WO2021249306A1 - 人群过密预测方法及装置 - Google Patents

人群过密预测方法及装置 Download PDF

Info

Publication number
WO2021249306A1
WO2021249306A1 PCT/CN2021/098382 CN2021098382W WO2021249306A1 WO 2021249306 A1 WO2021249306 A1 WO 2021249306A1 CN 2021098382 W CN2021098382 W CN 2021098382W WO 2021249306 A1 WO2021249306 A1 WO 2021249306A1
Authority
WO
WIPO (PCT)
Prior art keywords
area
tested
people
bayonet
flow
Prior art date
Application number
PCT/CN2021/098382
Other languages
English (en)
French (fr)
Inventor
杨昆霖
刘诗男
侯军
伊帅
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
Priority to EP21821878.2A priority Critical patent/EP4036794A4/en
Priority to BR112022011625A priority patent/BR112022011625A2/pt
Priority to JP2022532700A priority patent/JP2023503528A/ja
Priority to KR1020227013875A priority patent/KR20220063280A/ko
Publication of WO2021249306A1 publication Critical patent/WO2021249306A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • G06V20/53—Recognition of crowd images, e.g. recognition of crowd congestion
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0635—Risk analysis of enterprise or organisation activities
    • G—PHYSICS
    • G07—CHECKING-DEVICES
    • G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
    • G07C9/00—Individual registration on entry or exit

Definitions

  • the present disclosure relates to the field of computer vision, and in particular to a method and device for predicting crowd over-density.
  • Crowd density detection uses image recognition technology to detect whether the number of people in a closed area reaches the limit. When it is determined that the number of people in a closed area is too dense, officials can be reminded to limit the flow of people entering the closed area to avoid dangerous incidents.
  • a first aspect of the present disclosure provides a crowd over-density prediction method, which is applied to a server, and the method includes: acquiring a multi-frame image of a bayonet of an area to be tested; and according to the multi-frame image of a bayonet of the area to be tested , Determine the number of people accommodated in the area to be tested and the net inflow rate of people in the area to be tested; The net inflow speed of people flow determines the time when the crowds in the area to be tested are too dense.
  • determining the number of people accommodated in the area to be tested according to the multi-frame images of the bayonet of the area to be tested includes: from the bayonet of the area to be tested The number of people entering the area to be tested and the number of people leaving the area to be tested are identified in the multi-frame images; according to the number of people entering the area to be tested and the number of people leaving the area to be tested, the number of people in the area to be tested is determined The number of people has been accommodated.
  • determining the net inflow velocity of people flow in the area to be measured based on the multi-frame images of the bayonet of the area to be measured includes: Identifying the number of people entering the area to be tested and the number of people leaving the area to be tested in the target time period in the multiple frames of images; according to the number of people entering the area to be tested and leaving the area to be tested in the target time period The number of people in the area, determine the flow of people in the bayonet and the flow of people out of the bayonet; according to the flow of people in the bayonet and the flow of people out of the bayonet, determine the flow of people in the area to be tested Net inflow velocity.
  • determining the net inflow velocity of people flow in the area to be measured based on the multi-frame images of the bayonet of the area to be measured includes: Identifying the number of people entering the area to be tested and the number of people leaving the area to be tested in the target time period in the multiple frames of images; according to the number of people entering the area to be tested and leaving the area to be tested in the target time period The number of people in the area determines the net inflow of people in the area to be tested; the net inflow rate of people in the area to be tested is determined according to the net inflow of people in the area to be tested.
  • the secret time includes: determining the remaining capacity of the area to be tested according to the capacity of the area to be tested and the capacity of the area to be tested; according to the remaining capacity of the area to be tested and The net inflow speed of the people flow in the area to be tested determines the time when the people in the area to be tested are too dense.
  • the method before determining the time when the crowd in the area to be tested is too dense, the method further includes: determining the area to be tested if the net inflow velocity of people in the area to be tested is positive There is a risk of over-population.
  • the method before determining the time when the crowd in the area to be tested is too dense, the method further includes: if the net inflow velocity of people in the area to be tested is negative, determining the area to be tested There is no risk of over-population.
  • the method further includes: sending the over-density time of the crowd in the area to be tested to a terminal device.
  • determining the number of people accommodated in the area to be tested based on the multi-frame images of the bayonet of the area to be tested includes: For each bayonet of the multiple bayonet: identify the number of people entering the area to be tested from the bayonet and the number of people leaving the area to be tested from the bayonet from the multi-frame images of the bayonet ; According to the number of people entering the area to be tested from the checkpoint and the number of people leaving the area to be tested from the checkpoint, determine the net inflow of people from the checkpoint; add up the net inflow of people from all the checkpoints, Get the number of people accommodated in the area to be tested.
  • determining the number of people accommodated in the area to be tested based on the multi-frame images of the bayonet of the area to be tested includes: For each of the multiple bayonet ports, identify the number of people entering the area to be tested from the bayonet and the number of people leaving the area to be tested from the bayonet from the multi-frame images of the bayonet ; Add the number of people who enter the area to be tested from all the checkpoints to get the total number of people who enter the area to be tested; add the number of people who leave the area to be tested from all the checkpoints to get the number of people who leave the area to be tested The total number of people in the area; the total number of people entering the area to be tested is subtracted from the total number of people leaving the area to be tested to determine the number of people accommodated in the area to be tested.
  • determining the net inflow velocity of people flow in the area to be tested based on the multi-frame images of the bayonet of the area to be tested includes : For each bayonet of the multiple bayonet: identify the number of people entering the area to be tested from the bayonet in the target time period from the multiple frames of the bayonet and leaving the said bayonet from the bayonet The number of people in the area to be tested; the number of people who will enter the area to be tested from the bayonet is divided by the duration of the target time period to get the inflow speed of the bayonet; people who will leave the area to be tested from the bayonet Divide the number of people by the duration of the target time period to obtain the outflow speed of the flow of people at the bayonet; add the flow inflow speeds of all the checkpoints to obtain the flow inflow speed of the area to be measured; divide the flow out speed of all the bayonets Add up to obtain the outflow speed of the people
  • determining the net inflow velocity of people flow in the area to be tested based on the multi-frame images of the bayonet of the area to be tested includes : For each of the multiple bayonet ports: identify the number of people entering the area to be tested from the bayonet at the target time period from the multiple frames of the bayonet and leaving the waiting area from the bayonet The number of people in the area to be tested; the number of people entering the area to be tested from the bayonet is divided by the duration of the target time period to get the inflow speed of the bayonet; the number of people leaving the area to be tested from the bayonet Divide by the duration of the target time period to obtain the flow rate of people flow at the bayonet; subtract the flow rate of people flow from the bayonet from the flow rate of the bayonet to obtain the net flow rate of people flow at the bayonet; The net inflow velocity of people in the mouth is added to determine the net inflow velocity of people in the area to
  • a second aspect of the present disclosure provides a crowd over-density prediction device.
  • the device includes: an acquisition module for acquiring multi-frame images of a bayonet of an area to be tested; a first determining module for The multi-frame images of the bayonet determine the number of people accommodated in the area to be tested and the net inflow rate of people in the area to be tested; the second determining module is used to determine the number of people accommodated in the area to be tested, The number of persons that can be accommodated in the area to be tested and the net inflow speed of people in the area to be tested are used to determine the time when the crowd in the area to be tested is too dense.
  • the first determining module is specifically configured to identify the number of people entering the area to be tested and the number of people who leave the area to be tested from the multi-frame images of the bayonet of the area to be tested. The number of people in the area; according to the number of people entering the area to be tested and the number of people leaving the area to be tested, the number of people that have been accommodated in the area to be tested is determined.
  • the first determining module is specifically configured to identify the number of people entering and leaving the area to be tested in the target time period from the multi-frame images of the bayonet of the area to be tested.
  • the number of people in the area to be tested; the number of people entering the area to be tested and the number of people leaving the area to be tested within the target time period are used to determine the flow rate of people in the bayonet and the flow of people out of the bayonet Speed; Determine the net inflow speed of the people flow in the area to be measured according to the inflow speed of the people flow through the bayonet and the flow out speed of the people flow through the bayonet.
  • the second determining module is specifically configured to determine the remaining capacity of the area to be tested based on the capacity of the area to be tested and the capacity of the area to be tested ; According to the remaining number of people in the area to be tested and the net inflow rate of people in the area to be tested, determine the time when the crowd in the area to be tested is too dense.
  • the device further includes: a third determining module, configured to determine that there is a risk of over-density of crowds in the area to be tested if the net inflow velocity of the people flow in the area to be tested is positive.
  • the third determining module is further configured to determine that there is no risk of crowd density in the area to be tested if the net inflow velocity of the flow of people in the area to be tested is negative.
  • the device further includes: a sending module, configured to send the over-secret time of the crowd in the area to be tested to the terminal device.
  • a third aspect of the present disclosure provides an electronic device, including a memory and a processor; the memory is used to store executable instructions of the processor; the processor is configured to execute the first section of the present disclosure by executing the executable instructions On the one hand and on the first aspect, various optional crowd over-density prediction methods.
  • a fourth aspect of the present disclosure provides a storage medium in which a computer program is stored, and when the computer program is executed by a processor, the first aspect and various optional crowd crowd prediction methods of the first aspect are implemented.
  • a fifth aspect of the present disclosure provides a computer program that, when the computer program is executed by a processor, causes the processor to execute the first aspect and various optional crowd over-density prediction methods of the first aspect.
  • FIG. 1 is a schematic diagram of an application scenario of a method for predicting over-density of a crowd provided by an embodiment of the application;
  • FIG. 2 is a schematic flowchart of a method for predicting over-density of a crowd provided by an embodiment of the application;
  • FIG. 3 is a schematic flowchart of another method for predicting over-density of crowds provided by an embodiment of the application;
  • FIG. 4 is a schematic flowchart of another method for predicting over-density of crowds provided by an embodiment of the application
  • FIG. 5 is a schematic structural diagram of a crowd over-density prediction device provided by an embodiment of the application.
  • FIG. 6 is a schematic structural diagram of an electronic device provided by an embodiment of the application.
  • the image of the enclosed area can usually be recognized, so that the crowd counting method can be used to count the current number of people in the enclosed area.
  • the crowd counting method can be used to count the current number of people in the enclosed area.
  • only counting the current number of people in a closed area cannot accurately predict the time when the crowd in the closed area is too dense, and thus it is impossible to take measures in advance to prevent the crowd from being too dense.
  • the embodiments of the present application provide a method and device for predicting over-density of crowds, so as to solve the problem of not accurately predicting the time of over-density of crowds in closed areas.
  • the time that the crowd in the area to be tested is too dense is determined, which can improve Accuracy of prediction of crowd over-density time.
  • FIG. 1 is a schematic diagram of an application scenario of a method for predicting over-density of a crowd provided by an embodiment of the application.
  • the image acquisition device 101 can collect images of the bayonet of the area to be measured in real time, and send the image of the bayonet of the area to be measured to the server 102.
  • the server 102 judges whether there is a risk of over-density of crowds in the area to be tested according to the multi-frame images of the bayonet of the area to be tested. If there is a risk of over-density of crowds, it may further determine the time when the crowds are over-densified. Subsequently, the server 102 may send the crowd over-density time to the terminal device 103, so that the manager can take measures to prevent the crowd over-density based on the crowd over-density time displayed on the terminal device 103.
  • the image acquisition device 101 may include a camera component such as a camera.
  • the server 102 may be a server or a server in a cloud service platform.
  • the server 102 may receive the image of the bayonet of the area to be tested sent by the image acquisition device 101, and send the crowd over-secret time to the terminal device 103.
  • the terminal device 103 may be a mobile phone (mobile phone), a tablet computer (pad), a computer with wireless transceiver function, virtual reality (VR) terminal equipment, augmented reality (AR) terminal equipment, industrial control (industrial control) Wireless terminal in control), wireless terminal in self-driving (self-driving), wireless terminal in remote medical surgery, wireless terminal in smart grid (smart grid), smart home (smart home) Wireless terminal, etc.
  • the device used to implement the function of the terminal may be a terminal, or a device capable of supporting the terminal to implement the function, such as a chip system, and the device may be installed in the terminal.
  • the chip system may be composed of chips, or may include chips and other discrete devices.
  • the area to be tested can be a closed area, including closed areas such as buildings, parks, and libraries.
  • the application scenario of the embodiment of the present application may be the application scenario in FIG. 1, but is not limited to this, and the embodiment of the present application may also be applied to other scenarios that require over-population prediction.
  • the crowd over-density prediction method can be implemented by the crowd over-density prediction apparatus provided in the embodiment of the present application.
  • the crowd over-density prediction apparatus may be part or all of a certain device, such as a server or a processor in the server.
  • FIG. 2 is a schematic flowchart of a method for predicting over-density of a crowd provided by an embodiment of the application, and the execution subject of this embodiment is a server. As shown in Fig. 2, the method includes step S201 to step S203.
  • S201 Acquire multiple frames of images of the bayonet of the area to be measured.
  • the area to be tested may be a closed area, which may include closed areas such as buildings, parks, and libraries.
  • the bayonet can be the entrance and exit of the area to be tested.
  • the bayonet of the area to be measured may be provided with an image acquisition device that can collect images of the bayonet of the area to be measured in real time, and send the image of the bayonet of the area to be measured to the server for Make the server store the image of the bayonet of the area to be tested in its memory.
  • the server needs to predict the time when the crowd is too dense, it can extract multi-frame images of the bayonet of the area to be tested from the memory.
  • the embodiment of the present application does not limit the number of bayonet ports in the area to be tested, and it may be one or multiple.
  • the server needs to obtain multiple frames of images from multiple bayonet ports.
  • S202 Determine the number of people accommodated in the area to be measured and the net inflow velocity of people in the area to be measured according to the multi-frame images of the bayonet of the area to be measured.
  • the server After the server obtains the multi-frame images of the bayonet of the area to be tested, it can determine the number of people accommodated in the area to be tested and the net inflow of people in the area to be tested based on the multi-frame images of the bayonet of the area to be tested speed.
  • the net inflow speed of people flow can be the difference between the inflow speed of people flow and the outflow speed of people flow.
  • the server may first identify the number of people entering the area to be tested and the number of people leaving the area to be tested from the multi-frame images of the bayonet of the area to be tested. Subsequently, the server determines the number of people that have been accommodated in the area to be tested based on the number of people entering the area to be tested and the number of people leaving the area to be tested.
  • the server can determine the net inflow of each bayonet according to the number of people entering the area to be tested and the number of people leaving the area to be tested, and then calculate the net number of Add the number of inflows to get the number of people accommodated in the area to be tested. Or, the server can add the number of people entering the area to be tested for each bayonet to get the total number of people entering the area to be tested. Add the number of people who left the area to be tested for each bayonet to get the total number of people who left the area to be tested. Finally, the total number of people entering the area to be tested and the total number of people leaving the area to be tested are subtracted to determine the number of people that have been accommodated in the area to be tested.
  • the embodiment of the present application does not limit how to determine the number of people entering the area to be tested and the number of people leaving the area to be tested, and any available image recognition technology can be used.
  • the number of people entering the area to be tested and the number of people leaving the area to be tested can be determined based on the cross-line counting method.
  • the server can first identify from the multiple frames of images of the bayonet of the area to be tested that the target time period has entered the area to be tested. The number of people in the area and the number of people leaving the area to be tested, and then according to the number of people entering the area to be tested and the number of people leaving the area to be tested within the target time period, determine the inflow speed of the bayonet and the outflow speed of the bayonet. Subsequently, the server then determines the net inflow speed of the people flow in the area to be tested based on the inflow speed of the people flow through the bayonet and the flow out speed of the people flow through the bayonet.
  • the server may first identify the number of people entering the area to be tested and the number of people leaving the area to be tested in the target time period from the multi-frame images of the bayonet of the area to be tested, and then enter the target time period according to the number of people entering the area to be tested.
  • the number of people in the area and the number of people leaving the area to be tested determines the net inflow of the number of people in the area to be tested.
  • the server determines the net inflow rate of people in the area to be tested based on the net inflow of people in the area to be tested.
  • the present application does not limit the duration of the target time period, and the target time period may be a time period with the time point at which the crowd is too dense (referred to as the detection time point) as the end time point.
  • the target time period may be 60 seconds before the detection time point, or the target time period may be 120 seconds before the detection time point.
  • S203 Determine a time when the crowd in the area to be tested is over-secret based on the number of people accommodated in the area to be tested, the number of people that can be accommodated in the area to be tested, and the net inflow speed of people in the area to be tested.
  • the server determines the capacity of the area to be tested and the net inflow rate of people in the area to be tested, it can be based on the capacity of the area to be tested, the capacity of the area to be tested, and the net inflow of people in the area to be tested. Speed, to determine the time when the crowd in the area to be tested is too dense.
  • the capacity of the area to be tested can be preset, and this application does not limit the capacity of the area to be tested. Exemplarily, it can be predicted based on the area to be tested. For the area to be tested with a large area, a larger capacity can be set, and for the area to be tested with a small area, a smaller capacity can be set. Number of people.
  • the server may determine the remaining capacity of the area to be tested based on the number of people that have been accommodated in the area to be tested and the capacity of the area to be tested. Subsequently, the server then determines the time when the crowd in the area to be tested is too dense based on the remaining capacity of the area to be tested and the net inflow rate of people in the area to be tested.
  • the server may also detect the net inflow rate of people in the area to be tested. If the net inflow rate of people flow in the area to be tested is negative, it means that more people flow out of the area to be tested. The server determines that there is no risk of over-density of people in the area to be tested, and the server does not need to determine the time of over-density of people.
  • the server determines that there is a risk of over-density of people in the area to be tested, and the server needs to determine the time when the crowd is over-densified.
  • the server can determine that the people in the area to be tested are too dense.
  • the server may send the over-secret time of the crowd in the area to be tested to the terminal device, so that the terminal device can display the over-secret time of the crowd and inform the management personnel.
  • a reminder of over-density of crowds is issued so that managers can take timely measures to prevent over-density of crowds.
  • the crowd over-density prediction method In the crowd over-density prediction method provided by the embodiment of the present application, firstly, multi-frame images of the bayonet of the area to be tested are acquired. Secondly, according to the multi-frame images of the bayonet of the area to be tested, determine the number of people accommodated in the area to be tested and the net inflow rate of people in the area to be tested. Finally, according to the capacity of the area to be tested, the capacity of the area to be tested, and the net inflow rate of people in the area to be tested, determine the time when the crowd in the area to be tested is too dense. This application predicts the crowd over-secure time based on the capacity of the area to be tested, the capacity of the area to be tested, and the net inflow speed of the crowd in the area to be tested, which can improve the accuracy of the prediction of crowd over-secure time.
  • FIG. 3 is a schematic flowchart of another method for predicting over-density of a crowd provided by an embodiment of the application.
  • the execution subject of this embodiment is a server. As shown in FIG. 3, the method includes steps S301 to S306.
  • S301 Acquire multiple frames of images of the bayonet of the area to be measured.
  • Step S301 can be understood with reference to step S201 shown in FIG. 2, and the related content will not be repeated here.
  • S302 Identify the number of people entering the area to be tested and the number of people leaving the area to be tested from the multi-frame images of the bayonet of the area to be tested.
  • the embodiment of the present application does not limit how to recognize the number of people entering and leaving the area to be tested from the multi-frame images of the bayonet of the area to be tested, and any available image recognition technology can be used.
  • the number of people entering the area to be tested and the number of people leaving the area to be tested can be determined based on the cross-line counting method.
  • the number of people entering the area to be tested and the number of people leaving the area to be tested can be counted from the preset time node to the end of the time node when the crowd is over-secret time detected.
  • the embodiment of the present application does not limit the preset time node.
  • the preset time node may be 3 o'clock in the morning every day.
  • S303 Determine the number of people accommodated in the area to be tested based on the number of people entering the area to be tested and the number of people leaving the area to be tested.
  • the server may subtract the number of people leaving the area to be tested by the number of people entering the area to be tested, thereby obtaining the number of people accommodated in the area to be tested.
  • the server can target each bayonet of the multiple bayonet ports: firstly recognize from the multi-frame images of the bayonet that the bayonet enters the to-be-tested area. The number of people in the area and the number of people leaving the area to be tested from the checkpoint. Subsequently, the server can subtract the number of people who leave the area to be tested from the checkpoint from the number of people who enter the area to be tested from the checkpoint to obtain the net inflow of people from the checkpoint. Finally, the server can add the net influx of people from all the bayonet points to get the number of people accommodated in the area to be tested.
  • the server can first identify from the multiple frames of images of the bayonet that the bayonet enters the waiting area for each of the multiple bayonet ports. The number of people in the area to be tested and the number of people leaving the area to be tested from the checkpoint. Subsequently, the server can add the number of people who entered the area to be tested from all the checkpoints to get the total number of people who entered the area to be tested; add the number of people who left the area to be tested from all the checkpoints to get the total number of people who left the area to be tested. Finally, the server can subtract the total number of people leaving the area to be tested from the total number of people entering the area to be tested to obtain the number of people that have been accommodated in the area to be tested.
  • S304 Determine the inflow speed of the people flow through the bayonet and the outflow speed of the people flow through the bayonet according to the number of people entering the area to be measured and the number of people leaving the area to be measured within the target time period.
  • the target time period may be a time period with the detection time point as the end time point.
  • the target time period may be 60 seconds before the detection time point, or the target time period may be 120 seconds before the detection time point.
  • the server can recognize from the multi-frame images of the bayonet port to enter the said bayonet in the target time period.
  • the number of people in the area to be tested and the number of people who leave the area to be tested from the checkpoint Divide the number of people entering the area to be tested from the checkpoint by the duration of the target time period to get the inflow speed of the checkpoint. Divide the number of people leaving the area to be tested from the bayonet by the duration of the target time period to get the flow rate of the flow of people from the bayonet.
  • S305 Determine the net inflow speed of the people flow in the area to be measured according to the inflow speed of the people flow through the bayonet and the outflow speed of the people flow through the bayonet.
  • the server can determine the net inflow speed of people flow in the area to be measured according to the inflow speed of the people flow in the bayonet and the outflow speed of the people flow in the bayonet after determining the inflow speed of the people flow through the bayonet and the flow out speed of the people flow through the bayonet.
  • the server can add the inflow speeds of all bayonets to obtain the inflow velocity of people flow in the area to be tested, and add up the outflow speeds of all bayonets. Get the outflow speed of people flow in the area to be measured. Subsequently, the server can determine the net inflow speed of the people flow in the area to be measured by subtracting the flow out speed of the people flow in the area to be measured from the inflow speed of the people flow in the area to be measured.
  • the server can subtract the flow rate of people out of the bayonet from the flow rate of people out of the bayonet. , To get the net inflow speed of the people flow of the bayonet. Subsequently, the server can add up the net inflow speed of people flow from all the bayonet points to determine the net inflow speed of people flow in the area to be tested.
  • S306 Determine a time when the crowd in the area to be tested is over-secret based on the number of people that have been accommodated in the area to be tested, the number of people that can be accommodated in the area to be tested, and the net inflow rate of people in the area to be tested.
  • Step S306 can be understood with reference to step S203 shown in FIG. 2, and the related content will not be repeated here.
  • FIG. 4 is a schematic flowchart of another method for predicting over-density of crowds according to an embodiment of the application.
  • the execution subject of this embodiment is a server. As shown in FIG. 4, the method includes steps S401 to S404.
  • S401 Acquire multiple frames of images of the bayonet of the area to be measured.
  • S402 Determine the number of people accommodated in the area to be measured and the net inflow velocity of people in the area to be measured according to the multi-frame images of the bayonet of the area to be measured.
  • Steps S401-S402 can be understood with reference to steps S201-S202 shown in FIG. 2, and relevant content will not be repeated here.
  • S403 Determine the remaining capacity of the area to be tested according to the capacity of the area to be tested and the capacity of the area to be tested.
  • the server may obtain the remaining capacity of the area to be tested by subtracting the capacity Cnt_thresh of the area to be tested from the capacity Cnt_cur of the area to be tested.
  • S404 Determine a time when the crowd in the area to be tested is too dense according to the remaining capacity of the area to be tested and the net inflow speed of people in the area to be tested.
  • the server may input the remaining capacity of the area to be tested and the net inflow speed of people in the area to be tested into the algorithm model shown in formula (1) to determine the time when the crowds in the area to be tested are too dense.
  • the formula (1) is as follows:
  • T is the time when the crowd is over-secret
  • Cnt_thresh is the capacity of the area to be tested
  • Cnt_cur is the number of people accommodated in the area to be tested
  • Vin is the inflow rate of people in the area to be tested
  • V out is the outflow rate of people in the area to be tested.
  • multi-frame images of the bayonet of the area to be tested are acquired.
  • the capacity of the area to be tested, and the net inflow rate of people in the area to be tested determine the time when the crowd in the area to be tested is too dense.
  • the over-secret time of the crowd in the area to be tested is sent to the terminal device. This application predicts the crowd over-secure time based on the capacity of the area to be tested, the capacity of the area to be tested, and the net inflow speed of the crowd in the area to be tested, which can improve the accuracy of the prediction of crowd over-secure time.
  • a person of ordinary skill in the art can understand that all or part of the steps in the above method embodiments can be implemented by a program instructing relevant hardware.
  • the foregoing program can be stored in a computer readable storage medium. When the program is executed, the program is executed. Including the steps of the foregoing method embodiment; and the foregoing storage medium includes: ROM, RAM, magnetic disk, or optical disk and other media that can store program codes.
  • FIG. 5 is a schematic structural diagram of a crowd over-density prediction device provided by an embodiment of the application.
  • the device for predicting over-density of people can be implemented by software, hardware or a combination of the two to implement the method for predicting over-density of people in the foregoing embodiment. As shown in FIG.
  • the device for predicting over-density of people includes: an acquiring module 501 for acquiring multi-frame images of the bayonet of the area to be tested; a first determining module 502 for acquiring multi-frame images of the bayonet of the area to be tested Image to determine the number of people in the area to be tested and the net inflow rate of people in the area to be tested; the second determination module 503 is used to determine the number of people in the area to be tested, the number of people in the area to be tested, and the flow of people in the area to be tested The net inflow rate determines the time when the crowd in the area to be tested is too dense.
  • the first determining module 502 is specifically configured to identify the number of people entering the area to be tested and the number of people leaving the area to be tested from the multi-frame images of the bayonet of the area to be tested; The number of people and the number of people leaving the area to be tested determine the number of people that have been accommodated in the area to be tested.
  • the first determining module 502 is specifically configured to identify the number of people entering the area to be tested and the number of people leaving the area to be tested in the target time period from the multi-frame images of the bayonet of the area to be tested; The number of people entering the area to be tested and the number of people leaving the area to be tested within the target time period are determined to determine the inflow speed of the bayonet and the outflow speed of the bayonet; according to the inflow speed of the bayonet and the outflow speed of the bayonet, determine the number of people to be The net inflow velocity of people in the measurement area.
  • the first determining module 502 is specifically configured to identify the number of people entering the area to be tested and the number of people leaving the area to be tested in the target time period from the multi-frame images of the bayonet of the area to be tested; The number of people entering the area to be tested and the number of people leaving the area to be tested during the target time period determine the net inflow of the area to be tested; determine the net inflow rate of people in the area to be tested based on the net influx of the area to be tested.
  • the second determining module 503 is specifically configured to determine the remaining capacity of the area to be tested according to the number of persons that can be accommodated in the area to be tested and the capacity of the area to be tested; according to the remaining number of persons in the area to be tested The number of people that can be accommodated and the net inflow rate of people in the area to be tested determine the time when the crowd in the area to be tested is too dense.
  • the device further includes: a third determining module 505, configured to determine that if the net inflow velocity of the flow of people in the area to be measured is positive, there is a risk of overcrowding in the area to be measured.
  • the third determining module 505 is further configured to determine that if the net inflow velocity of the flow of people in the area to be measured is negative, it is determined that there is no risk of overcrowding in the area to be measured.
  • the device further includes: a sending module 504, configured to send the over-secret time of the crowd in the area to be tested to the terminal device.
  • the first determining module 502 is specifically configured to target each bayonet of the multiple bayonet ports: to identify from the multiple frames of images of the bayonet The number of people entering the area to be tested from the checkpoint and the number of people leaving the area to be tested from the checkpoint; according to the number of people entering the area to be tested from the checkpoint and the number of people leaving the area to be tested from the checkpoint, determine the number of people who The number of net inflows of people; add the net inflows of all the checkpoints to get the number of people accommodated in the area to be tested.
  • the first determining module 502 is specifically configured to identify each bayonet of the multiple bayonet ports from the multi-frame images of the bayonet The number of people entering the area to be tested from the checkpoint and the number of people leaving the area to be tested from the checkpoint; the number of people entering the area to be tested from all the checkpoints is added to get the total number of people entering the area to be tested; from all the checkpoints Add the number of people who left the area to be tested to get the total number of people who left the area to be tested; subtract the total number of people who left the area to be tested from the total number of people who entered the area to be tested to determine the number of people that have been accommodated in the area to be tested.
  • the first determining module 502 is specifically configured to target each bayonet of the multiple bayonet ports: to identify from the multiple frames of images of the bayonet The number of people entering the area to be tested from the checkpoint and the number of people leaving the area to be tested from the checkpoint within the target time period; the number of people entering the area to be tested from the checkpoint is divided by the duration of the target time period to get the checkpoint’s Inflow speed of people flow; divide the number of people leaving the area to be measured from the bayonet by the duration of the target time period to get the outflow speed of the flow of people at the bayonet; add the inflow speeds of all the bayonets to get the inflow of people flow into the area to be measured Speed: Add the outflow speed of people flow from all bayonets to get the outflow speed of people flow in the area to be measured; subtract the outflow speed of people flow in the area to be measured from the outflow speed of people flow in the area to to
  • the first determining module 502 is specifically configured to target each bayonet of the multiple bayonet ports: to identify from the multiple frames of images of the bayonet The number of people who enter the area to be tested from the checkpoint and the number of people who leave the area to be tested from the checkpoint during the target time period; divide the number of people who enter the area to be tested from the checkpoint by the duration of the target time period to get the flow of people at the checkpoint Inflow speed; divide the number of people leaving the area to be tested from the bayonet by the duration of the target time period to get the flow rate of the flow of people at the bayonet; subtract the flow rate of the bayonet from the flow rate of the bayonet to get The net inflow velocity of people flow at this bayonet; the net inflow velocity of people flow at all bayonets is added together to determine the net inflow velocity of people flow in the area to be tested.
  • the crowd over-density prediction device provided in the embodiment of the present application can execute the crowd over-density prediction method in the foregoing method embodiment, and its implementation principles and technical effects are similar, and will not be repeated here.
  • FIG. 6 is a schematic structural diagram of an electronic device provided by an embodiment of the application. As shown in FIG. 6, the electronic device may include: at least one processor 601 and a memory 602. Figure 6 shows an electronic device with a processor as an example.
  • the memory 602 is used to store programs.
  • the program may include program code, and the program code includes computer operation instructions.
  • the memory 602 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), for example, at least one disk memory.
  • the processor 601 is configured to execute computer-executable instructions stored in the memory 602 to implement the above method for predicting crowd density; wherein, the processor 601 may be a central processing unit (CPU for short) or a specific integrated circuit ( Application Specific Integrated Circuit (ASIC for short), or one or more integrated circuits configured to implement the embodiments of the present application.
  • CPU central processing unit
  • ASIC Application Specific Integrated Circuit
  • the communication interface, the memory 602, and the processor 601 may be connected to each other through a bus and complete mutual communication.
  • the bus can be an Industry Standard Architecture (ISA) bus, Peripheral Component (PCI) bus, or Extended Industry Standard Architecture (EISA) bus, etc.
  • ISA Industry Standard Architecture
  • PCI Peripheral Component
  • EISA Extended Industry Standard Architecture
  • the bus can be divided into address bus, data bus, control bus, etc., but it does not mean that there is only one bus or one type of bus.
  • the communication interface, the memory 602, and the processor 601 are integrated on a single chip for implementation, the communication interface, the memory 602, and the processor 601 may complete communication through an internal interface.
  • the embodiment of the present application also provides a crowd over-density prediction system, which includes an image acquisition device, a server, and a terminal device.
  • the image acquisition device is used to collect the image of the bayonet of the area to be tested
  • the server is used to determine the time when the crowd is too dense and send it to the terminal device
  • the terminal device is used to receive and display the time when the crowd is too dense.
  • the embodiment of the present application also provides a chip including a processor and an interface.
  • the interface is used to input and output data or instructions processed by the processor.
  • the processor is used to execute the method provided in the above method embodiment.
  • the chip can be used in a crowd prediction device.
  • This application also provides a computer-readable storage medium, which may include: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), and a random access memory (RAM, Random Access Memory). ), magnetic disks or optical disks, and other media that can store program codes.
  • the computer-readable storage medium stores program information, and the program information is used in the aforementioned crowd over-density prediction method.
  • the embodiment of the present application also provides a program, which is used to execute the crowd over-density prediction method provided in the above method embodiment when the program is executed by the processor.
  • the embodiment of the present application also provides a program product, such as a computer-readable storage medium, in which instructions are stored, which when run on a computer, cause the computer to execute the crowd over-density prediction method provided by the foregoing method embodiments.
  • a program product such as a computer-readable storage medium, in which instructions are stored, which when run on a computer, cause the computer to execute the crowd over-density prediction method provided by the foregoing method embodiments.
  • the computer can be implemented in whole or in part by software, hardware, firmware, or any combination thereof.
  • software it can be implemented in the form of a computer program product in whole or in part.
  • the computer program product includes one or more computer instructions.
  • the computer program instructions When the computer program instructions are loaded and executed on the computer, the processes or functions according to the embodiments of the present disclosure are generated in whole or in part.
  • the computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.
  • Computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium.
  • computer instructions can be transmitted from a website, computer, server, or data center through a cable (such as Coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) to transmit to another website site, computer, server or data center.
  • the computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or data center integrated with one or more available media.
  • the usable medium may be a magnetic medium (for example, a floppy disk, a hard disk, and a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid state disk (SSD)).

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Theoretical Computer Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Tourism & Hospitality (AREA)
  • Development Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Multimedia (AREA)
  • Educational Administration (AREA)
  • Image Analysis (AREA)
  • Closed-Circuit Television Systems (AREA)
  • Alarm Systems (AREA)
  • Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)

Abstract

本公开提供一种人群过密预测方法及装置，方法包括：获取待测区域的卡口的多帧图像；根据待测区域的卡口的多帧图像，确定待测区域的已容纳人数和待测区域的人流净流入速度；根据待测区域的已容纳人数、待测区域的可容纳人数和待测区域的人流净流入速度，确定待测区域的人群过密时间。

Description

人群过密预测方法及装置
相关申请的交叉引用
本专利申请要求于2020年06月08日提交的、申请号为202010510936.4、发明名称为“人群过密预测方法、装置、电子设备及存储介质”的中国专利申请的优先权，该申请的全文以引用的方式并入本文中。
技术领域
本公开涉及计算机视觉领域，尤其涉及一种人群过密预测方法及装置。
背景技术
人群密度检测是通过图像识别技术来检测一个封闭区域内人数是否达到极限。当确定一个封闭区域内人数过密时，可以提醒官方人员注意，限制进入这个封闭区域的人流量，从而避免发生危险事件。
发明内容
本公开第一方面提供一种人群过密预测方法，应用于服务器，所述方法包括：获取待测区域的卡口的多帧图像；根据所述待测区域的卡口的所述多帧图像，确定所述待测区域的已容纳人数和所述待测区域的人流净流入速度；根据所述待测区域的已容纳人数、所述待测区域的可容纳人数和所述待测区域的人流净流入速度，确定所述待测区域的人群过密时间。
一种可选的实施方式中，根据所述待测区域的卡口的所述多帧图像，确定所述待测区域的已容纳人数，包括：从所述待测区域的卡口的所述多帧图像中识别出进入所述待测区域的人数和离开所述待测区域的人数；根据进入所述待测区域的人数和离开所述待测区域的人数，确定所述待测区域的已容纳人数。
一种可选的实施方式中，根据所述待测区域的卡口的所述多帧图像，确定所述待测区域的人流净流入速度，包括：从所述待测区域的卡口的所述多帧图像中识别出目标时间段内进入所述待测区域的人数和离开所述待测区域的人数；根据所述目标时间段内进入所述待测区域的人数和离开所述待测区域的人数，确定所述卡口的人流流入速度和所 述卡口的人流流出速度；根据所述卡口的人流流入速度和所述卡口的人流流出速度，确定所述待测区域的人流净流入速度。
一种可选的实施方式中，根据所述待测区域的卡口的所述多帧图像，确定所述待测区域的人流净流入速度，包括：从所述待测区域的卡口的所述多帧图像中识别出目标时间段内进入所述待测区域的人数和离开所述待测区域的人数；根据所述目标时间段内进入所述待测区域的人数和离开所述待测区域的人数，确定所述待测区域的净流入人数；根据所述待测区域的净流入人数，确定所述待测区域的人流净流入速度。
一种可选的实施方式中，根据所述待测区域的已容纳人数、所述待测区域的可容纳人数和所述待测区域的人流净流入速度，确定所述待测区域的人群过密时间，包括：根据所述待测区域的已容纳人数和所述待测区域的可容纳人数，确定所述待测区域的剩余可容纳人数；根据所述待测区域的剩余可容纳人数和所述待测区域的人流净流入速度，确定所述待测区域的人群过密时间。
一种可选的实施方式中，在确定所述待测区域的人群过密时间之前，所述方法还包括：若所述待测区域的人流净流入速度为正，则确定所述待测区域存在人群过密风险。
一种可选的实施方式中，在确定所述待测区域的人群过密时间之前，所述方法还包括：若所述待测区域的人流净流入速度为负，则确定所述待测区域不存在人群过密风险。
一种可选的实施方式中，在确定所述待测区域的人群过密时间之后，所述方法还包括：向终端设备发送所述待测区域的人群过密时间。
一种可选的实施方式中，若所述待测区域包括多个卡口，根据所述待测区域的卡口的所述多帧图像，确定所述待测区域的已容纳人数，包括：针对所述多个卡口中的每个卡口：从该卡口的多帧图像中识别出从该卡口进入所述待测区域的人数和从该卡口离开所述待测区域的人数；根据从该卡口进入所述待测区域的人数和从该卡口离开所述待测区域的人数，确定该卡口的人流净流入人数；将所有卡口的人流净流入人数相加，得到所述待测区域的已容纳人数。
一种可选的实施方式中，若所述待测区域包括多个卡口，根据所述待测区域的卡口的所述多帧图像，确定所述待测区域的已容纳人数，包括：针对所述多个卡口中的每个卡口，从该卡口的多帧图像中识别出从该卡口进入所述待测区域的人数和从该卡口离开所述待测区域的人数；将从所有卡口进入所述待测区域的人数相加，得到进入所述待测区域的总人数；将从所有卡口离开所述待测区域的人数相加，得到离开所述待测区域的 总人数；将进入所述待测区域的总人数减去离开所述待测区域的总人数，确定所述待测区域的已容纳人数。
一种可选的实施方式中，若所述待测区域包括多个卡口，根据所述待测区域的卡口的所述多帧图像，确定所述待测区域的人流净流入速度，包括：针对所述多个卡口中的每个卡口：从该卡口的多帧图像中识别出目标时间段内从该卡口进入所述待测区域的人数和从该卡口离开所述待测区域的人数；将从该卡口进入所述待测区域的人数除以所述目标时间段的时长，得到该卡口的人流流入速度；将从该卡口离开所述待测区域的人数除以所述目标时间段的时长，得到该卡口的人流流出速度；将所有卡口的人流流入速度相加，得到所述待测区域的人流流入速度；将所有卡口的人流流出速度相加，得到所述待测区域的人流流出速度；将所述待测区域的人流流入速度减去所述待测区域的人流流出速度，确定所述待测区域的人流净流入速度。
一种可选的实施方式中，若所述待测区域包括多个卡口，根据所述待测区域的卡口的所述多帧图像，确定所述待测区域的人流净流入速度，包括：针对所述多个卡口中的每个卡口：从该卡口的多帧图像中识别出目标时间段从该卡口进入所述待测区域的人数和从该卡口离开所述待测区域的人数；将从该卡口进入所述待测区域的人数除以所述目标时间段的时长，得到该卡口的人流流入速度；将从该卡口离开所述待测区域的人数除以所述目标时间段的时长，得到该卡口的人流流出速度；将该卡口的人流流入速度减去该卡口的人流流出速度，得到该卡口的人流净流入速度；将所有卡口的人流净流入速度相加，确定所述待测区域的人流净流入速度。
本公开第二方面提供一种人群过密预测装置，所述装置包括：获取模块，用于获取待测区域的卡口的多帧图像；第一确定模块，用于根据所述待测区域的卡口的所述多帧图像，确定所述待测区域的已容纳人数和所述待测区域的人流净流入速度；第二确定模块，用于根据所述待测区域的已容纳人数、所述待测区域的可容纳人数和所述待测区域的人流净流入速度，确定所述待测区域的人群过密时间。
一种可选的实施方式中，所述第一确定模块具体用于从所述待测区域的卡口的所述多帧图像中识别出进入所述待测区域的人数和离开所述待测区域的人数；根据进入所述待测区域的人数和离开所述待测区域的人数，确定所述待测区域的已容纳人数。
一种可选的实施方式中，所述第一确定模块具体用于从所述待测区域的卡口的所述多帧图像中识别出目标时间段内进入所述待测区域的人数和离开所述待测区域的人数；根据所述目标时间段内进入所述待测区域的人数和离开所述待测区域的人数，确定所述 卡口的人流流入速度和所述卡口的人流流出速度；根据所述卡口的人流流入速度和所述卡口的人流流出速度，确定所述待测区域的人流净流入速度。
一种可选的实施方式中，所述第二确定模块具体用于根据所述待测区域的已容纳人数和所述待测区域的可容纳人数，确定所述待测区域的剩余可容纳人数；根据所述待测区域的剩余可容纳人数和所述待测区域的人流净流入速度，确定所述待测区域的人群过密时间。
一种可选的实施方式中，所述装置还包括：第三确定模块，用于若所述待测区域的人流净流入速度为正，则确定所述待测区域存在人群过密风险。
一种可选的实施方式中，所述第三确定模块还用于若所述待测区域的人流净流入速度为负，则确定所述待测区域不存在人群过密风险。
一种可选的实施方式中，所述装置还包括：发送模块，用于向终端设备发送所述待测区域的人群过密时间。
本公开第三方面提供一种电子设备，包括存储器与处理器；所述存储器用于存储所述处理器的可执行指令；所述处理器配置为经由执行所述可执行指令来执行本公开第一方面及第一方面各种可选的人群过密预测方法。
本公开第四方面提供一种存储介质，所述存储介质中存储有计算机程序，所述计算机程序被处理器执行时实现第一方面及第一方面各种可选的人群过密预测方法。
本公开第五方面提供一种计算机程序，所述计算机程序被处理器执行时，使得所述处理器执行第一方面及第一方面各种可选的人群过密预测方法。
附图说明
为了更清楚地说明本公开中的实施例，下面将对实施例中所需要使用的附图作一简单地介绍，显而易见地，下面描述中的附图是本公开的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。
图1为本申请实施例提供的一种人群过密预测方法的应用场景示意图；
图2为本申请实施例提供的一种人群过密预测方法的流程示意图；
图3为本申请实施例提供的另一种人群过密预测方法的流程示意图；
图4为本申请实施例提供的再一种人群过密预测方法的流程示意图；
图5为本申请实施例提供的一种人群过密预测装置的结构示意图；
图6为本申请实施例提供的一种电子设备的结构示意图。
具体实施方式
下面将结合本公开实施例中的附图，对本公开实施例进行清楚、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。基于本公开中的实施例，本领域普通技术人员在没有做出创造性劳动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。
通常可以对封闭区域的图像进行识别，从而使用人群计数法来统计封闭区域内的当前人数。然而，只统计出封闭区域内的当前人数并不能准确的预测出封闭区域的人群过密时间，进而无法提前采取措施进行人群过密的预防。
本申请实施例提供一种人群过密预测方法及装置，以解决无法准确的预测出封闭区域的人群过密时间的问题。本申请中，基于待测区域的已容纳人数、所述待测区域的可容纳人数和所述待测区域的人流净流入速度，确定出所述待测区域的人群过密时间，从而可以提高人群过密时间的预测准确性。
下面对本申请实施例的应用场景进行说明。
图1为本申请实施例提供的一种人群过密预测方法的应用场景示意图。如图1所示，图像采集设备101可以实时采集待测区域的卡口的图像，并将待测区域的卡口的图像发送给服务器102。服务器102根据待测区域的卡口的多帧图像判断待测区域是否存在人群过密风险，若存在人群过密风险，则可以进一步确定出人群过密时间。随后，服务器102可以将人群过密时间发送给终端设备103，以使管理人员可以基于终端设备103显示的人群过密时间采取措施进行人群过密的预防。
其中，图像采集设备101可包括例如摄像头等摄像组件。
服务器102可以是一台服务器，或者是云服务平台中的服务器。服务器102可以接收图像采集设备101发送的待测区域的卡口的图像，并向终端设备103发送人群过密时间。
终端设备103可以是手机(mobile phone)、平板电脑(pad)、带无线收发功能的电脑、虚拟现实(virtual reality，VR)终端设备、增强现实(augmented reality，AR)终端设备、工业控制(industrial control)中的无线终端、无人驾驶(self driving)中的 无线终端、远程手术(remote medical surgery)中的无线终端、智能电网(smart grid)中的无线终端、智慧家庭(smart home)中的无线终端等。本申请实施例中，用于实现终端的功能的装置可以是终端，也可以是能够支持终端实现该功能的装置，例如芯片系统，该装置可以被安装在终端中。本申请实施例中，芯片系统可以由芯片构成，也可以包括芯片和其他分立器件。
待测区域可以为封闭区域，可包括例如大厦、公园、图书馆等封闭区域。
需要说明的是，本申请实施例的应用场景可以是图1中的应用场景，但并不限于此，本申请实施例还可以应用于其他需要进行人群过密预测的场景。
可以理解，上述人群过密预测方法可以通过本申请实施例提供的人群过密预测装置实现，人群过密预测装置可以是某个设备的部分或全部，例如可以是服务器或者服务器内的处理器。
下面以集成或安装有相关执行代码的服务器为例，对本申请实施例进行详细说明。下面这几个实施例可以相互结合，对于相同或相似的概念或过程可能在某些实施例不再赘述。
图2为本申请实施例提供的一种人群过密预测方法的流程示意图，本实施例的执行主体是服务器。如图2所示，该方法包括步骤S201至步骤S203。
S201、获取待测区域的卡口的多帧图像。
其中，待测区域可以为封闭区域，可包括例如大厦、公园、图书馆等封闭区域。卡口可以为待测区域的出入口。
在一些实施例中，待测区域的卡口可以设置有图像采集设备，该图像采集设备可以实时采集待测区域的卡口的图像，并将待测区域的卡口的图像发送给服务器，以使服务器将待测区域的卡口的图像存储在其存储器中。当服务器需要对人群过密时间进行预测时，可以从存储器中提取待测区域的卡口的多帧图像。
此外，本申请实施例对于待测区域的卡口的数量不做限制，可以为一个，也可以为多个。当待测区域包括多个卡口时，服务器需要获取多个卡口的多帧图像。
S202、根据待测区域的卡口的多帧图像，确定待测区域的已容纳人数和待测区域的人流净流入速度。
在本步骤中，当服务器获取到待测区域的卡口的多帧图像后，可以根据待测区域的 卡口的多帧图像，确定待测区域的已容纳人数和待测区域的人流净流入速度。
其中，人流净流入速度可以为人流流入速度和人流流出速度的差值。
本申请实施例对于如何确定待测区域的已容纳人数不做限制。在一些实施例中，服务器可以先从待测区域的卡口的多帧图像中识别出进入待测区域的人数和离开待测区域的人数。随后，服务器再根据进入待测区域的人数和离开待测区域的人数，确定出待测区域的已容纳人数。若待测区域包括多个卡口，服务器可以根据每个卡口的进入待测区域的人数和离开待测区域的人数，确定每个卡口的净流入人数，再将每个卡口的净流入人数相加得到待测区域的已容纳人数。或者，服务器可以将每个卡口的进入待测区域的人数相加，得到进入待测区域的总人数。将每个卡口的离开待测区域的人数相加，得到离开待测区域的总人数。最后，将进入待测区域的总人数和离开待测区域的总人数相减，确定待测区域的已容纳人数。
需要说明的是，本申请实施例对于如何确定进入待测区域的人数和离开待测区域的人数不做限制，可以采用任意可用的图像识别技术。在一些可选的实施方式中，可以基于跨线计数的方式来确定进入待测区域的人数和离开待测区域的人数。
本申请实施例对于如何确定待测区域的人流净流入速度也不做限制，在一些实施例中，服务器可以先从待测区域的卡口的多帧图像中识别出目标时间段内进入待测区域的人数和离开待测区域的人数，再根据目标时间段内进入待测区域的人数和离开待测区域的人数，确定卡口的人流流入速度和卡口的人流流出速度。随后，服务器再根据卡口的人流流入速度和卡口的人流流出速度，确定待测区域的人流净流入速度。
在一些实施例中，服务器可以先从待测区域的卡口的多帧图像中识别出目标时间段内进入待测区域的人数和离开待测区域的人数，再根据目标时间段内进入待测区域的人数和离开待测区域的人数，确定待测区域的净流入人数。随后，服务器再根据待测区域的净流入人数，确定待测区域的人流净流入速度。
其中，本申请对于目标时间段的时长不做限制，目标时间段可以为以检测人群过密时间的时间点(简称检测时间点)为结束时间点的时间段。示例性的，目标时间段可以为检测时间点的前60秒，或者，目标时间段可以为检测时间点的前120秒。
S203、根据待测区域的已容纳人数、待测区域的可容纳人数和待测区域的人流净流入速度，确定待测区域的人群过密时间。
在本步骤中，当服务器确定待测区域的已容纳人数和待测区域的人流净流入速度， 可以根据待测区域的已容纳人数、待测区域的可容纳人数和待测区域的人流净流入速度，确定待测区域的人群过密时间。
其中，待测区域的可容纳人数可以预先设置，本申请对于待测区域的可容纳人数不做限制。示例性的，可以根据待测区域的占地面积进行预测，针对占地面积大的待测区域可以设置较大的可容纳人数，针对占地面积小的待测区域可以设置较小的可容纳人数。
在一些实施例中，服务器可以根据待测区域的已容纳人数和待测区域的可容纳人数，确定待测区域的剩余可容纳人数。随后，服务器再根据待测区域的剩余可容纳人数和待测区域的人流净流入速度，确定待测区域的人群过密时间。
在一种可选的实施方式中，在确定待测区域的人群过密时间之前，服务器还可以检测待测区域的人流净流入速度。若待测区域的人流净流入速度为负，则说明人流离开待测区域的更多，则服务器确定待测区域不存在人群过密风险，进而服务器无需再确定人群过密时间。若待测区域的人流净流入速度为正，则说明人流进入待测区域的更多，则服务器确定待测区域存在人群过密风险，服务器需确定人群过密时间。
此外，在一些实施例中，若待测区域的已容纳人数大于待测区域的可容纳人数，则服务器可以确定待测区域的人群已经过密。
在一种可选的实施方式中，在确定待测区域的人群过密时间之后，服务器可以向终端设备发送待测区域的人群过密时间，以使终端设备显示人群过密时间以及向管理人员发出人群过密提醒，从而使管理人员可以及时采取措施进行人群过密的预防。
本申请实施例提供的人群过密预测方法，首先获取待测区域的卡口的多帧图像。其次根据待测区域的卡口的多帧图像，确定待测区域的已容纳人数和待测区域的人流净流入速度。最后根据待测区域的已容纳人数、待测区域的可容纳人数和待测区域的人流净流入速度，确定待测区域的人群过密时间。本申请基于待测区域的已容纳人数、待测区域的可容纳人数和待测区域的人流净流入速度预测出人群过密时间，可以提高人群过密时间的预测准确性。
在上述实施例的基础上，下面对于如何确定待测区域的已容纳人数和待测区域的人流净流入速度进行说明。图3为本申请实施例提供的另一种人群过密预测方法的流程示意图，本实施例的执行主体是服务器，如图3所示，该方法包括步骤S301至步骤S306。
S301、获取待测区域的卡口的多帧图像。
步骤S301可参照图2所示的步骤S201理解，在此不再累述相关内容。
S302、从待测区域的卡口的多帧图像中识别出进入待测区域的人数和离开待测区域的人数。
本申请实施例对于如何从待测区域的卡口的多帧图像中识别出进入待测区域的人数和离开待测区域的人数不做限制，可以采用任意可用的图像识别技术。在一些可选的实施方式中，可以基于跨线计数的方式来确定进入待测区域的人数和离开待测区域的人数。
需要说明的是，进入待测区域的人数和离开待测区域的人数可以从预设的时间节点开始统计直至检测人群过密时间的时间节点结束。本申请实施例对于预设的时间节点不做限制，示例性的，预设的时间节点可以为每天的上午3点。
S303、根据进入待测区域的人数和离开待测区域的人数，确定待测区域的已容纳人数。
示例性的，服务器可以通过进入待测区域的人数减去离开待测区域的人数，从而得到待测区域的已容纳人数。
在一些实施例中，若待测区域包括多个卡口，则服务器可以针对多个卡口中的每个卡口：先从该卡口的多帧图像中识别出从该卡口进入待测区域的人数和从该卡口离开待测区域的人数。随后，服务器可以将从该卡口进入待测区域的人数减去从该卡口离开待测区域的人数，得到该卡口的人流净流入人数。最后，服务器可以将所有卡口的人流净流入人数相加，得到待测区域的已容纳人数。
在另一些实施例中，若待测区域包括多个卡口，则服务器可以针对多个卡口中的每个卡口，先从该卡口的多帧图像中识别出从该卡口进入待测区域的人数和从该卡口离开待测区域的人数。随后，服务器可以将从所有卡口进入待测区域的人数相加，得到进入待测区域的总人数；将从所有卡口离开待测区域的人数相加，得到离开待测区域的总人数。最后，服务器可以将进入待测区域的总人数减去离开待测区域的总人数，得到待测区域的已容纳人数。
S304、根据目标时间段内进入待测区域的人数和离开待测区域的人数，确定卡口的人流流入速度和卡口的人流流出速度。
其中，本申请对于目标时间段的时长不做限制，目标时间段可以为以检测的时间点为结束时间点的时间段。示例性的，目标时间段可以为检测时间点的前60秒，或者， 目标时间段可以为检测时间点的前120秒。
示例性的，若待测区域包括多个卡口，针对多个卡口中的每个卡口：服务器可以从该卡口的多帧图像中识别出目标时间段内从该卡口进入所述待测区域的人数和从该卡口离开所述待测区域的人数。将从该卡口进入待测区域的人数除以目标时间段的时长，得到该卡口的人流流入速度。将从该卡口离开待测区域的人数除以目标时间段的时长，得到该卡口的人流流出速度。
S305、根据卡口的人流流入速度和卡口的人流流出速度，确定待测区域的人流净流入速度。
在本步骤中，服务器在确定卡口的人流流入速度和卡口的人流流出速度之后，可以根据卡口的人流流入速度和卡口的人流流出速度，确定待测区域的人流净流入速度。
在一些实施例中，若待测区域包括多个卡口，则服务器可以将所有卡口的人流流入速度相加，得到待测区域的人流流入速度，将所有卡口的人流流出速度相加，得到待测区域的人流流出速度。随后，服务器可以通过将待测区域的人流流入速度减去待测区域的人流流出速度，确定待测区域的人流净流入速度。
在另一些实施例中，若待测区域包括多个卡口，则针对多个卡口中的每个卡口：服务器可以通过将该卡口的人流流入速度减去该卡口的人流流出速度，得到该卡口的人流净流入速度。随后，服务器可以将所有卡口的人流净流入速度相加，确定待测区域的人流净流入速度。
S306、根据待测区域的已容纳人数、待测区域的可容纳人数和待测区域的人流净流入速度，确定待测区域的人群过密时间。
步骤S306可参照图2所示的步骤S203理解，在此不再累述相关内容。
在上述实施例的基础上，下面对于如何确定待测区域的人群过密时间进行说明。图4为本申请实施例提供的再一种人群过密预测方法的流程示意图，本实施例的执行主体是服务器，如图4所示，该方法包括步骤S401至步骤S404。
S401、获取待测区域的卡口的多帧图像。
S402、根据待测区域的卡口的多帧图像，确定待测区域的已容纳人数和待测区域的人流净流入速度。
步骤S401-S402可参照图2所示的步骤S201-S202理解，在此不再累述相关内容。
S403、根据待测区域的已容纳人数和待测区域的可容纳人数，确定待测区域的剩余可容纳人数。
示例性的，服务器可以通过将待测区域的已容纳人数Cnt_cur减去待测区域的可容纳人数Cnt_thresh，得到待测区域的剩余可容纳人数。
S404、根据待测区域的剩余可容纳人数和待测区域的人流净流入速度，确定待测区域的人群过密时间。
示例性的，服务器可以将待测区域的剩余可容纳人数和待测区域的人流净流入速度输入如公式(1)所示的算法模型，确定待测区域的人群过密时间。公式(1)如下所示：
T＝(Cnt_thresh-Cnt_cur)/(V in-V out)      (1)
其中，T为人群过密时间，Cnt_thresh为待测区域的可容纳人数，Cnt_cur为待测区域的已容纳人数，V in为待测区域的人流流入速度，V out为待测区域的人流流出速度。
本申请实施例提供的人群过密预测方法，首先获取待测区域的卡口的多帧图像。其次根据待测区域的卡口的多帧图像，确定待测区域的已容纳人数和待测区域的人流净流入速度。再次根据待测区域的已容纳人数、待测区域的可容纳人数和待测区域的人流净流入速度，确定待测区域的人群过密时间。最后，向终端设备发送待测区域的人群过密时间。本申请基于待测区域的已容纳人数、待测区域的可容纳人数和待测区域的人流净流入速度预测出人群过密时间，可以提高人群过密时间的预测准确性。
本领域普通技术人员可以理解：实现上述方法实施例的全部或部分步骤可以通过程序指令相关的硬件来完成，前述的程序可以存储于一计算机可读取存储介质中，该程序在执行时，执行包括上述方法实施例的步骤；而前述的存储介质包括：ROM、RAM、磁碟或者光盘等各种可以存储程序代码的介质。
图5为本申请实施例提供的一种人群过密预测装置的结构示意图。该人群过密预测装置可以通过软件、硬件或者两者的结合实现，以执行上述实施例中的人群过密预测方法。如图5所示，该人群过密预测装置包括：获取模块501，用于获取待测区域的卡口的多帧图像；第一确定模块502，用于根据待测区域的卡口的多帧图像，确定待测区域的已容纳人数和待测区域的人流净流入速度；第二确定模块503，用于根据待测区域的已容纳人数、待测区域的可容纳人数和待测区域的人流净流入速度，确定待测区域的人群过密时间。
一种可选的实施方式中，第一确定模块502具体用于从待测区域的卡口的多帧图像 中识别出进入待测区域的人数和离开待测区域的人数；根据进入待测区域的人数和离开待测区域的人数，确定待测区域的已容纳人数。
一种可选的实施方式中，第一确定模块502具体用于从待测区域的卡口的多帧图像中识别出目标时间段内进入待测区域的人数和离开待测区域的人数；根据目标时间段内进入待测区域的人数和离开待测区域的人数，确定卡口的人流流入速度和卡口的人流流出速度；根据卡口的人流流入速度和卡口的人流流出速度，确定待测区域的人流净流入速度。
一种可选的实施方式中，第一确定模块502具体用于从待测区域的卡口的多帧图像中识别出目标时间段内进入待测区域的人数和离开待测区域的人数；根据目标时间段内进入待测区域的人数和离开待测区域的人数，确定待测区域的净流入人数；根据待测区域的净流入人数，确定待测区域的人流净流入速度。
一种可选的实施方式中，第二确定模块503具体用于根据待测区域的已容纳人数和待测区域的可容纳人数，确定待测区域的剩余可容纳人数；根据待测区域的剩余可容纳人数和待测区域的人流净流入速度，确定待测区域的人群过密时间。
一种可选的实施方式中，该装置还包括：第三确定模块505，用于若待测区域的人流净流入速度为正，则确定待测区域存在人群过密风险。
一种可选的实施方式中，第三确定模块505还用于若待测区域的人流净流入速度为负，则确定待测区域不存在人群过密风险。
一种可选的实施方式中，该装置还包括：发送模块504，用于向终端设备发送待测区域的人群过密时间。
一种可选的实施方式中，若待测区域包括多个卡口，第一确定模块502具体用于针对多个卡口中的每个卡口：从该卡口的多帧图像中识别出从该卡口进入待测区域的人数和从该卡口离开待测区域的人数；根据从该卡口进入待测区域的人数和从该卡口离开待测区域的人数，确定该卡口的人流净流入人数；将所有卡口的人流净流入人数相加，得到待测区域的已容纳人数。
一种可选的实施方式中，若待测区域包括多个卡口，第一确定模块502具体用于针对多个卡口中的每个卡口，从该卡口的多帧图像中识别出从该卡口进入待测区域的人数和从该卡口离开待测区域的人数；将从所有卡口进入待测区域的人数相加，得到进入待测区域的总人数；将从所有卡口离开待测区域的人数相加，得到离开待测区域的总 人数；将进入待测区域的总人数减去离开待测区域的总人数，确定待测区域的已容纳人数。
一种可选的实施方式中，若待测区域包括多个卡口，第一确定模块502具体用于针对多个卡口中的每个卡口：从该卡口的多帧图像中识别出目标时间段内从该卡口进入待测区域的人数和从该卡口离开待测区域的人数；将从该卡口进入待测区域的人数除以目标时间段的时长，得到该卡口的人流流入速度；将从该卡口离开待测区域的人数除以目标时间段的时长，得到该卡口的人流流出速度；将所有卡口的人流流入速度相加，得到待测区域的人流流入速度；将所有卡口的人流流出速度相加，得到待测区域的人流流出速度；将待测区域的人流流入速度减去待测区域的人流流出速度，确定待测区域的人流净流入速度。
一种可选的实施方式中，若待测区域包括多个卡口，第一确定模块502具体用于针对多个卡口中的每个卡口：从该卡口的多帧图像中识别出目标时间段从该卡口进入待测区域的人数和从该卡口离开待测区域的人数；将从该卡口进入待测区域的人数除以目标时间段的时长，得到该卡口的人流流入速度；将从该卡口离开待测区域的人数除以目标时间段的时长，得到该卡口的人流流出速度；将该卡口的人流流入速度减去该卡口的人流流出速度，得到该卡口的人流净流入速度；将所有卡口的人流净流入速度相加，确定待测区域的人流净流入速度。
本申请实施例提供的人群过密预测装置，可以执行上述方法实施例中的人群过密预测方法，其实现原理和技术效果类似，在此不再赘述。
图6为本申请实施例提供的一种电子设备的结构示意图。如图6所示，该电子设备可以包括：至少一个处理器601和存储器602。图6示出的是以一个处理器为例的电子设备。
存储器602，用于存放程序。具体地，程序可以包括程序代码，程序代码包括计算机操作指令。
存储器602可能包含高速RAM存储器，也可能还包括非易失性存储器(non-volatile memory)，例如至少一个磁盘存储器。
处理器601用于执行存储器602存储的计算机执行指令，以实现上述人群过密预测方法；其中，处理器601可能是一个中央处理器(Central Processing Unit，简称为CPU)，或者是特定集成电路(Application Specific Integrated Circuit，简称为ASIC)， 或者是被配置成实施本申请实施例的一个或多个集成电路。
可选的，在具体实现上，如果通信接口、存储器602和处理器601独立实现，则通信接口、存储器602和处理器601可以通过总线相互连接并完成相互间的通信。总线可以是工业标准体系结构(Industry Standard Architecture，简称为ISA)总线、外部设备互连(Peripheral Component，简称为PCI)总线或扩展工业标准体系结构(Extended Industry Standard Architecture，简称为EISA)总线等。总线可以分为地址总线、数据总线、控制总线等，但并不表示仅有一根总线或一种类型的总线。
可选的，在具体实现上，如果通信接口、存储器602和处理器601集成在一块芯片上实现，则通信接口、存储器602和处理器601可以通过内部接口完成通信。
本申请实施例还提供了一种人群过密预测系统，包括图像采集设备、服务器和终端设备。其中，图像采集设备用于采集待测区域的卡口的图像，服务器用于确定人群过密时间并发送给终端设备，终端设备用于接收并显示人群过密时间。
本申请实施例还提供了一种芯片，包括处理器和接口。其中接口用于输入输出处理器所处理的数据或指令。处理器用于执行以上方法实施例中提供的方法。该芯片可以应用于人群过密预测装置中。
本申请还提供了一种计算机可读存储介质，该计算机可读存储介质可以包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁盘或者光盘等各种可以存储程序代码的介质，具体的，该计算机可读存储介质中存储有程序信息，程序信息用于上述人群过密预测方法。
本申请实施例还提供一种程序，该程序在被处理器执行时用于执行以上方法实施例提供的人群过密预测方法。
本申请实施例还提供一种程序产品，例如计算机可读存储介质，该程序产品中存储有指令，当其在计算机上运行时，使得计算机执行上述方法实施例提供的人群过密预测方法。
在上述实施例中，可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行计算机程序指令时，全部或部分地产生按照本公开实施例的流程或功能。计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。计算机指令可以存储在计算机可读存储介质中，或 者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线(DSL))或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。可用介质可以是磁性介质，(例如，软盘、硬盘、磁带)、光介质(例如，DVD)、或者半导体介质(例如固态硬盘Solid State Disk(SSD))等。
最后应说明的是：以上各实施例仅用以说明本公开的技术方案，而非对其限制；尽管参照前述各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：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分或者全部技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本公开各实施例技术方案的范围。

Claims (16)

  1. 一种人群过密预测方法，应用于服务器，所述方法包括：
    获取待测区域的卡口的多帧图像；
    根据所述待测区域的卡口的所述多帧图像，确定所述待测区域的已容纳人数和所述待测区域的人流净流入速度；
    根据所述待测区域的已容纳人数、所述待测区域的可容纳人数和所述待测区域的人流净流入速度，确定所述待测区域的人群过密时间。
  2. 根据权利要求1所述的方法，其中，根据所述待测区域的卡口的所述多帧图像，确定所述待测区域的已容纳人数，包括：
    从所述待测区域的卡口的所述多帧图像中识别出进入所述待测区域的人数和离开所述待测区域的人数；
    根据进入所述待测区域的人数和离开所述待测区域的人数，确定所述待测区域的已容纳人数。
  3. 根据权利要求1所述的方法，其中，根据所述待测区域的卡口的所述多帧图像，确定所述待测区域的人流净流入速度，包括：
    从所述待测区域的卡口的所述多帧图像中识别出目标时间段内进入所述待测区域的人数和离开所述待测区域的人数；
    根据所述目标时间段内进入所述待测区域的人数和离开所述待测区域的人数，确定所述卡口的人流流入速度和所述卡口的人流流出速度；
    根据所述卡口的人流流入速度和所述卡口的人流流出速度，确定所述待测区域的人流净流入速度。
  4. 根据权利要求1所述的方法，其中，根据所述待测区域的卡口的所述多帧图像，确定所述待测区域的人流净流入速度，包括：
    从所述待测区域的卡口的所述多帧图像中识别出目标时间段内进入所述待测区域的人数和离开所述待测区域的人数；
    根据所述目标时间段内进入所述待测区域的人数和离开所述待测区域的人数，确定所述待测区域的净流入人数；
    根据所述待测区域的净流入人数，确定所述待测区域的人流净流入速度。
  5. 根据权利要求1-4任一项所述的方法，其中，根据所述待测区域的已容纳人数、所述待测区域的可容纳人数和所述待测区域的人流净流入速度，确定所述待测区域的人群过密时间，包括：
    根据所述待测区域的已容纳人数和所述待测区域的可容纳人数，确定所述待测区域的剩余可容纳人数；
    根据所述待测区域的剩余可容纳人数和所述待测区域的人流净流入速度，确定所述待测区域的人群过密时间。
  6. 根据权利要求1所述的方法，其中，在确定所述待测区域的人群过密时间之前，所述方法还包括：
    若所述待测区域的人流净流入速度为正，则确定所述待测区域存在人群过密风险。
  7. 根据权利要求1所述的方法，其中，在确定所述待测区域的人群过密时间之前，所述方法还包括：
    若所述待测区域的人流净流入速度为负，则确定所述待测区域不存在人群过密风险。
  8. 根据权利要求1所述的方法，其中，在确定所述待测区域的人群过密时间之后，所述方法还包括：
    向终端设备发送所述待测区域的人群过密时间。
  9. 根据权利要求1所述的方法，其中，若所述待测区域包括多个卡口，根据所述待测区域的卡口的所述多帧图像，确定所述待测区域的已容纳人数，包括：
    针对所述多个卡口中的每个卡口：
    从该卡口的多帧图像中识别出从该卡口进入所述待测区域的人数和从该卡口离开所述待测区域的人数；
    根据从该卡口进入所述待测区域的人数和从该卡口离开所述待测区域的人数，确定该卡口的人流净流入人数；
    将所有卡口的人流净流入人数相加，得到所述待测区域的已容纳人数。
  10. 根据权利要求1所述的方法，其中，若所述待测区域包括多个卡口，根据所述待测区域的卡口的所述多帧图像，确定所述待测区域的已容纳人数，包括：
    针对所述多个卡口中的每个卡口，从该卡口的多帧图像中识别出从该卡口进入所述待测区域的人数和从该卡口离开所述待测区域的人数；
    将从所有卡口进入所述待测区域的人数相加，得到进入所述待测区域的总人数；
    将从所有卡口离开所述待测区域的人数相加，得到离开所述待测区域的总人数；
    将进入所述待测区域的总人数减去离开所述待测区域的总人数，确定所述待测区域的已容纳人数。
  11. 根据权利要求1所述的方法，其中，若所述待测区域包括多个卡口，根据所述 待测区域的卡口的所述多帧图像，确定所述待测区域的人流净流入速度，包括：
    针对所述多个卡口中的每个卡口：
    从该卡口的多帧图像中识别出目标时间段内从该卡口进入所述待测区域的人数和从该卡口离开所述待测区域的人数；
    将从该卡口进入所述待测区域的人数除以所述目标时间段的时长，得到该卡口的人流流入速度；
    将从该卡口离开所述待测区域的人数除以所述目标时间段的时长，得到该卡口的人流流出速度；
    将所有卡口的人流流入速度相加，得到所述待测区域的人流流入速度；
    将所有卡口的人流流出速度相加，得到所述待测区域的人流流出速度；
    将所述待测区域的人流流入速度减去所述待测区域的人流流出速度，确定所述待测区域的人流净流入速度。
  12. 根据权利要求1所述的方法，其中，若所述待测区域包括多个卡口，根据所述待测区域的卡口的所述多帧图像，确定所述待测区域的人流净流入速度，包括：
    针对所述多个卡口中的每个卡口：
    从该卡口的多帧图像中识别出目标时间段从该卡口进入所述待测区域的人数和从该卡口离开所述待测区域的人数；
    将从该卡口进入所述待测区域的人数除以所述目标时间段的时长，得到该卡口的人流流入速度；
    将从该卡口离开所述待测区域的人数除以所述目标时间段的时长，得到该卡口的人流流出速度；
    将该卡口的人流流入速度减去该卡口的人流流出速度，得到该卡口的人流净流入速度；
    将所有卡口的人流净流入速度相加，确定所述待测区域的人流净流入速度。
  13. 一种人群过密预测装置，包括：
    获取模块，用于获取待测区域的卡口的多帧图像；
    第一确定模块，用于根据所述待测区域的卡口的所述多帧图像，确定所述待测区域的已容纳人数和所述待测区域的人流净流入速度；
    第二确定模块，用于根据所述待测区域的已容纳人数、所述待测区域的可容纳人数和所述待测区域的人流净流入速度，确定所述待测区域的人群过密时间。
  14. 一种电子设备，其特征在于，包括存储器与处理器；所述存储器用于存储所述 处理器的可执行指令；所述处理器配置为经由执行所述可执行指令来执行权利要求1-12任一所述的方法。
  15. 一种存储介质，其上存储有计算机程序，其特征在于，所述程序被处理器执行时实现权利要求1-12任一所述的方法。
  16. 一种计算机程序，所述计算机程序被处理器执行时，使得所述处理器执行如权利要求1至12任一项所述的方法。
PCT/CN2021/098382 2020-06-08 2021-06-04 人群过密预测方法及装置 WO2021249306A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
EP21821878.2A EP4036794A4 (en) 2020-06-08 2021-06-04 CROWD OVERDENSITY PREDICTION METHOD AND DEVICE
BR112022011625A BR112022011625A2 (pt) 2020-06-08 2021-06-04 Métodos e aparelhos de predição de superlotação
JP2022532700A JP2023503528A (ja) 2020-06-08 2021-06-04 群集過密の予測方法、装置、電子機器、記憶媒体及びコンピュータプログラム
KR1020227013875A KR20220063280A (ko) 2020-06-08 2021-06-04 군중 과밀 예측 방법 및 장치

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010510936.4A CN111652161A (zh) 2020-06-08 2020-06-08 人群过密预测方法、装置、电子设备及存储介质
CN202010510936.4 2020-06-08

Publications (1)

Publication Number Publication Date
WO2021249306A1 true WO2021249306A1 (zh) 2021-12-16

Family

ID=72344845

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/098382 WO2021249306A1 (zh) 2020-06-08 2021-06-04 人群过密预测方法及装置

Country Status (7)

Country Link
EP (1) EP4036794A4 (zh)
JP (1) JP2023503528A (zh)
KR (1) KR20220063280A (zh)
CN (1) CN111652161A (zh)
BR (1) BR112022011625A2 (zh)
TW (1) TW202211082A (zh)
WO (1) WO2021249306A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111652161A (zh) * 2020-06-08 2020-09-11 上海商汤智能科技有限公司 人群过密预测方法、装置、电子设备及存储介质
CN114900669A (zh) * 2020-10-30 2022-08-12 深圳市商汤科技有限公司 场景监测方法、装置、电子设备及存储介质
CN113344649A (zh) * 2021-08-05 2021-09-03 江西合一云数据科技有限公司 社会调查大数据构建系统
KR102594435B1 (ko) 2023-09-05 2023-10-27 비티에스 유한회사 Ai 기반 재난안전 및 방범용 영상감시시스템과 그 방법

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105791022A (zh) * 2016-04-14 2016-07-20 北京中电万联科技股份有限公司 一种拥挤度检测预警系统
US20160261984A1 (en) * 2015-03-05 2016-09-08 Telenav, Inc. Computing system with crowd mechanism and method of operation thereof
CN108345857A (zh) * 2018-02-09 2018-07-31 北京天元创新科技有限公司 一种基于深度学习的区域人群密度预测方法及装置
CN111652161A (zh) * 2020-06-08 2020-09-11 上海商汤智能科技有限公司 人群过密预测方法、装置、电子设备及存储介质

Family Cites Families (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH077080B2 (ja) * 1990-11-27 1995-01-30 技研トレーディング株式会社 平均滞留時間の測定装置
CN101325690A (zh) * 2007-06-12 2008-12-17 上海正电科技发展有限公司 监控视频流中人流分析与人群聚集过程的检测方法及系统
CN101861594B (zh) * 2007-09-19 2014-06-25 联合工艺公司 用于占用估计的系统和方法
CN100583171C (zh) * 2008-09-04 2010-01-20 上海交通大学 基于客流预测和自适应仿真的拥挤预警系统
KR101480348B1 (ko) * 2013-05-31 2015-01-09 삼성에스디에스 주식회사 사람 검출 장치 및 방법과 사람 계수 장치 및 방법
CN105095991A (zh) * 2015-07-20 2015-11-25 百度在线网络技术（北京）有限公司 用于人群风险预警的方法及装置
CN105512772B (zh) * 2015-12-22 2020-09-15 重庆邮电大学 一种基于移动网络信令数据的动态人流量预警方法
CN105763853A (zh) * 2016-04-14 2016-07-13 北京中电万联科技股份有限公司 一种公共区域拥挤、踩踏事件应急预警方法
CN106251578B (zh) * 2016-08-19 2019-05-07 深圳奇迹智慧网络有限公司 基于探针的人流预警分析方法和系统
CN107862437B (zh) * 2017-10-16 2022-02-22 中国人民公安大学 基于风险概率评估的公共区域人群聚集预警方法及系统
CN108388852B (zh) * 2018-02-09 2021-03-23 北京天元创新科技有限公司 一种基于深度学习的区域人群密度预测方法及装置
CN109034355B (zh) * 2018-07-02 2022-08-02 百度在线网络技术（北京）有限公司 致密人群的人数预测方法、装置、设备以及存储介质
CN109446989A (zh) * 2018-10-29 2019-03-08 上海七牛信息技术有限公司 人群聚集检测方法、装置及存储介质
CN109697435B (zh) * 2018-12-14 2020-10-23 重庆中科云从科技有限公司 人流量监测方法、装置、存储介质及设备
CN110414715B (zh) * 2019-06-28 2023-06-09 武汉大学 一种基于社团检测的客流量预警方法
CN110544001A (zh) * 2019-07-15 2019-12-06 中国平安财产保险股份有限公司 客流量预警方法、装置、计算机装置及存储介质
CN110705494A (zh) * 2019-10-10 2020-01-17 北京东软望海科技有限公司 人流量监测方法、装置、电子设备及计算机可读存储介质
CN110929648B (zh) * 2019-11-22 2021-03-16 广东睿盟计算机科技有限公司 监控数据处理方法、装置、计算机设备以及存储介质
CN110956122B (zh) * 2019-11-27 2022-08-02 深圳市商汤科技有限公司 图像处理方法及装置、处理器、电子设备、存储介质
CN111178276B (zh) * 2019-12-30 2024-04-02 上海商汤智能科技有限公司 图像处理方法、图像处理设备及计算机可读存储介质

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160261984A1 (en) * 2015-03-05 2016-09-08 Telenav, Inc. Computing system with crowd mechanism and method of operation thereof
CN105791022A (zh) * 2016-04-14 2016-07-20 北京中电万联科技股份有限公司 一种拥挤度检测预警系统
CN108345857A (zh) * 2018-02-09 2018-07-31 北京天元创新科技有限公司 一种基于深度学习的区域人群密度预测方法及装置
CN111652161A (zh) * 2020-06-08 2020-09-11 上海商汤智能科技有限公司 人群过密预测方法、装置、电子设备及存储介质

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4036794A4 *

Also Published As

Publication number Publication date
CN111652161A (zh) 2020-09-11
EP4036794A1 (en) 2022-08-03
KR20220063280A (ko) 2022-05-17
BR112022011625A2 (pt) 2022-12-13
TW202211082A (zh) 2022-03-16
EP4036794A4 (en) 2023-01-25
JP2023503528A (ja) 2023-01-30

Similar Documents

Publication Publication Date Title
WO2021249306A1 (zh) 人群过密预测方法及装置
CN111724522B (zh) 一种门禁控制系统、方法、装置、控制设备及存储介质
US9424464B2 (en) Monitoring system, monitoring method, monitoring program, and recording medium in which monitoring program is recorded
TW202019198A (zh) 人數統計方法、裝置及電腦設備
CN110348519A (zh) 金融产品欺诈团伙的识别方法和装置
CN113537172B (zh) 人群密度确定方法、装置、设备及存储介质
CN109302423A (zh) 一种漏洞扫描能力测试方法和装置
CN112767585A (zh) 通行检测方法及装置、电子设备及计算机可读存储介质
CN112767586A (zh) 通行检测方法及装置、电子设备及计算机可读存储介质
CN110879986A (zh) 人脸识别的方法、设备和计算机可读存储介质
CN111540020A (zh) 目标行为的确定方法及装置、存储介质、电子装置
US10176379B1 (en) Integrating computer vision and wireless data to provide identification
CN114049658A (zh) 基于人脸识别的流动人口管理方法、装置、计算机设备和存储介质
CN112530074A (zh) 排队叫号提醒方法、装置、设备及存储介质
CN106127866A (zh) 检票方法和通道管理设备
US20220327450A1 (en) Method for increasing or decreasing number of workers and inspectors in crowdsourcing-based project for creating artificial intelligence learning data
JP2020167447A (ja) 解析装置、制御方法、及びプログラム
CN114582038A (zh) 巡检管理方法及装置、电子设备及计算机可读存储介质
CN113483760A (zh) 巡更监测方法及装置、电子设备及计算机可读存储介质
CN113591713A (zh) 图像处理方法及装置、电子设备及计算机可读存储介质
CN104486512B (zh) 呼叫检测方法及装置
CN112565715A (zh) 一种景点客流量监控方法、装置、电子设备及存储介质
CN115150504A (zh) 通道控制方法、装置、设备、存储介质及系统
CN109508703A (zh) 一种视频中的人脸确定方法及装置
CN111225290B (zh) 一种疫情宣传视频数据安全处理方法及设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21821878

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 20227013875

Country of ref document: KR

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 2021821878

Country of ref document: EP

Effective date: 20220426

ENP Entry into the national phase

Ref document number: 2022532700

Country of ref document: JP

Kind code of ref document: A

REG Reference to national code

Ref country code: BR

Ref legal event code: B01A

Ref document number: 112022011625

Country of ref document: BR

ENP Entry into the national phase

Ref document number: 112022011625

Country of ref document: BR

Kind code of ref document: A2

Effective date: 20220613

NENP Non-entry into the national phase

Ref country code: DE