标信息。轨迹点的时序信息可以包括采集到轨迹点的时间节点，以及多个轨迹点之间的先后顺序信息。基于轨迹点对应的坐标信息和时序信息，可以确定目标对象的运行轨迹。
延续上例，在确定目标对象之后，***实时跟踪该目标对象，定位目标对象，以确定目标对象在预设区域内的轨迹点(包括坐标信息)，这里，每一轨迹点的生成都对应一个时间节点，将该目标对象的若干个轨迹点按照时间节点顺序连接，可以生成目标对象的运行轨迹。
S104：基于目标对象的运行轨迹和客流判定标志，确定目标区域内的客流信息。
本步骤中，客流信息可以包括但不限于以下至少一种：有效客流量、有效客流对象 在目标区域内的停留时长、有效客流对象在目标区域内的平均停留时长、目标区域的进入客流量和目标区域的离开客流量等。这里，有效客流对象为具有进入和离开目标区域行为的目标对象。有效客流量为有效客流对象的数量。
上述S101～S104通过与目标区域的出入口相邻，且不在目标区域内的预设区域中产生的目标对象的运行轨迹、以及与目标区域对应的客流判定标志，能够较为准确地预测出目标对象相对于目标区域的具体的行为，比如预测目标对象进入或离开目标区域，进而能够在目标设备未对目标区域进行拍摄的情况下，获得目标区域内的客流信息。
参见图4所示，其为确定目标区域内的客流信息的流程图，包括步骤S401～S403，具体地：
S401：基于运行轨迹和客流判定标志，判断运行轨迹与客流判定标志是否相交。
由于目标对象的运行轨迹是目标对象的若干个轨迹点按照时间节点顺序连接而成的，因此目标对象的运行轨迹是有方向的，即从前一时刻的轨迹点指向后一时刻的轨迹点。可以参见图5所示，其为目标对象在预设区域内的运行轨迹的示意图。
在已知运行轨迹和客流判定标志的情况下，可以通过判断运行轨迹与客流判定标志是否相交来预测出目标对象的行为。
S402：在确定运行轨迹与客流判定标志相交的情况下，确定运行轨迹相对于客流判定标志的移动方向。
本步骤中，在确定运行轨迹与客流判定标志相交的情况下，能够预测出目标对象有进入或离开目标区域的行为。
移动方向可以包括指示目标对象进入目标区域的第一方向和/或指示目标对象离开目标区域的第二方向，这里，第一方向和第二方向的展示效果可以参见图3a或图3b所示，重复部分不再赘述。
S403：基于移动方向，确定目标区域内的客流信息。
示例性的，如图5所示，圆圈虚线框指示目标对象的运行轨迹与客流判定标志相交，其中，运行轨迹1与客流判定标志相交，可以确定运行轨迹相对于客流判定标志的移动方向为第一方向，进而可以预测目标对象具有进入目标区域的行为；运行轨迹2与客流判定标志相交，可以确定运行轨迹相对于客流判定标志的移动方向为第二方向，进而可以预测目标对象具有离开目标区域的行为。
上述S401～S403，在运行轨迹与客流判定标志相交的情况下，说明目标对象有进入或离开目标区域的行为，之后，通过目标对象的运行轨迹相对于客流判定标志的移动方向，能够进一步的预测出目标对象的具体行为，比如进入目标区域或离开目标区域，进而能够较为准确地获取到目标设备无法拍摄到的目标区域内的客流信息，实现对该目标区域的客流统计。
在一个实施例中，在已知客流统计任务为统计进入目标区域的客流量的情况下，先判断目标对象的运行轨迹是否与客流判定标志相交，并在确定相交的情况下，进一步判断运行轨迹相对于客流判定标志的移动方向。具体地，首先，可以筛选出目标对象的运行轨迹相对于客流判定标志的移动方向中指示目标对象进入目标区域的第一方向，能够预测出目标对象有进入目标区域的行为，此时，可以将对应的目标对象作为目标区域的进入客流对象；之后，将进入客流对象的数量，作为目标区域内的进入客流量。示例性的，可以在一个周期内统计出进入目标区域的客流量。
在另一个实施例中，在已知客流统计任务为统计离开目标区域的客流量的情况下，先判断目标对象的运行轨迹是否与客流判定标志相交，并在确定相交的情况下，进一步判断运行轨迹相对于客流判定标志的移动方向。具体地，首先，可以从目标对象的运行轨迹与客流判定标志的移动方向中筛选出指示目标对象离开目标区域的第二方向，能够预测出目标对象有离开目标区域的行为，此时，可以将对应的目标对象作为目标区域的离开客流对象；之后，将离开客流对象的数量，作为目标区域内的离开客流量。示例性的，可以在一个周期内统计出离开目标区域的客流量。
在一些实施例中，客流信息包括目标区域内的有效客流量。在已知客流统计任务为统计目标区域内的有效客流量的情况下，先判断目标对象的运行轨迹是否与客流判定标志相交，并在确定相交的情况下，进一步判断运行轨迹相对于客流判定标志的移动方向。具体地，在移动方向包括指示目标对象进入目标区域的第一方向和指示目标对象离开目标区域的第二方向的情况下，将对应的目标对象作为目标区域的有效客流对象，即可以包括进入客流对象和离开客流对象；将有效客流对象的数量，作为目标区域内的有效客流量。
示例性的，在高端商场应用场景，可以统计品牌店铺的有效客流量，已知品牌店铺不能安装摄像头，商场管理人员要想统计该品牌店铺的有效客流量，可将摄像头安装在店铺门口的通道中，获取进入和离开品牌店铺的顾客信息，以辅助商场运营人员对该品牌店铺的客流做出评估，具体评估内容，比如判断商场引进的商品是否符合顾客需求、到访顾客的年龄段信息、以及到访顾客的性别信息等。在已知顾客的运行轨迹相对于客流判定标志的移动方向既有进入品牌店铺的第一方向，又有离开品牌店铺的第二方向，则可以确定该顾客为该品牌店铺的有效客流对象，按照上述统计方法，能够在一定的周期内，统计出至少部分(例如所有)有效客流对象的数量，进而确定出该店铺的有效客流量。
在一些实施例中，在确定完目标区域内的客流信息之后，还可以基于客流信息，生成并上传客流统计报告，这里，客流统计报告可以包括客流信息。延续上例，在客流统计报告中的客流信息包括有效客流量时，可以利用该客流统计报告为品牌店铺提供业务参考数据，比如进店顾客人数多但销售额低，则一种可能的原因是货品不符合顾客需求，此时，管理人员可以根据该业务参考数据，及时对货品做出调整。
在一些实施例中，客流信息还包括有效客流对象在目标区域内的停留时长。在已知客流统计任务为统计有效客流对象在目标区域内的停留时长的情况下，先确定有效客流对象的运行轨迹与客流判定标志相交的相交时间(即下述的第一时间和第二时间)，进一步根据时间差确定停留时长。具体地，基于有效客流对象的运行轨迹和客流判定标志，确定有效客流对象的运行轨迹在第一方向上与客流判定标志相交的第一时间；基于有效客流对象的运行轨迹和客流判定标志，确定有效客流对象的运行轨迹在第二方向上与客流判定标志相交的第二时间；基于第一时间和第二时间，确定有效客流对象在目标区域内的停留时长。具体实施时，由于有效客流对象的运行轨迹是由若干个轨迹点组成的，首先确定在第一方向上与客流判定标志相交的轨迹点A，并确定该轨迹点A与客流判定标志相交的相交时间，即第一时间t 1；之后，确定该有效客流对象的运行轨迹在第二方向上与客流判定标志相交的轨迹点B，并确定该轨迹点B与客流判定标志相交的相交时间，即第二时间t 2；之后，基于第一时间和第二时间的，可以确定出有效客流对象在目 标区域的停留时长T，即T＝t 2-t 1。
上述，由于第一方向为进入目标区域的方向，第二方向为离开目标区域的方向，因此，通过第一时间与第二时间的时间差，就能够确定出有效客流对象在目标区域内的停留时长，进而有针对性的管理目标区域。示例性的，在高端商场应用场景，客流统计报告中的客流信息包括每个顾客在店铺内的停留时长，如果停留时长较短，则说明该店铺货品不满足该顾客需求，此时，该店铺可以具有针对性的为停留时长较短的顾客设计满足其要求的货品。
在一些实施例中，客流信息还包括有效客流对象在目标区域内的平均停留时长。在已知客流统计任务为统计有效客流对象在目标区域内的平均停留时长的情况下，具体地，先确定有效客流对象的运行轨迹与客流判定标志相交的相交时间，即为第一时间和第二时间，进一步根据时间差确定停留时长，之后，基于至少部分(例如各个)有效客流对象在目标区域的停留时长和有效客流对象的数量，确定有效客流对象在目标区域内的平均停留时长
Figure PCTCN2022092492-appb-000001
即
Figure PCTCN2022092492-appb-000002
其中，T i表示第i个有效客流对象的停留时长，i小于N，且i与N均为正整数。N表示周期范围内统计的有效客流对象的数量，N取正整数，其中，周期可以根据统计任务需要进行设定，比如一天或者一周等，在此不进行具体限定。
上述，通过确定出的有效客流对象在目标区域内的平均停留时长，能够有针对性的管理目标区域。示例性的，在高端商场应用场景，客流统计报告中的客流信息包括一天内光顾店铺的顾客的平均停留时长，如果平均停留时长较长，则说明该店铺货品比较符合顾客需求，可以将该时期的货品重点作为推广货品。
上述客流统计报告中除了包括为用户提供的客流信息之外，还可以包括但不仅限于以下至少一项：报告标识信息、视频流展示设备身份标识ID、目标设备身份标识ID、目标区域身份标识ID。这里，报告标识信息可以包括客流统计报告的名称、编号等信息。视频流展示设备比如发光二极管(Light-Emitting Diode，LED)显示屏、电脑的显示屏或手机等移动设备的屏幕等。
在一些实施例中，在启动客流统计任务之前，用户需要先配置参数，比如，在边缘硬件(比如电脑、手机等移动设备)的本地页面新建用于存储摄像头拍摄到的视频以及该客流统计任务的相关信息的文件，配置摄像头名称、视频流地址，添加视频任务等。之后，用户将需要进行监测的场地CAD图上传到***，并在CAD图上画出目标区域，并配置目标区域的名称，比如身份标识ID。上述信息可以对应生成在客流统计报告中。
在参数配置成功后，***可以实时追踪目标对象的运行轨迹，输出完成的轨迹报文，该轨迹报文包括不同目标设备对预设区域进行拍摄得到的各个目标对象在不同时间点产生的进入目标区域的行为和离开目标区域的行为对应的报文，并在业务层将轨迹还原，还原过程为对各个目标对象在目标区域外产生的轨迹点进行整合，确定目标对象的运行轨迹。之后，判断运行轨迹是否与客流判定标志相交，以及运行轨迹相对于客流判定标志的移动方向，完成对进入或离开目标区域的行为的判定。当一个目标对象ID(person id)同时有进入和离开目标区域的行为时，则该目标对象计为该目标区域的有效客流对象。
这里，进入目标区域的行为或离开目标区域的行为的报文包括：报文产生时间戳(timestamp)，边缘硬件ID(deviceID)，摄像头ID(cameraID)，进入目标区域的行为(in)，离开目标区域的行为(out)，目标区域ID(areaID)等。
本领域技术人员可以理解，在具体实施方式的上述方法中，各步骤的撰写顺序并不 意味着严格的执行顺序而对实施过程构成任何限定，各步骤的具体执行顺序应当以其功能和可能的内在逻辑确定。
基于同一发明构思，本公开实施例中还提供了与获取客流信息的方法对应的获取客流信息的装置，由于本公开实施例中的装置解决问题的原理与本公开实施例上述获取客流信息的方法相似，因此装置的实施可以参见方法的实施，重复之处不再赘述。
参见图6所示，为本公开实施例提供的一种获取客流信息的装置的示意图，所述装置包括：视频流获取模块601、标志确定模块602，运行轨迹确定模块603和客流信息确定模块604；其中，
视频流获取模块601，用于获取至少一个目标设备对预设区域进行拍摄得到的视频流；所述预设区域与目标区域的出入口相邻，并且不在所述目标区域内；
标志确定模块602，用于确定与所述目标区域对应的客流判定标志；
运行轨迹确定模块603，用于识别所述视频流中的目标对象，并确定所述目标对象的运行轨迹；
客流信息确定模块604，用于基于所述目标对象的运行轨迹和所述客流判定标志，确定所述目标区域内的客流信息。
一种可选的实施方式中，所述客流信息确定模块604，用于在确定所述运行轨迹与所述客流判定标志相交的情况下，确定所述运行轨迹相对于所述客流判定标志的移动方向；基于所述移动方向，确定所述目标区域内的客流信息。
一种可选的实施方式中，所述客流信息包括所述目标区域内的有效客流量；
所述客流信息确定模块604，用于在所述移动方向包括指示所述目标对象进入所述目标区域的第一方向和指示所述目标对象离开所述目标区域的第二方向的情况下，将对应的所述目标对象作为所述目标区域的有效客流对象；将所述有效客流对象的数量，作为所述目标区域内的有效客流量。
一种可选的实施方式中，所述客流信息还包括所述有效客流对象在所述目标区域内的停留时长；
所述客流信息确定模块604，还用于基于所述有效客流对象的运行轨迹和所述客流判定标志，分别确定所述有效客流对象的运行轨迹在所述第一方向上与所述客流判定标志相交的第一时间，以及确定所述有效客流对象的运行轨迹在所述第二方向上与所述客流判定标志相交的第二时间；基于所述第一时间和所述第二时间，确定所述有效客流对象在所述目标区域内的停留时长。
一种可选的实施方式中，所述客流信息还包括所述有效客流对象在所述目标区域内的平均停留时长；
所述客流信息确定模块604，还用于基于至少部分所述有效客流对象在所述目标区域的停留时长和所述有效客流对象的数量，确定所述有效客流对象在所述目标区域内的平均停留时长。
一种可选的实施方式中，所述客流信息还包括所述目标区域的进入客流量和离开客流量中的至少一项；
所述客流信息确定模块604，还用于在所述移动方向包括指示所述目标对象进入所述目标区域的第一方向的情况下，将对应的所述目标对象作为所述目标区域的进入客流对象；将所述进入客流对象的数量，作为所述目标区域内的进入客流量；在所述移动方 向包括指示所述目标对象离开所述目标区域的第二方向的情况下，将对应的所述目标对象作为所述目标区域的离开客流对象；将所述离开客流对象的数量，作为所述目标区域内的离开客流量。
一种可选的实施方式中，所述运行轨迹确定模块603，用于识别所述视频流中的目标对象，确定所述目标对象的轨迹点以及所述轨迹点的时序信息；基于所述目标对象的轨迹点以及所述轨迹点的时序信息，确定所述目标对象的运行轨迹。
一种可选的实施方式中，所述标志确定模块602，用于确定所述预设区域与所述目标区域的出入口的相对位置信息；基于所述相对位置信息和所述预设区域的位置信息，确定与所述目标区域对应的所述客流判定标志。
一种可选的实施方式中，所述获取客流信息的装置还包括报告生成模块605，用于在基于所述目标对象的运行轨迹和所述客流判定标志，确定所述目标区域内的客流信息之后，基于所述客流信息，生成并上传客流统计报告；
所述客流统计报告包括所述客流信息，或者，包括所述客流信息以及以下至少一项：报告标识信息、视频流展示设备身份标识ID、目标设备身份标识ID、目标区域身份标识ID。
关于获取客流信息的装置中的各模块的处理流程、以及各模块之间的交互流程的描述可以参照上述获取客流信息的方法实施例中的相关说明，这里不再详述。
本公开实施例还提供了一种计算机设备，如图7所示，为本公开实施例提供的一种计算机设备结构示意图，包括：
处理器71和存储器72；所述存储器72存储有处理器71可执行的机器可读指令，处理器71用于执行存储器72中存储的机器可读指令，所述机器可读指令被处理器71执行时，处理器71执行下述步骤：
获取至少一个目标设备对预设区域进行拍摄得到的视频流；预设区域与目标区域的出入口相邻，并且不在目标区域内；
确定与目标区域对应的客流判定标志；
识别视频流中的目标对象，并确定目标对象的运行轨迹；
基于目标对象的运行轨迹和客流判定标志，确定目标区域内的客流信息。
上述存储器72包括内存721和外部存储器722；这里的内存721也称内存储器，用于暂时存放处理器71中的运算数据，以及与硬盘等外部存储器722交换的数据，处理器71通过内存721与外部存储器722进行数据交换。
上述指令的具体执行过程可以参考本公开实施例中所述的获取客流信息的方法的步骤，此处不再赘述。
本公开实施例还提供一种计算机可读存储介质，该计算机可读存储介质上存储有计算机程序，该计算机程序被处理器运行时执行上述方法实施例中所述的获取客流信息的方法的步骤。其中，该存储介质可以是易失性或非易失的计算机可读取存储介质。
本公开实施例所提供的获取客流信息的方法的计算机程序产品，包括计算机可读代码，或者存储了程序代码的计算机可读存储介质，当所述计算机可读代码在电子设备的处理器中运行时，所述程序代码包括的指令可用于执行上述方法实施例中所述的获取客流信息的方法的步骤，具体可参见上述方法实施例，在此不再赘述。
该计算机程序产品可以具体通过硬件、软件或其结合的方式实现。在一个可选实施 例中，所述计算机程序产品具体体现为计算机存储介质，在另一个可选实施例中，计算机程序产品具体体现为软件产品，例如软件开发包(Software Development Kit，SDK)等等。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的装置的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。在本公开所提供的几个实施例中，应该理解到，所揭露的装置和方法，可以通过其它的方式实现。以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，又例如，多个单元或组件可以结合或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些通信接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本公开各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个处理器可执行的非易失的计算机可读取存储介质中。基于这样的理解，本公开的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本公开各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
最后应说明的是：以上所述实施例，仅为本公开的具体实施方式，用以说明本公开的技术方案，而非对其限制，本公开的保护范围并不局限于此，尽管参照前述实施例对本公开进行了详细的说明，本领域的普通技术人员应当理解：任何熟悉本技术领域的技术人员在本公开揭露的技术范围内，其依然可以对前述实施例所记载的技术方案进行修改或可轻易想到变化，或者对其中部分技术特征进行等同替换；而这些修改、变化或者替换，并不使相应技术方案的本质脱离本公开实施例技术方案的精神和范围，都应涵盖在本公开的保护范围之内。因此，本公开的保护范围应所述以权利要求的保护范围为准。

Claims (13)

  1. 一种获取客流信息的方法，其特征在于，包括：
    获取至少一个目标设备对预设区域进行拍摄得到的视频流；所述预设区域与目标区域的出入口相邻，并且不在所述目标区域内；
    确定与所述目标区域对应的客流判定标志；
    识别所述视频流中的目标对象，并确定所述目标对象的运行轨迹；
    基于所述目标对象的运行轨迹和所述客流判定标志，确定所述目标区域内的客流信息。
  2. 根据权利要求1所述的方法，其特征在于，所述基于所述目标对象的运行轨迹和所述客流判定标志，确定所述目标区域内的客流信息，包括：
    在确定所述运行轨迹与所述客流判定标志相交的情况下，确定所述运行轨迹相对于所述客流判定标志的移动方向；
    基于所述移动方向，确定所述目标区域内的客流信息。
  3. 根据权利要求2所述的方法，其特征在于，所述客流信息包括所述目标区域内的有效客流量；
    所述基于所述移动方向，确定所述目标区域内的客流信息，包括：
    在所述移动方向包括指示所述目标对象进入所述目标区域的第一方向和指示所述目标对象离开所述目标区域的第二方向的情况下，将对应的所述目标对象作为所述目标区域的有效客流对象；
    将所述有效客流对象的数量，作为所述目标区域内的有效客流量。
  4. 根据权利要求3所述的方法，其特征在于，所述客流信息还包括所述有效客流对象在所述目标区域内的停留时长；
    所述基于所述移动方向，确定所述目标区域内的客流信息，还包括：
    基于所述有效客流对象的运行轨迹和所述客流判定标志，分别确定所述有效客流对象的运行轨迹在所述第一方向上与所述客流判定标志相交的第一时间，以及确定所述有效客流对象的运行轨迹在所述第二方向上与所述客流判定标志相交的第二时间；
    基于所述第一时间和所述第二时间，确定所述有效客流对象在所述目标区域内的停留时长。
  5. 根据权利要求4所述的方法，其特征在于，所述客流信息还包括所述有效客流对象在所述目标区域内的平均停留时长；
    所述基于所述移动方向，确定所述目标区域内的客流信息，还包括：
    基于至少部分所述有效客流对象在所述目标区域的停留时长和所述有效客流对象的数量，确定所述有效客流对象在所述目标区域内的平均停留时长。
  6. 根据权利要求3所述的方法，其特征在于，所述客流信息还包括所述目标区域的进入客流量和离开客流量中的至少一项；
    所述基于所述移动方向，确定所述目标区域内的客流信息，包括如下至少一项：
    在所述移动方向包括指示所述目标对象进入所述目标区域的第一方向的情况下，将对应的所述目标对象作为所述目标区域的进入客流对象；将所述进入客流对象的数量，作为所述目标区域内的进入客流量；
    在所述移动方向包括指示所述目标对象离开所述目标区域的第二方向的情况下，将 对应的所述目标对象作为所述目标区域的离开客流对象；将所述离开客流对象的数量，作为所述目标区域内的离开客流量。
  7. 根据权利要求1所述的方法，其特征在于，所述识别所述视频流中的目标对象，并确定所述目标对象的运行轨迹，包括：
    识别所述视频流中的目标对象，确定所述目标对象的轨迹点以及所述轨迹点的时序信息；
    基于所述目标对象的轨迹点以及所述轨迹点的时序信息，确定所述目标对象的运行轨迹。
  8. 根据权利要求1所述的方法，其特征在于，所述确定与所述目标区域对应的客流判定标志，包括：
    确定所述预设区域与所述目标区域的出入口的相对位置信息；
    基于所述相对位置信息和所述预设区域的位置信息，确定与所述目标区域对应的所述客流判定标志。
  9. 根据权利要求1至8任一项所述的方法，其特征在于，在基于所述目标对象的运行轨迹和所述客流判定标志，确定所述目标区域内的客流信息之后，还包括：
    基于所述客流信息，生成并上传客流统计报告；
    所述客流统计报告包括所述客流信息，或者，包括所述客流信息以及以下至少一项：报告标识信息、视频流展示设备身份标识ID、目标设备身份标识ID、目标区域身份标识ID。
  10. 一种获取客流信息的装置，其特征在于，包括：
    视频流获取模块，用于获取至少一个目标设备对预设区域进行拍摄得到的视频流；所述预设区域与目标区域的出入口相邻，并且不在所述目标区域内；
    标志确定模块，用于确定与所述目标区域对应的客流判定标志；
    运行轨迹确定模块，用于识别所述视频流中的目标对象，并确定所述目标对象的运行轨迹；
    客流信息确定模块，用于基于所述目标对象的运行轨迹和所述客流判定标志，确定所述目标区域内的客流信息。
  11. 一种计算机设备，其特征在于，包括：处理器、存储器，所述存储器存储有所述处理器可执行的机器可读指令，所述处理器用于执行所述存储器中存储的机器可读指令，所述机器可读指令被所述处理器执行时，所述处理器执行如权利要求1至9任意一项所述的获取客流信息的方法的步骤。
  12. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质上存储有计算机程序，所述计算机程序被计算机设备运行时，所述计算机设备执行如权利要求1至9任意一项所述的获取客流信息的方法的步骤。
  13. 一种计算机程序产品，包括计算机可读代码，或者承载有计算机可读代码的计算机可读存储介质，当所述计算机可读代码在电子设备的处理器中运行时，所述电子设备中的处理器执行用于实现权利要求1至9中的任一权利要求所述的获取客流信息的方法的步骤。
PCT/CN2022/092492 2021-08-30 2022-05-12 获取客流信息的方法、装置、计算机设备以及存储介质 WO2023029574A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111007027.X 2021-08-30
CN202111007027.XA CN113705470A (zh) 2021-08-30 2021-08-30 获取客流信息的方法、装置、计算机设备以及存储介质

Publications (1)

Publication Number Publication Date
WO2023029574A1 true WO2023029574A1 (zh) 2023-03-09

Family

ID=78657025

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/092492 WO2023029574A1 (zh) 2021-08-30 2022-05-12 获取客流信息的方法、装置、计算机设备以及存储介质

Country Status (2)

Country Link
CN (1) CN113705470A (zh)
WO (1) WO2023029574A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113705470A (zh) * 2021-08-30 2021-11-26 北京市商汤科技开发有限公司 获取客流信息的方法、装置、计算机设备以及存储介质
CN114155488A (zh) * 2021-11-30 2022-03-08 北京市商汤科技开发有限公司 获取客流数据的方法、装置、电子设备以及存储介质
CN114821487B (zh) * 2022-06-29 2022-10-04 珠海视熙科技有限公司 客流统计方法、装置、智能终端、***、设备及介质
CN117523472A (zh) * 2023-09-19 2024-02-06 浙江大华技术股份有限公司 客流数据统计方法、计算机设备及计算机可读存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111738134A (zh) * 2020-06-18 2020-10-02 北京市商汤科技开发有限公司 获取客流数据的方法、装置、设备及介质
CN112052838A (zh) * 2020-10-10 2020-12-08 腾讯科技（深圳）有限公司 一种对象流量数据监控方法、装置以及可读存储介质
CN112257520A (zh) * 2020-09-30 2021-01-22 洛伦兹（北京）科技有限公司 人流量统计方法、装置及***
WO2021043089A1 (zh) * 2019-09-02 2021-03-11 平安科技（深圳）有限公司 一种客流量统计方法、装置、设备及计算机可读存储介质
CN112541440A (zh) * 2020-12-16 2021-03-23 中电海康集团有限公司 一种基于视频行人识别的地铁人流网络融合方法及人流预测方法
CN113705470A (zh) * 2021-08-30 2021-11-26 北京市商汤科技开发有限公司 获取客流信息的方法、装置、计算机设备以及存储介质

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108985218A (zh) * 2018-07-10 2018-12-11 上海小蚁科技有限公司 人流量统计方法及装置、存储介质、计算设备
CN110942055A (zh) * 2019-12-31 2020-03-31 北京市商汤科技开发有限公司 展示区域的状态识别方法、装置、设备及存储介质
CN111476183A (zh) * 2020-04-13 2020-07-31 腾讯科技（深圳）有限公司 客流信息处理方法及装置
CN112508626A (zh) * 2020-12-18 2021-03-16 北京市商汤科技开发有限公司 一种信息处理方法、装置、电子设备及存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021043089A1 (zh) * 2019-09-02 2021-03-11 平安科技（深圳）有限公司 一种客流量统计方法、装置、设备及计算机可读存储介质
CN111738134A (zh) * 2020-06-18 2020-10-02 北京市商汤科技开发有限公司 获取客流数据的方法、装置、设备及介质
CN112257520A (zh) * 2020-09-30 2021-01-22 洛伦兹（北京）科技有限公司 人流量统计方法、装置及***
CN112052838A (zh) * 2020-10-10 2020-12-08 腾讯科技（深圳）有限公司 一种对象流量数据监控方法、装置以及可读存储介质
CN112541440A (zh) * 2020-12-16 2021-03-23 中电海康集团有限公司 一种基于视频行人识别的地铁人流网络融合方法及人流预测方法
CN113705470A (zh) * 2021-08-30 2021-11-26 北京市商汤科技开发有限公司 获取客流信息的方法、装置、计算机设备以及存储介质

Also Published As

Publication number Publication date
CN113705470A (zh) 2021-11-26

Similar Documents

Publication Publication Date Title
WO2023029574A1 (zh) 获取客流信息的方法、装置、计算机设备以及存储介质
US10735694B2 (en) System and method for activity monitoring using video data
US11562021B2 (en) Coordinating communication and/or storage based on image analysis
US10726271B2 (en) Virtual turnstile system and method
KR101759453B1 (ko) 자동 이미지 크로핑 및 공유
US9210385B2 (en) Method and system for metadata extraction from master-slave cameras tracking system
JP6397581B2 (ja) 混雑状況可視化装置、混雑状況可視化システム、混雑状況可視化方法、および混雑状況可視化プログラム
US11100368B2 (en) Accelerated training of an image classifier
KR102097781B1 (ko) 객체 정보 제공 장치 및 방법
JP2018518694A (ja) スマートフィルムスクリーンを利用した広告ディスプレイシステム
JP2012252613A (ja) 顧客行動追跡型映像配信システム
WO2022213540A1 (zh) 目标检测、属性识别与跟踪方法及***
CN111435435B (zh) 一种同伴识别方法、装置、服务器及***
Chandran et al. Real-time identification of pedestrian meeting and split events from surveillance videos using motion similarity and its applications
WO2023173616A1 (zh) 一种人群统计方法及装置、电子设备和存储介质
TW202226114A (zh) 資訊處理方法、裝置、電子設備及儲存媒體、計算機程式
CN113486799A (zh) 设备联动方法、装置、设备、存储介质及程序产品
CN111178113B (zh) 一种信息处理方法、装置及存储介质
US20210158399A1 (en) System and method for interactive perception and content presentation
CN113822367B (zh) 一种基于人脸的区域行为分析方法、***及介质
CN112749363B (zh) 网络广告发布与获取方法、装置、存储介质和计算机设备
AU2020202221A1 (en) Privacy preserving visitor recognition and movement pattern analysis based on computer vision
JP2015162241A (ja) 情報提供効果を測定する測定システム及び情報提供効果測定方法
CN114820692B (zh) 跟踪目标的状态分析方法、装置、存储介质和终端
Marcenaro Access to data sets

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE