WO2022088653A1 - 场景监测方法、装置、电子设备、存储介质及程序 - Google Patents

场景监测方法、装置、电子设备、存储介质及程序 Download PDF

Info

Publication number
WO2022088653A1
WO2022088653A1 PCT/CN2021/094699 CN2021094699W WO2022088653A1 WO 2022088653 A1 WO2022088653 A1 WO 2022088653A1 CN 2021094699 W CN2021094699 W CN 2021094699W WO 2022088653 A1 WO2022088653 A1 WO 2022088653A1
Authority
WO
WIPO (PCT)
Prior art keywords
monitoring
people
event
point
data
Prior art date
Application number
PCT/CN2021/094699
Other languages
English (en)
French (fr)
Inventor
柴龙龙
龚超
周静
Original Assignee
深圳市商汤科技有限公司
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 深圳市商汤科技有限公司 filed Critical 深圳市商汤科技有限公司
Priority to JP2021576933A priority Critical patent/JP7305808B2/ja
Priority to KR1020217042832A priority patent/KR20220058859A/ko
Publication of WO2022088653A1 publication Critical patent/WO2022088653A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
    • G08B21/02—Alarms for ensuring the safety of persons
    • 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
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B13/00—Burglar, theft or intruder alarms
    • G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
    • G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
    • G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
    • G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
    • G08B13/19602—Image analysis to detect motion of the intruder, e.g. by frame subtraction
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources

Definitions

  • the present disclosure relates to the technical field of computer vision, and in particular, to a scene monitoring method, apparatus, electronic device, storage medium and program.
  • embodiments of the present disclosure provide at least a scene monitoring method, apparatus, electronic device, storage medium, and program.
  • An embodiment of the present disclosure provides a scene monitoring method, the method is executed by an electronic device, and the method includes:
  • the monitoring event occurs in the monitoring area corresponding to the at least one monitoring point, obtain the monitoring data of the number of people matching the monitoring event within a preset time period;
  • the people flow status data of the at least one monitoring device is determined. In this way, by obtaining the monitoring video collected by the monitoring equipment, when a monitoring event is detected in the monitoring area corresponding to at least one monitoring point based on the collected monitoring video, the monitoring data of the number of people matching the monitoring time within a preset time period is obtained, and Based on the number of people monitoring data matching the monitoring event within a preset time period, the crowd flow status data of at least one monitoring device is determined, and the determined crowd flow status data is used to characterize the status of the monitoring event, so as to monitor the monitoring video.
  • the method further includes: generating crowd flow state alarm information when the crowd flow state data satisfies an alarm condition.
  • crowd flow state alarm information is generated, and based on the generated crowd flow state alarm information, the target monitoring area can be regulated, reducing the occurrence rate of safety accidents and improving the safety of crowd flow in the target monitoring area. sex.
  • determining whether a monitoring event occurs in the monitoring area corresponding to the at least one monitoring point based on the monitoring video includes: based on the monitoring video The monitoring video determines whether there is a target object that crosses the target position matching the pre-drawn entry and exit boundary in the monitoring area corresponding to the at least one monitoring point; if the target object exists, determine the at least one monitoring point The cross-line event occurs in the monitoring area corresponding to the bit.
  • acquiring the monitoring data of the number of people matching the monitoring event within a preset time period includes: acquiring different numbers within the preset time period The number of people entering and exiting at the collection time point; wherein, the number of people entering the flow refers to the number of people who cross the pre-drawn entry and exit boundaries along the pre-drawn entry direction at the different collection time points; the number of exiting people is Refers to the number of people who cross the pre-drawn entry-exit boundary along the pre-drawn exit direction at the different collection time points.
  • the monitoring event is a cross-line event
  • the number of incoming and outgoing human flows at different collection time points within a preset time period can be obtained, which provides data support for subsequent determination of the human flow status data corresponding to the cross-line event.
  • determining the crowd flow status data of the at least one monitoring device based on the number of people monitoring data includes: Describe the number of people flow, and determine the total number of people entering and exiting within the preset time period in the monitoring area corresponding to the monitoring point; and generating the number of people flow when the state data of people flow satisfies the alarm condition.
  • Crowd flow state alarm information including: generating the crowd flow state alarm information when one of the following conditions is met: the total number of people entering the flow is greater than the set first crowd flow threshold, and the total number of people leaving the flow is greater than the set first crowd flow threshold Two crowd flow thresholds; the total number of incoming crowds is greater than the first crowd flow threshold, or the total outgoing crowd flow quantity is greater than the second crowd flow threshold.
  • the total number of inflows and the total number of inflows in the preset time period in the monitoring area corresponding to the monitoring point is determined. The number of outflows.
  • the total number of people entering the preset time period is greater than the set first threshold
  • the total number of people leaving the preset time period is greater than at least one of the set second threshold
  • generate a traffic state alarm message which can realize the early warning of the number of people entering and leaving the surveillance video, so as to facilitate the flow of people based on the generated warning information of the flow of people, and reduce the safety caused by the large number of people entering or leaving in a short time. accident rate.
  • determining the crowd flow status data of the at least one monitoring device based on the number of people monitoring data includes: The number of people flow is described, and the speed of people entering and leaving the monitoring area corresponding to the monitoring point is determined. In this way, the inflow and outflow speeds in the monitoring area corresponding to the monitoring point can be determined based on the number of inflows and outflows at different collection time points within a preset time period, so as to realize the difference between the inflow and outflow speeds. Monitor and reduce the incidence of safety accidents caused by high inflow speed or high outflow speed.
  • determining the people flow state data of the at least one monitoring device based on the number of people monitoring data includes: for each monitoring point position, based on the number of people entering the flow and the number of people leaving The historical number of people in the target monitoring area within a preset time period, and the total number of incoming and outgoing people in the preset time period corresponding to each of the monitoring points respectively, to determine the personnel in the target monitoring area Net stock; when the people flow status data satisfies the alarm condition, generating the people flow status warning information includes: in the case that the people net stock is greater than a set net stock threshold, generating the people flow status warning information.
  • the generation of people flow status alarm information can realize the early warning of the net stock of people in the target monitoring area, so that when the net stock of people is large, people can be dredged based on the generated people flow status warning information, and the safety caused by the large number of people in the target monitoring area can be reduced. accident rate.
  • the monitoring event when the monitoring event is an over-density event, determining, based on the monitoring video, whether a monitoring event occurs in the monitoring area corresponding to the at least one monitoring point, including: based on the monitoring video Monitoring video, determining whether the number of target objects in the monitoring area corresponding to the at least one monitoring point exceeds the over-density threshold; in the case that the number of the target objects exceeds the over-density threshold, determining the at least one monitoring point An over-density event occurred in the monitoring area corresponding to the point.
  • acquiring the monitoring data of the number of people matching the monitoring event within a preset time period includes: counting different numbers within the preset time period The number of the target objects at the collection time point. In this way, when the monitoring event is an over-density event, the number of the target objects at different collection time points within a preset time period can be counted, which can provide data support for subsequent determination of the crowd flow state data corresponding to the over-density event.
  • determining the people flow state data of the at least one monitoring device based on the number of people monitoring data includes: based on the number of the target objects , determining the average number of people in the monitoring area corresponding to the monitoring point within the preset time period; and generating the warning information of the people flow state when the people flow state data satisfies the alarm condition, including: in the average When the number of people is greater than the set first threshold of the number of people, the alarm information about the flow of people is generated.
  • the average number of people in the monitoring area corresponding to the monitoring point within the preset time period is determined based on the number of target objects at different collection time points within the preset time period;
  • the alarm information of the flow of people is generated, which can realize the monitoring of the average number of people in the detection area of the monitoring video, so that based on the generated
  • the crowd flow status alarm information diverts the flow of people in the detection area and reduces the occurrence rate of safety accidents caused when the people in the detection area are crowded.
  • determining the people flow state data of the at least one monitoring device based on the number of people monitoring data includes: for each monitoring point position, based on the number of the target objects, determine the average number of people in the monitoring area corresponding to the monitoring point within the preset time period; based on the average number of people, determine the total real-time number of people in the target monitoring area;
  • generating the people flow state alarm information includes: when the total real-time number of people is greater than a set second number of people threshold, generating the people flow state alarm information.
  • the total real-time number of people in the target monitoring area can be determined based on the average number of people corresponding to a plurality of the monitoring points respectively; And when it is determined that the total real-time number of people in the target monitoring area is greater than the set second threshold of the number of people, the alarm information of the flow of people is generated, which can realize the early warning of the total real-time number of people in the target monitoring area, so that when the total real-time number of people is large, Based on the generated human flow status alarm information, the personnel in the target monitoring area are diverted, and the occurrence rate of security accidents caused by a large number of real-time people in the target monitoring area is reduced.
  • the method further includes: averaging the crowd flow state data at the same collection time point in recent multiple historical dates to obtain predicted crowd flow state data corresponding to each collection time point; based on the prediction
  • the crowd flow state data constitutes prediction data of the crowd flow state data in the future; wherein the prediction data is used to generate a crowd flow diversion plan.
  • a crowd flow diversion plan can be generated based on the forecast data of the crowd flow state data in the future.
  • the embodiment of the present disclosure also provides a scene monitoring device, including:
  • a first acquisition module configured to acquire monitoring video collected by monitoring equipment set at at least one monitoring point
  • a detection module configured to determine whether a monitoring event occurs in the monitoring area corresponding to the at least one monitoring point based on the monitoring video
  • a second acquisition module configured to acquire the monitoring data of the number of people matching the monitoring event within a preset time period when the monitoring event occurs in the monitoring area corresponding to the at least one monitoring point;
  • a determination module configured to determine the flow state data of the at least one monitoring device based on the number of people monitoring data.
  • the scene monitoring apparatus after determining the crowd flow state data of the at least one monitoring device, the scene monitoring apparatus further includes: an alarm module, configured to: when the crowd flow state data satisfies an alarm condition, Generates crowd flow status alarm information.
  • the detection module when the monitoring event is a cross-line event, is configured to determine, based on the monitoring video, whether there is a cross-connection in the monitoring area corresponding to the at least one monitoring point, based on the monitoring video.
  • the pre-drawn target object at the target position matched with the entry and exit boundary; in the case of the existence of the target object, it is determined that the cross-line event occurs in the monitoring area corresponding to the at least one monitoring point.
  • the second acquisition module is configured to acquire the number of incoming and outgoing traffic at different collection time points within the preset time period;
  • the number of people entering and exiting refers to the number of people who cross the pre-drawn entry-exit boundary along the pre-drawn entry direction at the different collection time points; The number of people whose exit direction crosses the pre-drawn entry and exit boundaries.
  • the determining module is configured to determine, based on the number of inflows and the number of outflows, in the monitoring area corresponding to the monitoring point The total number of people entering and exiting within the preset time period;
  • an alarm module configured to generate the crowd flow state alarm information when one of the following conditions is met: the total number of people entering the crowd is greater than the set first crowd flow threshold, and the total number of people leaving the crowd is greater than the set second crowd flow Threshold; the total number of people entering is greater than the first threshold, or the total number of people leaving is greater than the second threshold.
  • the determining module is configured to determine, based on the number of inflows and the number of outflows, in the monitoring area corresponding to the monitoring point The speed of inflow and outflow of people.
  • the determining module is configured to, for each monitoring point, determine based on the number of inflows and the number of outflows The total number of people entering and leaving the preset time period in the monitoring area corresponding to each of the monitoring points; based on the historical number of people in the target monitoring area within the preset time period, and each The total number of incoming and outgoing people in the preset time period corresponding to the monitoring points respectively, to determine the net stock of personnel in the target monitoring area;
  • the alarm module is configured to generate the alarm information of the people flow state when the net personnel stock is greater than the set net stock threshold.
  • the detection module when the monitoring event is an over-density event, is configured to, based on the monitoring video, determine the number of target objects in the monitoring area corresponding to the at least one monitoring point Whether the number exceeds the over-density threshold; in the case that the number of the target objects exceeds the over-density threshold, it is determined that an over-density event occurs in the monitoring area corresponding to the at least one monitoring point.
  • the second acquisition module is configured to count the number of the target objects at different acquisition time points within the preset time period.
  • the determining module is configured to, based on the number of the target objects, determine that the monitoring area corresponding to the monitoring point is within the preset the average number of people in the time period;
  • An alarm module configured to generate the alarm information of the crowd flow state when the average number of people is greater than the set first threshold of the number of people.
  • a determining module is configured to, for each monitoring point, determine the monitoring point based on the number of the target objects The average number of people in the corresponding monitoring area within the preset time period; based on the average number of people, determine the total real-time number of people in the target monitoring area;
  • the alarm module is configured to generate the alarm information of the crowd flow state when the total real-time number of people is greater than the set second threshold of the number of people.
  • the scene detection device further includes: an early warning module, configured to average the crowd flow status data at the same collection time point in recent multiple historical dates to obtain the predicted crowd flow corresponding to each collection time point state data; based on the predicted crowd flow state data, the prediction data of the crowd flow state data in the future date is formed; wherein, the predicted data is used to generate a crowd flow diversion plan.
  • an early warning module configured to average the crowd flow status data at the same collection time point in recent multiple historical dates to obtain the predicted crowd flow corresponding to each collection time point state data; based on the predicted crowd flow state data, the prediction data of the crowd flow state data in the future date is formed; wherein, the predicted data is used to generate a crowd flow diversion plan.
  • Embodiments of the present disclosure further provide an electronic device, including: a processor, a memory, and a bus, where the memory stores machine-readable instructions executable by the processor, and when the electronic device runs, the processor communicates with The memories communicate with each other through a bus, and when the machine-readable instructions are executed by the processor, the scene monitoring method according to any one of the foregoing embodiments is executed.
  • Embodiments of the present disclosure further provide a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is run by a processor, the scene monitoring method described in any of the foregoing embodiments is executed.
  • Embodiments of the present disclosure further provide a computer program, where the computer program includes computer-readable codes, and when the computer-readable codes are executed in an electronic device, the processor of the electronic device executes any of the foregoing implementations The scene monitoring method described in the example.
  • the embodiments of the present disclosure provide at least one scene monitoring method, device, electronic device, storage medium, and program.
  • acquiring the monitoring video collected by the monitoring equipment based on the collected monitoring video, it is detected that at least one monitoring point occurs in the monitoring area corresponding to the monitoring point.
  • obtain the number of people monitoring data that matches the monitoring time within a preset time period and based on the number of people monitoring data, determine the crowd flow status data of at least one monitoring device, and use the determined crowd flow status data to characterize the status of the monitoring event to achieve monitoring.
  • Video surveillance By acquiring the monitoring video collected by the monitoring equipment, based on the collected monitoring video, it is detected that at least one monitoring point occurs in the monitoring area corresponding to the monitoring point.
  • FIG. 1 shows a schematic flowchart of a scene monitoring method provided by an embodiment of the present disclosure
  • FIG. 2 shows a schematic diagram of a system architecture to which the scene monitoring method according to the embodiment of the present disclosure can be applied;
  • Fig. 3A shows a kind of interface schematic diagram showing the interface schematic diagram of the video screen screenshot drawn with monitoring identification provided by the embodiment of the present disclosure
  • 3B is a schematic interface diagram showing another interface schematic diagram of a video screen screenshot drawn with a monitoring logo provided by an embodiment of the present disclosure
  • 3C shows a schematic interface diagram of early warning information corresponding to human flow data shown in an embodiment of the present disclosure
  • FIG. 4 shows a schematic diagram of an interface for displaying detailed information of a crowd flow state alarm provided by an embodiment of the present disclosure
  • 5A shows a schematic diagram of an interface for displaying detailed information of a crowd flow state alarm provided by an embodiment of the present disclosure
  • FIG. 5B is a schematic diagram of another interface for displaying detailed information of a crowd flow state alarm provided by an embodiment of the present disclosure
  • FIG. 5C shows a schematic diagram of an interface for displaying alarm details provided by an embodiment of the present disclosure
  • FIG. 6 shows a schematic diagram of an interface for displaying detailed information of a crowd flow state alarm provided by an embodiment of the present disclosure
  • FIG. 7A shows a schematic diagram of an interface for displaying detailed information of a crowd flow state alarm provided by an embodiment of the present disclosure
  • FIG. 7B shows a schematic diagram of an interface for displaying alarm details provided by an embodiment of the present disclosure
  • FIG. 8A shows a schematic interface diagram of changes in the real-time total number of people and the total stock of people flow over time provided by an embodiment of the present disclosure
  • FIG. 8B shows a schematic diagram of an interface for predicting human flow data in a future time period provided by an embodiment of the present disclosure
  • FIG. 9 shows a schematic structural diagram of a scene monitoring apparatus 900 provided by an embodiment of the present disclosure.
  • FIG. 10 shows a schematic structural diagram of an electronic device 1900 provided by an embodiment of the present disclosure.
  • Crowd analysis has important applications in many fields of life. It can timely analyze the real-time number of people in the area, reduce major events such as stampede, and ensure people's safety. It can also be used in business scenarios to help merchants analyze customer behavior.
  • the existing crowd analysis-related applications are more concerned with estimating the number of people in the video area or performing early warning of regional intrusions.
  • This mechanism can only be viewed and processed for real-time situations, and cannot be predicted to prevent security problems, nor can it be Perform data analysis to summarize the flow patterns.
  • the embodiments of the present disclosure provide a scene monitoring method, apparatus, electronic device, storage medium and program.
  • the execution subject of the scene monitoring method provided by the embodiment of the present disclosure is generally a computer device with a certain computing capability. , mobile devices, user terminals, terminals, cellular phones, cordless phones, Personal Digital Assistant (PDA), handheld devices, computing devices, in-vehicle devices, wearable devices, etc.
  • the scene monitoring method may be implemented by a processor invoking computer-readable instructions stored in a memory.
  • the method includes S101 to S104, and specifically includes:
  • S101 acquiring surveillance video collected by a surveillance device set at at least one surveillance point.
  • S102 based on the monitoring video, determine whether a monitoring event occurs in a monitoring area corresponding to at least one monitoring point.
  • the monitoring data of the number of people matching the monitoring time within a preset time period is obtained.
  • determine the crowd flow status data of at least one monitoring device and use the determined crowd flow status data to represent the status of the monitoring event to monitor the surveillance video.
  • the crowd flow status data can be the total number of people entering the flow. When the number of people entering the flow is large, the occurrence of monitoring events is more frequent.
  • FIG. 2 shows a schematic diagram of a system architecture to which a scene monitoring method according to an embodiment of the present disclosure can be applied; as shown in FIG.
  • the monitoring video acquisition terminal 201 and the scene monitoring terminal 203 establish a communication connection through the network 202
  • the monitoring video acquisition terminal 201 reports the monitoring video to the scene monitoring terminal 203 through the network 202
  • the scene monitoring terminal 203 responds to the monitoring video.
  • determine whether a monitoring event occurs in the monitoring area corresponding to at least one monitoring point based on the monitoring video and in the case where a monitoring event occurs in the monitoring area corresponding to at least one monitoring point, obtain the number of people who match the monitoring event within a preset time period.
  • the scene monitoring terminal 203 uploads the people flow state data of the monitoring video to the network 202 , and sends the data to the monitoring video obtaining terminal 201 through the network 202 .
  • the surveillance video acquisition terminal 201 may include a video capture device, and the scene monitoring terminal 203 may include a visual processing device or a remote server with visual information processing capability.
  • Network 202 may employ wired or wireless connections.
  • the monitoring video acquisition terminal 201 can communicate with the visual processing device through a wired connection, such as data communication through a bus; when the scene monitoring terminal 203 is a remote server, the monitoring video The acquisition terminal 201 can perform data interaction with a remote server through a wireless network.
  • the surveillance video acquisition terminal 201 may be a vision processing device with a video acquisition module, and is specifically implemented as a host with a camera.
  • the scene monitoring method of the embodiment of the present disclosure may be executed by the monitoring video acquisition terminal 201 , and the above-mentioned system architecture may not include the network 202 and the scene monitoring terminal 203 .
  • the method can be used to detect a target monitoring area, and the target monitoring area can be any area in a real scene, for example, the target monitoring area can be shopping malls, beaches, parks, and subway stations Wait.
  • multiple monitoring points may be set at the target monitoring area, and a monitoring device may be installed at each monitoring point, so that the monitoring device can monitor the corresponding monitoring area and monitor the target monitoring area.
  • the setting of monitoring points can be determined according to actual needs. For example, when the target monitoring area is a shopping mall, one monitoring point can be set at each door of the shopping mall, and one monitoring point can be set at at least one of each elevator entrance. point etc.
  • the monitoring device may be a device such as a monitoring camera.
  • the monitoring video in the corresponding monitoring area can be collected by the monitoring device, so that the monitoring video collected by each monitoring device can be obtained, that is, the monitoring equipment set at at least one monitoring point can be obtained. surveillance video.
  • the monitoring event can include at least one of an over-density event and a cross-line event; an over-density event means that the density of people in the area is greater than the set value, that is, the density of people in the area is relatively large; cross-line means that there are pedestrians crossing the area in the area. set baseline.
  • a reference line can be set on the platform at a preset distance (for example, 1 meter) from the subway to monitor Whether someone crossed the baseline (that is, whether someone crossed the baseline into or out of the subway), and if so, a crossing event occurred.
  • a target monitoring area can be set on the beach, and when the number of people in the target monitoring area is greater than the set number of people, it is determined that an over-density event has occurred.
  • a function button may also be set for each monitoring video, and by triggering the function button, it is determined to monitor the monitoring event in the monitoring area of the monitoring point.
  • the first function button corresponding to the cross-line event can be set (monitoring the cross-line event for a single surveillance video). After the first function button corresponding to the monitoring video A is triggered, it is determined to perform the cross-line event on the monitoring video A.
  • the second function button corresponding to the cross-line event can also be set (to monitor the cross-line event for the monitoring video group), and the second function corresponding to the monitoring video group A formed by monitoring video A, monitoring video B, etc. After the button is triggered, it is determined to monitor the cross-line event of the monitoring video group A.
  • a third function button corresponding to the over-confidence event may also be set (monitoring of the over-encryption event for a single surveillance video), after the third function button corresponding to the surveillance video A is triggered, it is determined that Monitor surveillance video A for over-confidence events; alternatively, you can also set a fourth function button corresponding to over-dense events (monitoring over-dense events for surveillance video groups), and monitor the surveillance video A, surveillance video B, etc. After the fourth function button corresponding to the video group A is triggered, it is determined that the monitoring video group A is monitored for an over-densification event.
  • a monitoring identification corresponding to the monitoring video may be drawn.
  • the monitoring mark may be a pre-drawn entry and exit boundary, inbound and outbound directions; for over-density events, the monitoring mark may be any pre-drawn polygon, or, for over-density events, no corresponding monitoring identification.
  • the monitoring identifiers corresponding to different surveillance videos are different, that is, for each surveillance video, a corresponding monitoring identifier can be drawn for the surveillance video (at least one of the monitoring identifier corresponding to the cross-line event and the monitoring identifier corresponding to the over-density event).
  • a frame of video screenshot can be collected from the surveillance video, and the video screenshot can be displayed, so that the user can draw a monitoring logo on the video screenshot according to actual needs.
  • the position information of the monitoring mark in the video screen shot in the video screen shot can be determined by acquiring the video screen shot that is pre-drawn with the monitoring mark, wherein the position information can be that the monitoring mark is in the pixel coordinate system corresponding to the video screen shot.
  • the set of coordinates for example, can be the position information of the entry and exit boundaries, etc.
  • the target location information matching the monitoring identification can be determined in the video picture of the monitoring video.
  • the location information of the monitoring logo in the screenshot of the video screen may be the target location information of the monitoring logo in the video screen of the monitoring video. Furthermore, it can be determined whether a monitoring event occurs in the monitoring area corresponding to the monitoring point based on the monitoring video collected for the monitoring device and the determined target location information.
  • the monitoring identification when the monitoring identification includes a monitoring identification corresponding to a cross-line event, see FIG. 3A , a schematic interface diagram showing a screenshot of a video screen drawn with a monitoring identification.
  • the monitoring mark 31 includes a drawn entry and exit boundary line and an arrow mark indicating the entry and exit direction.
  • an entry threshold that is, the first threshold
  • an exit threshold that is, the second threshold
  • At least one of the surveillance videos is monitored.
  • 3A also includes prompt information set on the line crossing event located above the screenshot of the video screen, so that the user can draw a monitoring mark according to the displayed prompt information set on the line crossing event.
  • the monitoring identification when the monitoring identification includes a monitoring identification corresponding to an over-density event, refer to another interface schematic diagram shown in FIG. 3B in which a screenshot of a video screen with a monitoring identification is drawn.
  • the monitoring mark 32 includes a polygon indicating a detection area, wherein the number of the detection area may be multiple.
  • you can also set the number of people for early warning by grades on the displayed interface, that is, the number of early warning people corresponding to general risks, the number of early warning people corresponding to major risks, and the number of early warning people corresponding to major risks.
  • Surveillance video for monitoring This FIG.
  • 3B also includes prompt information set on the overcrowding event located above the screenshot of the video screen, so that the user can draw a monitoring mark indicating the detection area according to the displayed prompt information set on the line crossing event.
  • the drawn monitoring logo can be stored in the multiplexing area, so that when the monitoring logo is determined next time, the function button of the multiplexing area can be directly triggered to realize the reuse of the monitoring logo.
  • the function button of human body annotation in FIG. 3B is used to display the setting information of human body annotation.
  • the size of the human body in the surveillance video screen is related to the height and angle of the monitoring equipment, and the distance between the same human body and the monitoring equipment is different, the size of the area in the monitoring video screen is different, that is, when the distance from the monitoring equipment is relatively close.
  • the area of the human body is large, so human body labeling is the basic setting for cross-line events and over-density events.
  • the area and location of each pedestrian's body frame may be estimated from the body frames of a plurality of pedestrians marked at different depth positions. It is convenient for algorithms (for example, image recognition algorithms for human body recognition) to use human body annotation results to identify different monitoring devices in different situations and improve the recognition accuracy. The more human body frames, the higher the accuracy.
  • the number of marked pedestrian frames can be set as required, for example, the number of marked pedestrian frames can be set to range from 3 to 10.
  • the area of the human body frame of multiple pedestrians and the depth information of each pedestrian can be used to detect the real-time number of people included in the detection area of the video frame per second in the surveillance video.
  • the predicted area of the drawn area in the real scene can be calculated based on the human body samples marked in the human body annotation, and the predicted area of the drawn area in the real scene can be calculated and displayed in the "area area estimation" at the bottom of Figure 3B. ”, the predicted area is displayed, and the density of people in the detection area can be calculated in the subsequent alarms of over-density of points and over-density of video groups, etc.
  • the FIG. 3B also includes a “correct area” function button, after triggering the “correct area” function button, the predicted area displayed at the area area estimate can be corrected.
  • FIG. 3C when the monitoring identification of the video screen includes both the over-density event and the monitoring identification corresponding to the cross-line time, see FIG. 3C for an early warning corresponding to the flow of people data shown in the embodiment of the disclosure.
  • Figure 3C includes the warning level and the number of people corresponding to the overcrowded event and the cross-line event, respectively.
  • Figure 3C also includes two buttons to activate the valid time of the overcrowded event and the cross-line event on the video screen, namely "long-term valid" and "custom", so that the user can flexibly set relevant monitoring parameters and time information.
  • determining whether a monitoring event occurs in a monitoring area corresponding to at least one monitoring point based on the monitoring video includes: determining, based on the monitoring video, corresponding to at least one monitoring point Whether there is a target object that crosses the target position matching the pre-drawn entry and exit boundary in the monitoring area of ?
  • the monitoring event when the monitoring event is a cross-line event, for the monitoring video collected at each monitoring point, it may be determined, based on the monitoring video, whether there is a cross-over or entry or exit in the monitoring area corresponding to the monitoring point.
  • the target object at the target location that matches the boundary line such as detecting whether a pedestrian crosses the drawn entry and exit boundary line in the surveillance video. No cross-line event occurred in the monitoring area corresponding to the monitoring point.
  • the monitoring area corresponding to the monitoring point can be the detection area that can be monitored by the monitoring equipment set at the monitoring point; the monitoring area corresponding to the monitoring point is related to the installation position and installation angle of the monitoring equipment. The angles correspond to different monitoring areas.
  • the monitoring area corresponding to at least one monitoring point determined based on the monitoring video when there is a target object that crosses the target position matching the entry and exit boundary, it is determined that the monitoring area corresponding to the at least one monitoring point occurs.
  • Cross-line events can realize real-time monitoring of cross-line events and improve the accuracy of cross-line event monitoring.
  • determining whether a monitoring event occurs in a monitoring area corresponding to at least one monitoring point based on the monitoring video includes: determining, based on the monitoring video, at least one monitoring point Whether the number of target objects in the corresponding monitoring area exceeds the over-density threshold; if the number of target objects exceeds the over-density threshold, it is determined that an over-density event occurs in the monitoring area corresponding to at least one monitoring point.
  • the monitoring event when the monitoring event is an over-density event, for the monitoring video collected at each monitoring point, it may be determined, based on the monitoring video, whether the number of target objects in the monitoring area corresponding to the monitoring point exceeds Over-density threshold, for example, determine the number of humans in the monitoring area in the surveillance video, and determine whether the number of humans is greater than the preset over-density threshold. If so, determine that an over-density event has occurred in the monitoring area corresponding to the monitoring point; Then it is determined that the monitoring area corresponding to the monitoring point does not have an over-density event.
  • Over-density threshold for example, determine the number of humans in the monitoring area in the surveillance video, and determine whether the number of humans is greater than the preset over-density threshold. If so, determine that an over-density event has occurred in the monitoring area corresponding to the monitoring point; Then it is determined that the monitoring area corresponding to the monitoring point does not have an over-density event.
  • the monitoring area corresponding to the monitoring point may be the detection area matching the drawn polygon; when the monitoring logo is not drawn, the monitoring area corresponding to the monitoring point is the monitoring point.
  • the set detection area that can be monitored by the monitoring device that is, the area corresponding to the monitoring interface of the monitoring video is the monitoring area).
  • the monitoring data of the number of people matching the monitoring event within a preset time period can be obtained; the monitoring data of the number of people includes the data corresponding to the cross-line event At least one of the population monitoring data and the population monitoring data corresponding to the overcrowding event. Further, based on the number of people monitoring data matching the monitoring event within a preset time period, determine the crowd flow state data of at least one monitoring device; the crowd flow state data includes at least one of the crowd flow state data corresponding to the cross-line event and the crowd flow state data corresponding to the over-density event. one.
  • the preset time period can be set as required.
  • the preset time period can be the time period from the time when the monitoring event is determined to occur to one hour later. If the time when the monitoring event is determined to occur is 13:10:00 , the preset time period is a time period from 13:10:00 to 14:10:00. For example, the preset time period may be the time period from the time when the monitoring event is determined to occur to one minute later. If it is determined that the time when the monitoring event occurs is 13:10:00, the preset time period is from 13:10 The time period from 00 seconds to 13:11:00.
  • the monitoring data of the number of people matching the cross-line event within a preset time period can be obtained; Matching traffic status data.
  • the monitoring data of the number of people matching the over-density events within a preset time period can be obtained; and based on the number of people monitoring data matching the over-density events within the preset time period, determine at least one monitoring device that matches the over-density events. Matching traffic status data.
  • the method further includes: generating crowd flow state alarm information when the crowd flow state data satisfies an alarm condition.
  • the crowd flow status alarm information is generated, so that the user can generate a grooming plan based on the crowd flow status alarm information to reduce the occurrence of stampede, congestion and other events in the target monitoring area. probability.
  • crowd flow status alarm information is generated, and based on the generated crowd flow status alarm information, the target monitoring area can be regulated to reduce the occurrence rate of safety accidents and improve the safety of crowd flow in the target monitoring area. sex.
  • obtaining the monitoring data of the number of people matching the monitoring event within the preset time period includes: obtaining the number of incoming and outgoing traffic at different collection time points within the preset time period, wherein the incoming and outgoing traffic
  • the number of people flow refers to the number of people who cross the pre-drawn entry-exit boundary along the pre-drawn incoming direction at different collection time points; .
  • the monitoring identification corresponding to the cross-line event may include a preset entry and exit boundary and an entry and exit direction (at least one of the entry and exit directions, and the exit direction is the opposite direction of the entry direction).
  • the inbound direction in the inbound and outbound direction may be the direction from the outbound area into the inbound area
  • the outbound direction in the inbound and outbound direction may be the direction from the inbound area into the outbound area.
  • the number of incoming and outgoing traffic (that is, the number of incoming traffic) and outgoing traffic at each collection time point in the preset time period in the surveillance video can be determined, and the incoming traffic at different collection time points.
  • Quantity refers to the number of people who cross the inbound and outbound boundary along the incoming direction at different collection time points;
  • a trained target tracking algorithm can be used to detect surveillance video based on a set surveillance identifier, and within a preset time period, output a detection result every preset time interval, and the preset time period
  • the multiple detection results in this can be the number of inflows and outflows at different collection time points within a preset time period, and each detection result is associated with an output time (the output time is the collection time point), and then the preset time can be obtained. The number of incoming and outgoing traffic at different collection time points in the segment.
  • the monitoring event is a cross-line event
  • the number of incoming and outgoing human flows at different collection time points within a preset time period can be obtained, which can provide data support for subsequent determination of the human flow state data corresponding to the cross-line event.
  • determining the crowd flow status data of at least one monitoring device includes: determining the corresponding monitoring point based on the number of incoming and outgoing crowds. The total number of incoming and outgoing people in the monitoring area of the preset time period.
  • generating crowd flow status alarm information including: generating crowd flow status alarm information when one of the following conditions is met: the total number of people entering the flow is greater than the set first crowd flow threshold, and the total number of people leaving The number of people flow is greater than the set second flow threshold; the total number of people entering is greater than the first set threshold, or the total number of people leaving is greater than the second set threshold.
  • the total number of inflows and outflows in the preset time period in the monitoring area corresponding to the monitoring point can be determined based on the number of inflows and outflows at different collection time points within the preset time period.
  • the trained target tracking algorithm can output a detection result every 3 seconds (a collection time point is determined every 3 seconds), and the detection result can be the number of people entering and leaving the flow within the 3 seconds,
  • the detection result can be: the number of incoming traffic between 08:10:01 and 08:10:03 (including 10:01 and 10:03) is 20, the number of outgoing traffic is 50, and the associated output
  • the time (collection time point) is 08:10:03; then multiple detection results within a preset time period can be obtained, that is, the number of incoming and outgoing traffic at different collection time points within the preset time period.
  • the number of inbound flows at different collection time points can be added to obtain the total number of inflows in the preset time period; and Add up the number of outbound people at different collection time points to obtain the total number of outbound people in the preset time period.
  • the first crowd flow threshold and the second crowd flow threshold are preset, and the first crowd flow threshold and the second crowd flow threshold may be set according to actual needs. After obtaining the total number of people entering and exiting within the preset time period, it can be determined whether the total number of people entering the preset time period is greater than the set first threshold of people flow, and/or, the preset time can be judged. Whether the total number of people leaving the segment is greater than the set second threshold.
  • the generated crowd status alarm information may be information in the format of text, voice, video, etc.
  • the generated crowd status alarm information may be "Attention, the number of people entering the flow is large”.
  • the alarm event type of the crowd flow state alarm information is: point crossing line alarm.
  • detailed information of the crowd flow state alarm may be displayed, including but not limited to the alarm point (namely, the name of the alarm monitoring device, etc.) Alarm time and alarm event type.
  • the alarm event type is a point crossing alarm
  • the detailed information also includes the number of incoming and outgoing traffic in the unit time.
  • the total number of inflows and the total number of inflows in the preset time period in the monitoring area corresponding to the monitoring point is determined.
  • determining the crowd flow status data of at least one monitoring device includes: determining the corresponding monitoring point based on the number of incoming and outgoing crowds. The speed of inflow and outflow of people in the monitoring area.
  • the inflow speed and the outflow speed in the monitoring area corresponding to the monitoring point may also be determined based on the number of inflows and outflows at different collection time points within a preset time period.
  • the multiple detection results can be classified according to the output time (collection time point), and Integrate, get the number of incoming and outgoing traffic in a unit time (such as one minute), and get the speed of incoming and outgoing traffic.
  • the output results from 08:10:00 to 08:11:00 can be classified and integrated, i.e.
  • the output time is 08:10:03, 08:10:06,... Carry out integration to obtain the number of inflows and outflows within 1 minute (unit time) between 08:10:00 and 08:11:00, that is, the corresponding inflow speed at 08:10 ( Unit: person/minute) and outflow speed (unit: person/minute).
  • the inflow speed and the outflow speed in the monitoring area corresponding to the monitoring point can be determined based on the number of inflows and outflows at different collection time points within a preset time period, so as to realize the control of the inflow and outflow.
  • Speed monitoring can reduce the incidence of safety accidents caused by high inflow speed or high outflow speed.
  • FIG. 4 includes the alarm details and the statistics of the cross-line event period of the day.
  • the alarm details include the alarm point, event type, alarm time, and duration ( The cross-line event duration), inflow peak, outflow peak, etc.
  • the current cross-line event period statistics include cross-line event alarms from the zero point of the day to the current time of the statistics.
  • the figure also includes a screenshot of the video screen, and the screenshot of the video screen displays the information of the outgoing flow (the number of outgoing flows and the speed of the outgoing flow) and the information of the incoming flow (the number of the incoming flow and the speed of the incoming flow) corresponding to the current moment; Displays multiple frames of alarm pictures, in which the number of alarm pictures is related to the duration of the alarm. For example, when the duration of the cross-line event is 17 minutes, one frame of alarm pictures can be extracted every one minute as an alarm record, that is, 17 frames of alarm pictures are displayed below the video screenshot.
  • information such as the name of the monitoring device, the installation location, the point information such as the collected monitoring video, and the number of outgoing traffic and incoming traffic per unit time can also be persistently stored in the search server (for example, elasticsearch) for subsequent search queries.
  • search server For example, elasticsearch
  • determining the flow state data of at least one monitoring device includes:
  • Step 1 For each monitoring point, based on the number of inflows and outflows, determine the total number of inflows and total outflows in the preset time period in the monitoring area corresponding to each monitoring point.
  • Step 2 Determine the net stock of personnel in the target monitoring area based on the historical number of people in the target monitoring area within the preset time period, and the total number of incoming and outgoing people in the preset time period corresponding to each monitoring point. .
  • generating the people flow state alarm information includes: when the net personnel stock is greater than the set net stock threshold, generating the people flow status warning information.
  • the multiple monitoring devices may be set up in a site or place
  • people flow analysis can be performed on the surveillance videos collected by the multiple monitoring devices respectively, so as to obtain the crowd flow status data of the multiple surveillance videos.
  • the monitoring videos collected by multiple monitoring devices respectively constitute a video group, that is, the crowd flow analysis can be performed on the video group to obtain the crowd flow status data corresponding to the video group.
  • the cross-line analysis function of each surveillance video in the video group can be enabled by triggering the open button of the cross-line event corresponding to the video group set on the display interface.
  • the different collection time points may be The total number of inflows in the preset time period corresponding to the monitoring point is obtained by adding up the number of inflows of the monitoring point; and the number of outflows at different collection time points can be added to obtain the preset time period corresponding to the monitoring point.
  • the total number of inflows and the total number of outflows in preset time periods corresponding to each monitoring point can be obtained.
  • the total number of incoming and outgoing people from the time point from 08:11:00 to 08:12:00 corresponding to each monitoring device can be obtained, and the total number of outgoing people can be obtained at 08:00.
  • the time period between the time point of 11 minutes 00 seconds and the time point of 08:12 minutes 00 seconds is the preset time period.
  • the target monitoring area is determined based on the historical number of people in the target monitoring area within a preset time period, and the total number of incoming and outgoing people in the preset time period corresponding to each monitoring point. The net stock of people in the area. For example, for each surveillance video in a video group, the total number of incoming traffic and the total number of outgoing traffic in the preset time period corresponding to the surveillance video can be subtracted to obtain the traffic of the surveillance video in the preset time period.
  • the change amount add up the change amount of people flow in the preset time period corresponding to each surveillance video to obtain the change amount of total people flow corresponding to the video group (that is, the total change amount of people flow corresponding to the venue or venue corresponding to the video group), and then add The total change of people flow corresponding to the video group is added to the historical number of people in the target monitoring area within the preset time period to obtain the net stock of personnel in the target monitoring area (that is, the current number of people at the current time point corresponding to the venue or venue corresponding to the video group is obtained. ).
  • the preset time period may be the time period between the time point of 08:11:00 seconds and the time point of 08:12:00 seconds, and then the time period of 08:11:00 seconds may be obtained.
  • the current number of people corresponding to the time point that is, the historical number of people in the target monitoring area within the preset time period
  • the corresponding 08:11:00 to 08:12:00 (preset time) of each surveillance video in the video group can be obtained.
  • the total number of incoming and outgoing traffic in the segment based on the historical number of people in the target monitoring area within the preset time period (that is, the obtained net stock of personnel at the time point of 08:11:00), and within the video group The total number of people entering and exiting from 08:11:00 to 08:12:00 corresponding to each surveillance video, to determine the net stock of people in the target monitoring area at 08:12:00.
  • the generated crowd flow state alarm information may be "Attention, there are a lot of net people in the xx venue at the current time”.
  • the alarm event type of the crowd status alarm information is: video group cross-line alarm.
  • multi-level alarm risks can be set for cross-line alarms of video groups.
  • multi-level alarm risks include: increasing inventory, inventory warning, and inventory overheating.
  • Different net alarms can be set for different alarm risks.
  • the inventory threshold for example, the net inventory threshold corresponding to the increasing inventory can be 100, the net inventory threshold corresponding to the inventory warning can be 200, and the net inventory threshold corresponding to the inventory overheating can be 500.
  • the warning information of the crowd status corresponding to the increasing stock may be: alarm information in text format; the warning information of the crowd status corresponding to the stock warning may be: alarm information in voice format; the warning information of the crowd status corresponding to the overheating stock may be: video format alarm information.
  • detailed information of the crowd flow state alarm may be displayed, including but not limited to the alarm point (namely, the name of the alarm monitoring device, etc.) Alarm time and alarm event type.
  • the alarm time type is video group cross-line alarm
  • the detailed information may also include: the net personnel stock at the current time point.
  • the generation of people flow status alarm information can realize the early warning of the net stock of people in the target monitoring area, so that when the net stock of people is large, people can be dredged based on the generated people flow status warning information, and the safety caused by the large number of people in the target monitoring area can be reduced. accident rate.
  • FIG. 5A a schematic interface diagram showing the detailed information of the human flow state alarm shown in FIG. 5A
  • the detailed information of the human flow state alarm is displayed in a map mode in FIG. 5A
  • FIG. 5B shows the detailed information of the crowd flow state alarm in a list mode, wherein the list shown in FIG. 5B includes over-density events and line crossing events.
  • the alarm details shown in FIG. 5C may be displayed, and FIG.
  • the alarm details displayed in 5C include group name (namely the name corresponding to the video group), event type, alarm time, duration, peak value of the total stock of people flow, and statistics of the total stock of people flow on the day.
  • acquiring the monitoring data of the number of people matching the monitoring event within a preset time period includes: counting target objects at different collection time points within the preset time period number of.
  • the target object (human) in the detection area corresponding to the monitoring identifier can be detected based on the monitoring identifier and the surveillance video corresponding to the over-density event, and each collection time can be obtained.
  • the number of target objects in the detection area when you click.
  • the entire surveillance image of the surveillance video is a detection area, and the surveillance video can be detected to obtain the number of target objects in the detection area at each collection time point.
  • a trained deep learning algorithm for recognizing target objects can be used to detect the detection area in the surveillance video, and the detection result can be output in real time, and the detection result can be each acquisition in the surveillance video The number of people in the detection area at the time point.
  • the deep learning algorithm may output detection results periodically, for example, the deep learning algorithm may output detection results every second, or may output detection results every two seconds, and so on.
  • the detection result may be: the number of people in the detection area at 08:10:00 (the collection time point) is 50; the number of people in the detection area at 08:10:01 is 54, and so on.
  • the number of target objects at different collection time points within a preset time period may be counted.
  • the preset time period is from 08:10:00 to 08:11:00.
  • Set the time points per second in the time point as a collection time point that is, the number of target objects at 08:10:00 (collection time point 1) and the number of target objects at 08:10:01 (collection time point 2) can be counted.
  • the monitoring event is an over-density event
  • the number of target objects at different collection time points within a preset time period can be counted, which can provide data support for subsequent determination of the crowd flow state data corresponding to the over-density event.
  • determining the flow state data of at least one monitoring device based on the number of people monitoring data includes: determining the monitoring point corresponding to the monitoring point based on the number of target objects The average number of people in the area over a preset time period.
  • generating crowd flow state alarm information includes: generating crowd flow state alarm information when the average number of people is greater than the set first threshold of the number of people.
  • the number of target objects at different collection time points within the preset time period may be averaged to obtain the average number of people in the monitoring area corresponding to the monitoring point within the preset time period.
  • the average number of people in the preset time period is monitored, and when the average number of people is greater than the set first number of people threshold, alarm information about the flow of people is generated.
  • the length of the preset time period may be set as required, for example, the length of the preset time period may be 5 seconds, 10 seconds, 60 seconds, 5 minutes, and the like.
  • the length of the preset time period corresponding to the cross-line event and the preset time period corresponding to the overcrowding event may be the same or different.
  • the number of target objects at different collection time points within the preset time period includes: the number of target objects at 08:10:01 is 50, and the number of target objects at 08:10:02 The number of target objects is 53, the number of target objects at 08:10:03 is 52, the number of target objects at 08:10:04 is 51, and the number of target objects at 08:10:05 is 54 , you can take the average of the five detection results, and the average is 52. It is determined that the average number of people in the monitoring area corresponding to the monitoring point is 52 within 08:10:01 to 08:10:05.
  • the average number of people in the monitoring area corresponding to the monitoring point within a preset time period can be monitored, and when the average number of people is greater than the set first number of people threshold, alarm information about the flow of people is generated.
  • the generated crowd status alarm information may be "Attention, there are many people in the xx area at the current time”.
  • the alarm event type of the crowd flow status alarm information is: point overload alarm.
  • detailed information of the crowd flow state alarm may be displayed, including but not limited to the alarm point (namely, the name of the alarm monitoring device, etc.) Alarm time and alarm event type.
  • the alarm event type is an alarm of over-density of points
  • the detailed information may also include: the real-time number of people in the detection area at the current time point.
  • FIG. 6 as a schematic diagram of an interface displaying detailed information of a crowd flow state alarm.
  • This FIG. 6 includes alarm details and statistics on the time period of overcrowded events of the day.
  • the alarm details include alarm point, event type, alarm time, and overcrowding duration. , peak number of people, peak density, etc.
  • the statistics of the current overcrowding event period include overcrowding event alarms from the zero point of the day to the current time of the statistics.
  • 6 also includes a screenshot of the video screen, and multiple frames of alarm pictures are displayed below the screenshot of the video screen, wherein the number of alarm pictures is related to the duration of the over-density event, for example, when the duration of the over-density event is 17 minutes , you can extract a frame of alarm pictures every one minute as an alarm record, that is, you can display 17 frames of alarm pictures below the screenshot of the video screen.
  • the name of the monitoring device, the installation location, the point information such as the collected monitoring video, and the real-time number of people per minute, the maximum real-time number of people, and the minimum real-time number of people of the monitoring device can also be included.
  • Persistent associations of information such as values are stored in a search server (e.g., elasticsearch) for subsequent search queries.
  • the average number of people in the monitoring area corresponding to the monitoring point within the preset time period is determined based on the number of target objects at different collection time points within the preset time period;
  • the alarm information of the crowd flow state is generated, which can realize the monitoring of the average number of people in the detection area of the monitoring video, so as to be based on
  • the generated crowd flow status alarm information diverts the flow of people in the detection area, and reduces the occurrence rate of safety accidents caused when the people in the detection area are crowded.
  • determining the flow state data of at least one monitoring device includes:
  • Step 1 For each monitoring point, based on the number of target objects, determine the average number of people in the monitoring area corresponding to the monitoring point within a preset time period.
  • Step 2 Determine the total real-time number of people in the target monitoring area based on the average number of people.
  • generating the people flow state alarm information includes: when the total real-time number of people is greater than the set second number of people threshold, generating the people flow state alarm information.
  • the average number of people in the monitoring area corresponding to the monitoring point within the preset time period can be determined based on the number of target objects at different collection time points within the preset time period; The average number of people corresponding to each monitoring point is added to determine the total real-time number of people in the target monitoring area.
  • the traffic state alarm information is generated.
  • the function of analyzing the overcrowding of people in each surveillance video in the video group can be enabled by triggering the open button of the overcrowding event corresponding to the video group set on the display interface.
  • the monitoring videos respectively collected by the multiple monitoring devices constitute a video group.
  • the trained deep learning algorithm for identifying the target object can be used to detect the detection area indicated by the monitoring mark in the monitoring video,
  • the detection result is output in real time, and the detection result can be the number of target objects in the detection area at the collection time point in the surveillance video.
  • the average number of people in the monitoring area corresponding to the monitoring point within a preset time period may be determined based on the periodically obtained detection results.
  • the average number of people corresponding to each surveillance video included in the video group can be added to determine the total real-time number of people in the target surveillance area. Further, the total real-time number of people in the target monitoring area can be monitored, and when the total real-time number of people in the target monitoring area is greater than the set second threshold of the number of people, alarm information on the flow of people is generated.
  • the generated crowd status alarm information may be "Attention, there are many people in the scene xx at the current time".
  • the alarm event type of the crowd flow state alarm information is: video group overcrowded alarm.
  • a multi-level alarm risk may be set for an over-densified video group alarm.
  • the multi-level alarm risk includes: a first-level risk, a major risk, and a major risk, and different alarm risks are set for different alarm risks.
  • the second population threshold for example, the second population threshold corresponding to the first-level risk may be 100, the second population threshold corresponding to the second-level risk may be 200, and the second population threshold corresponding to the major risk may be 500.
  • the warning information of the people flow state corresponding to the first-level risk may be: the warning information of text format; the warning information of the people flow state corresponding to the second-level risk may be: the warning information of the people flow state; the warning information of the people flow state corresponding to the third-level risk may be: Alarm information in video format.
  • detailed information of the crowd flow state alarm may be displayed, including but not limited to the alarm point (namely, the name of the alarm monitoring device, etc.)
  • Alarm time, alarm event type, when the alarm event type is a video group over-density alarm the detailed information may also include: the total real-time number of people in the real scene.
  • FIG. 7A a schematic diagram of an interface showing the detailed information of the human flow state alarm
  • the detailed information of the human flow state alarm is displayed in a map mode in FIG. 7A
  • FIG. 5B shows the detailed information of the crowd flow state alarm in a list mode, wherein the list shown in FIG. 5B includes over-density events and line crossing events.
  • the alarm details shown in FIG. 7B may be displayed, and FIG.
  • the alarm details displayed in 7B include group name (namely the name corresponding to the video group), event type, alarm time, duration, peak number of people, peak density, real-time total number of people statistics on the day, and video source statistics.
  • the change of the people flow state data over time can also be generated.
  • the schematic diagram of the change of the flow state data over time includes a first schematic diagram of the real-time total number of people over time, and the first schematic diagram of the change includes the relationship between the peak number of people and the change of the valley value of the number of people over time.
  • a second schematic diagram of the change of the total stock of people flow over time includes the change relationship of the total flow of people over time, the change relationship of the total flow of people over time, and the change of the total stock of people flow over time relation.
  • the time interval set by the first change graph and the second change graph may be 5 minutes, 10 minutes, 30 minutes, 1 hour, and the like.
  • FIG. 8A shows a schematic interface diagram of the real-time total number of people and the total stock of people flow over time provided by an embodiment of the present disclosure; wherein, FIG. 8A provides a schematic interface diagram of the real-time total number of people changing over time, and the total number of people flow Interface diagram of stock change over time.
  • the peak and valley values of the number of people in the overcrowded event can be viewed, where the peak number of people is the highest value of the real-time number of people in a certain time period, and the lowest value of the real-time number of people in a certain time period, etc. , which can monitor the surveillance video, so that users can monitor the flow status data in real time.
  • FIG. 8B shows a schematic diagram of an interface for predicting human flow data in a future time period provided by an embodiment of the present disclosure; the time period given in FIG. 8B is from 1:00 on April 16, 20 to 11:00 on April 16, 20
  • the crowd flow state data in this way, can generate a crowd flow diversion plan based on prediction data of the crowd flow state data in a future date.
  • the total real-time number of people in the target monitoring area may be determined based on the average number of people corresponding to the multiple monitoring points respectively; And when it is determined that the total real-time number of people in the target monitoring area is greater than the set second threshold of the number of people, the alarm information of the flow of people is generated, which can realize the early warning of the total real-time number of people in the target monitoring area, so that when the total real-time number of people is large, Based on the generated human flow status alarm information, the personnel in the target monitoring area are diverted, and the occurrence rate of security accidents caused by a large number of real-time people in the target monitoring area is reduced.
  • the method further includes: averaging the crowd flow state data at the same collection time point in recent multiple historical dates to obtain predicted crowd flow state data corresponding to each collection time point; based on the predicted crowd flow state data, which constitutes the prediction data of the crowd flow state data in the future date; wherein, the prediction data is used to generate the crowd flow diversion plan.
  • multiple historical periods can be set as required.
  • multiple historical periods can be the data on the flow of people in the last 7 days (one historical period corresponds to one day), that is, at 00:00 on October 8, it can be obtained From October 1st to October 7th (7 historical periods), the crowd flow status data at the same collection time point in the last 7 historical dates are averaged to obtain the predicted crowd flow status data corresponding to each collection time point.
  • the predicted people flow state data corresponding to each collection time point respectively constitutes the prediction data of the people flow state data in the future date.
  • the crowd flow status data from October 1st to October 7th can be averaged at the same collection time point to obtain the average value of each collection time point, and the average value is the predicted crowd flow status data corresponding to the collection time point.
  • the predicted people flow state data corresponding to each collection time point respectively constitutes the prediction data of the people flow state data in the future date. For example, generate the forecast data of the total number of inflows in the future date (one day in the future), generate the forecast data of the total number of outgoing people in the future date (one day in the future), and generate the forecast data of the net stock of people in the future date (one day in the future) forecast data.
  • a people flow diversion plan can be generated. For example, if it is known from the forecast data that the total number of people in real time is the largest at 15:00, the number of people entering the target monitoring area can be controlled at 15:00. .
  • the method can be applied to shopping malls, halls and other scenarios.
  • the following takes a shopping mall as an example to illustrate the line crossing event of a surveillance video and the line crossing event of a video group.
  • the shopping mall has two doors, you can set up a monitoring device at each door position (monitoring point). That is, monitoring equipment 1 (monitoring equipment 1 set up at one monitoring point) collects the surveillance video of gate A, and monitoring equipment 2 (monitoring equipment 2 set up at monitoring point 2) collects the surveillance video of gate B.
  • the monitoring equipment 1 The second monitoring device can monitor the pedestrians entering and leaving the door.
  • the first surveillance video collected by the first monitoring device and the second surveillance video collected by the second monitoring device may be obtained.
  • For surveillance video 1 draw the entry and exit boundaries and entry and exit directions on the screenshots of surveillance video 1, and determine the total number of people entering and exiting within the preset time period in the monitoring area corresponding to the monitoring point in surveillance video 1; Further, in the case that the total number of people entering the preset time period is greater than the set first threshold, and the total number of people leaving the preset time period is greater than at least one of the set second thresholds, generate a traffic state alarm message .
  • surveillance video 2 set the entry-exit boundary and entry-exit direction on the video screenshot of surveillance video 2, and determine the total number of people entering and exiting within the preset time period in the monitoring area corresponding to the monitoring point in surveillance video 2. . Further, when the total number of people entering the preset time period is greater than the set first threshold, and when the total number of people leaving within the preset time period is greater than at least one of the set second thresholds, generate a traffic state alarm information.
  • surveillance video 1 and surveillance video 2 constitute a video group, and the video group can be analyzed to determine the crowd flow status data in the target surveillance area corresponding to surveillance device 1 and surveillance device 2.
  • surveillance video 1 determine the total number of people entering and exiting within a preset time period in the monitoring area corresponding to surveillance point 1; for surveillance video 2, determine surveillance point 2 The total number of incoming and outgoing people in the preset time period in the corresponding monitoring area. Further, based on the historical number of people in the target monitoring area within the preset time period, and the total number of incoming and outgoing people in the preset time period corresponding to the multiple monitoring points, the net personnel stock in the target monitoring area is determined.
  • the net stock of people in the mall is determined.
  • the alarm information of the pedestrian flow state is generated, so that after receiving the alarm information of the pedestrian flow status, the pedestrians in the shopping mall can be regulated to reduce the occurrence rate of congestion events.
  • it can also obtain the net stock of personnel and the real-time total number of people in each time period of the current day, so as to dynamically summarize the rules and reasonably control the number of people, adjust the working hours of the mall staff, and adjust the number of staff in different time periods.
  • the following takes the lobby as an example to describe the over-densification event of a surveillance video and the over-densification event of a video group respectively.
  • monitoring devices are installed in the four corners (four monitoring points) of the hall, that is, the four monitoring devices detect the four detection areas of the hall, and the monitoring videos collected by each monitoring device in the four monitoring devices constitute a video group.
  • the monitoring area corresponding to the monitoring point is within the preset time period.
  • alarm information about the flow of people corresponding to the surveillance video is generated. That is, it is possible to realize the monitoring of over-density events for each surveillance video in the video group.
  • the monitoring logo can be drawn on the video screenshot, that is, the area corresponding to the monitoring logo is the detection area; or the monitoring logo can not be drawn on the video screenshot, that is, the monitoring video does not have a corresponding datum plane logo.
  • the entire video is defaulted to The screen is the detection area.
  • the video group can be monitored for over-density events, and the total real-time number of people in the target monitoring area corresponding to the video group can be determined.
  • the average number of people in the monitoring area corresponding to the monitoring point within a preset time period; and based on the average number of people corresponding to the four monitoring points respectively, Determine the total real-time number of people in the target surveillance area. That is, the total real-time number of people in multiple detection areas in the hall is determined.
  • the writing order of each step does not mean a strict execution order and constitutes any limitation on the implementation process, and the specific execution order of each step should be based on its function. and possible internal logic to determine.
  • an embodiment of the present disclosure also provides a scene monitoring apparatus.
  • a schematic structural diagram of a scene monitoring apparatus 900 provided by an embodiment of the present disclosure includes a first acquisition module 901, a detection module 902, The second acquiring module 903 and the determining module 904 specifically include:
  • the first acquisition module 901 is configured to acquire monitoring video collected by monitoring equipment set at at least one monitoring point;
  • the detection module 702 is configured to determine whether a monitoring event occurs in a monitoring area corresponding to at least one monitoring point based on the monitoring video;
  • the second obtaining module 703 is configured to obtain the monitoring data of the number of people matching the monitoring event within a preset time period when a monitoring event occurs in the monitoring area corresponding to at least one monitoring point;
  • the determining module 704 is configured to determine the flow state data of the at least one monitoring device based on the monitoring data of the number of people.
  • the scene monitoring apparatus 900 further includes: an alarm module 905 configured to generate crowd flow state alarm information when the crowd flow state data satisfies the alarm condition .
  • the detection module 902 is configured to determine, based on the monitoring video, whether there is a crossing and a pre-drawn entry-exit boundary in the monitoring area corresponding to at least one monitoring point. The target object at the matched target position; in the presence of the target object, it is determined that a cross-line event occurs in the monitoring area corresponding to at least one monitoring point.
  • the second obtaining module 903 is configured to obtain the number of incoming and outgoing traffic at different collection time points within a preset time period;
  • the number of people flow refers to the number of people who cross the pre-drawn entry-exit boundary along the pre-drawn incoming direction at different collection time points; .
  • the determining module 904 is configured to determine, based on the number of inflows and the number of outflows, the total number of people in the monitoring area corresponding to the monitoring point within a preset time period. The number of people entering and the total number of people leaving;
  • the alarm module 905 is configured to generate alarm information on the state of people flow when one of the following conditions is met: the total number of people entering the flow is greater than the set first flow threshold, and the total number of people leaving the flow is greater than the second set threshold; the total flow of people entering The number is greater than the first crowd flow threshold, or, the total number of outgoing crowd flows is greater than the second crowd flow threshold.
  • the determining module 904 is configured to determine, based on the number of inflows and the number of outflows, the speed of inflow and outflow of people in the monitoring area corresponding to the monitoring point speed.
  • the determining module 904 is configured to, for each monitoring point, determine the corresponding monitoring point based on the number of incoming and outgoing traffic The total number of incoming and outgoing traffic in the preset time period in the monitoring area of The number of people flow and the total number of people leaving, to determine the net stock of people in the target monitoring area;
  • the alarm module 905 is configured to generate the alarm information of the people flow state when the net personnel stock is greater than the set net stock threshold.
  • the detection module 902 is configured to determine, based on the surveillance video, whether the number of target objects in the monitoring area corresponding to at least one monitoring point exceeds the over-density event Threshold; when the number of target objects exceeds the over-density threshold, it is determined that an over-density event occurs in the monitoring area corresponding to at least one monitoring point.
  • the second obtaining module 903 is configured to count the number of target objects at different collection time points within a preset time period.
  • the determining module 904 is configured to determine, based on the number of target objects, the average number of people in the monitoring area corresponding to the monitoring point within a preset time period;
  • the alarm module 905 is configured to generate alarm information about the flow of people when the average number of people is greater than the set first threshold of the number of people.
  • the determining module 904 is configured to, for each monitoring point, based on the number of target objects, determine that the monitoring area corresponding to the monitoring point is within the predetermined range. Set the average number of people in the time period; based on the average number of people, determine the total real-time number of people in the target monitoring area;
  • the alarm module 905 is configured to generate alarm information about the flow of people when the total real-time number of people is greater than the set second number of people threshold.
  • the scene monitoring device 900 further includes: an early warning module 906, configured to average the crowd flow status data at the same collection time point in multiple recent historical dates to obtain the predicted crowd flow corresponding to each collection time point Status data; based on the predicted human flow status data, the predicted data of the human flow status data in the future date is constituted; wherein, the predicted data is used to generate a human flow diversion plan.
  • an early warning module 906 configured to average the crowd flow status data at the same collection time point in multiple recent historical dates to obtain the predicted crowd flow corresponding to each collection time point Status data; based on the predicted human flow status data, the predicted data of the human flow status data in the future date is constituted; wherein, the predicted data is used to generate a human flow diversion plan.
  • the functions or templates included in the apparatus provided in the embodiments of the present disclosure may be configured to execute the methods described in the above method embodiments.
  • reference may be made to the above method embodiments. It is concise and will not be repeated here.
  • an embodiment of the present disclosure also provides an electronic device 1900 .
  • a schematic structural diagram of an electronic device 1900 provided by an embodiment of the present disclosure includes a processor 1901 , a memory 1902 , and a bus 1903 .
  • the memory 1902 is used to store the execution instructions, including the memory 1921 and the external memory 1922;
  • the memory 1921 here is also called the internal memory, which is used to temporarily store the operation data in the processor 1901 and the data exchanged with the external memory 1922 such as the hard disk,
  • the processor 1901 exchanges data with the external memory 1922 through the memory 1921.
  • the processor 1901 communicates with the memory 1902 through the bus 1903, so that the processor 1901 executes the following instructions:
  • a monitoring event occurs in the monitoring area corresponding to at least one monitoring point, obtain the monitoring data of the number of people matching the monitoring event within a preset time period;
  • the flow state data of at least one monitoring device is determined.
  • an embodiment of the present disclosure further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and the computer program is executed by a processor to execute the scene monitoring method described in the above method embodiments.
  • Embodiments of the present disclosure further provide a computer program, the computer program includes computer-readable codes, and when the computer-readable codes run in an electronic device, a processor in the electronic device executes any of the above Scene monitoring method.
  • the embodiments of the present disclosure also provide another computer program product, including a computer-readable storage medium storing program codes, wherein the instructions included in the program codes can be used to execute the scene monitoring method described in the above method embodiments, specifically Refer to the above method embodiments, which are not repeated here.
  • the devices involved in the embodiments of the present disclosure may be systems, methods and/or computer program products.
  • the computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of the present disclosure.
  • a computer-readable storage medium may be a tangible device that can hold and store instructions for use by the instruction execution device.
  • the computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing.
  • Computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), Erase programmable read only memory (Electrical Programmable Read Only Memory, EPROM) or flash memory, static random access memory (Static Random-Access Memory, SRAM), portable compact disk read only memory (Compact Disc Read-Only Memory, CD- ROM), Digital Video Disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or raised structures in grooves on which instructions are stored, and any suitable combination of the foregoing.
  • RAM random access memory
  • ROM read-only memory
  • EPROM Erase programmable read only memory
  • EPROM Electrical Programmable Read Only Memory
  • flash memory static random access memory
  • SRAM static random access memory
  • portable compact disk read only memory Compact Disc Read-Only Memory
  • CD- ROM Compact Disc Read-Only Memory
  • DVD Digital Video Disc
  • memory sticks floppy disks
  • mechanical encoding devices such as punch cards or raised structures
  • Computer-readable storage media are not to be construed as transient signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (eg, light pulses through fiber optic cables), or through electrical wires transmitted electrical signals.
  • the computer readable program instructions described herein may be downloaded to various computing/processing devices from a computer readable storage medium, or to an external computer or external storage device over a network such as the Internet, a local area network, a wide area network, and/or a wireless network.
  • the network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and/or edge servers.
  • a network adapter card or network interface in each computing/processing device receives computer-readable program instructions from a network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in each computing/processing device .
  • the computer program instructions for carrying out the operations of the present disclosure may be assembly instructions, Industry Standard Architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or in one or more source or object code written in any combination of programming languages, including object-oriented programming languages—such as Smalltalk, C++, etc., and conventional procedural programming languages, such as the “C” language or similar programming languages.
  • the computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server implement.
  • the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or Wide Area Network (WAN), or may be connected to an external computer (eg, using Internet service provider to connect via the Internet).
  • LAN Local Area Network
  • WAN Wide Area Network
  • electronic circuits such as programmable logic circuits, FPGAs, or Programmable Logic Arrays (PLAs), that can execute computer-readable Program instructions are read to implement various aspects of the present disclosure.
  • PDAs Programmable Logic Arrays
  • the units described as separate components may or may not be physically separated, and components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this embodiment.
  • each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit.
  • the functions, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a processor-executable non-volatile computer-readable storage medium.
  • the computer software products are stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present disclosure.
  • the aforementioned storage medium includes: U disk, removable hard disk, ROM, RAM, magnetic disk or optical disk and other media that can store program codes.
  • Embodiments of the present disclosure provide a method, device, electronic device, storage medium, and program for scene monitoring.
  • the method includes: acquiring a monitoring video collected by a monitoring device set at at least one monitoring point; whether a monitoring event occurs in the monitoring area corresponding to the at least one monitoring point; in the case where the monitoring event occurs in the monitoring area corresponding to the at least one monitoring point, obtain the number of people who match the monitoring event within a preset time period Monitoring data; based on the number of people monitoring data, determine the people flow state data of the at least one monitoring device.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Signal Processing (AREA)
  • Business, Economics & Management (AREA)
  • Emergency Management (AREA)
  • Theoretical Computer Science (AREA)
  • Alarm Systems (AREA)
  • Closed-Circuit Television Systems (AREA)
  • Image Analysis (AREA)

Abstract

本公开实施例提供了一种场景监测方法、装置、电子设备、存储介质及程序，该方法包括：获取设置于至少一个监控点位的监控设备采集的监控视频；基于所述监控视频，确定所述至少一个监控点位对应的监测区域是否发生监测事件；在所述至少一个监控点位对应的监测区域发生所述监测事件的情况下，获取预设时间段内与所述监测事件匹配的人数监测数据；基于所述人数监测数据，确定所述至少一个监控设备的人流状态数据。如此，在本公开实施例中，通过确定的人流状态数据表征监测事件的状态，实现对监控视频的监测。

Description

场景监测方法、装置、电子设备、存储介质及程序
相关申请的交叉引用
本专利申请要求2020年10月30日提交的中国专利申请号为202011190695.6、申请人为深圳市商汤科技有限公司，申请名称为“场景检测方法、装置、电子设备及存储介质”的优先权，该申请的全文以引用的方式并入本申请中。
技术领域
本公开涉及计算机视觉技术领域，具体而言，涉及一种场景监测方法、装置、电子设备、存储介质及程序。
背景技术
随着人们生活水平的提高，越来越多的大型活动在各地、各场所内举办。由于举办大型活动时，人流较为密集，使得举办大型活动的地方、场所容易发生事故，比如，踩踏事件、拥堵事件等。故为了提高各地、各场所的安全程度，对人流的有效监测越来越重要。
发明内容
有鉴于此，本公开实施例至少提供一种场景监测方法、装置、电子设备、存储介质及程序。
本公开实施例提供了一种场景监测方法，所述方法由电子设备执行，所述方法包括：
获取设置于至少一个监控点位的监控设备采集的监控视频；
基于所述监控视频，确定所述至少一个监控点位对应的监测区域是否发生监测事件；
在所述至少一个监控点位对应的监测区域发生所述监测事件的情况下，获取预设时间段内与所述监测事件匹配的人数监测数据；
基于所述人数监测数据，确定所述至少一个监控设备的人流状态数据。如此，通过获取监控设备采集的监控视频，在基于采集的监控视频，检测到至少一个监控点位对应的监测区域发生监测事件时，获取预设时间段内与监测时间匹配的人数监测数据，并基于预设时间段内与监测事件匹配的人数监测数据，确定至少一个监控设备的人流状态数据，通过确定的人流状态数据表征监测事件的状态，实现对监控视频的监测。
在本公开的一些实施例中，在确定所述至少一个监控设备的人流状态数据之后，还包括：在所述人流状态数据满足告警条件的情况下，生成人流状态告警信息。如此，在确定的人流状态数据满足告警条件时，生成人流状态告警信息，基于生成的人流状态告警信息，可以对目标监控区域进行调控，降低安全事故的发生率，提高目标监控区域下人流的安全性。
在本公开的一些实施例中，在所述监测事件为跨线事件的情况下，基于所述监控视频，确定所述至少一个监控点位对应的监测区域是否发生监测事件，包括：基于所述监控视频确定所述至少一个监控点位对应的监测区域内，是否存在跨越与预先绘制的进出界线匹配的目标位置的目标对象；在存在所述目标对象的情况下，确定所述至少一个监控点位对应的监测区域发生所述跨线事件。如此，在基于监控视频确定至少一个监控点位对应的监测区域内，存在跨越与进出界线匹配的目标位置的目标对象时，确定至少一个监控点位对应的监测区域发生跨线事件，实现了对跨线事件的实时监测，提高跨线事件监测的准确性。
在本公开的一些实施例中，在所述监测事件为跨线事件的情况下，获取预设时间段内与所述监测事件匹配的人数监测数据，包括：获取所述预设时间段内不同采集时间点的进人流数量和出人流数量；其中，所述进人流数量是指在所述不同采集时间点，沿预先绘制的进方向跨越预先绘制的进出界线的人数；所述出人流数量是指在所述不同采集时间点，沿预先绘制的出方向跨越预先绘制的进出界线的人数。如此，在监测事件为跨线事件时，可以获取预设时间段内不同采集时间点的进人流数量和出人流数量，为后续确定跨线事件对应的人流状态数据提供了数据支持。
在本公开的一些实施例中，在所述监控点位为一个的情况下，基于所述人数监测数据，确定所述至少一个监控设备的人流状态数据，包括：基于所述进人流数量和所述出人流数量，确定所述监控点 位对应的监测区域中所述预设时间段内的总进人流数量和总出人流数量；所述在所述人流状态数据满足告警条件的情况下，生成人流状态告警信息，包括：在满足以下条件之一的情况下，生成所述人流状态告警信息：所述总进人流数量大于设置的第一人流阈值，且所述总出人流数量大于设置的第二人流阈值；所述总进人流数量大于所述第一人流阈值，或，所述总出人流数量大于所述第二人流阈值。如此，在监控点位为一个时，基于预设时间段内不同采集时间点的进人流数量和出人流数量，确定监控点位对应的监测区域中预设时间段内的总进人流数量和总出人流数量。在预设时间段内的总进人流数量大于设置的第一人流阈值，和在预设时间段内的总出人流数量大于设置的第二人流阈值至少之一的情况下，生成人流状态告警信息，能够实现对该监控视频的进人流数量和出人流数量的预警，以便基于生成的人流状态告警信息进行人流的疏导，降低短时间内进人流数量较多，或者出人流数量较多造成的安全事故的发生率。
在本公开的一些实施例中，在所述监控点位为一个的情况下，基于所述人数监测数据，确定所述至少一个监控设备的人流状态数据，包括：基于所述进人流数量和所述出人流数量，确定所述监控点位对应的监测区域中的进人流速度和出人流速度。如此，可以基于预设时间段内不同采集时间点的进人流数量和出人流数量，确定监控点位对应的监测区域中的进人流速度和出人流速度，实现对进人流速度和出人流速度的监测，降低进人流速度较大，或者出人流速度较大造成的安全事故的发生率。
在本公开的一些实施例中，在所述监控点位为多个的情况下，基于所述人数监测数据，确定所述至少一个监控设备的人流状态数据，包括：针对每个所述监控点位，基于所述进人流数量和所述出人流数量，确定每个所述监控点位分别对应的监测区域中所述预设时间段内的总进人流数量和总出人流数量；基于所述预设时间段内目标监控区域的历史人数，以及每个所述监控点位分别对应的所述预设时间段内的总进人流数量和总出人流数量，确定所述目标监控区域内的人员净存量；所述在所述人流状态数据满足告警条件的情况下，生成人流状态告警信息，包括：在所述人员净存量大于设置的净存量阈值的情况下，生成所述人流状态告警信息。如此，在确定每个监控点位对应的监测区域中预设时间段内的总进人流数量和总出人流数量之后，可以基于预设时间段内目标监控区域的历史人数，以及每个监控点位分别对应的预设时间段内的总进人流数量和总出人流数量，确定目标监控区域内的人员净存量，并在目标监控区域内的人员净存量大于设置的净存量阈值的情况下，生成人流状态告警信息，能够实现对目标监控区域中的人员净存量的预警，以便在人员净存量较多时，基于生成的人流状态告警信息进行人员的疏导，降低目标监控区域中人员较多时造成安全事故的发生率。
在本公开的一些实施例中，在所述监测事件为过密事件的情况下，基于所述监控视频，确定所述至少一个监控点位对应的监测区域是否发生监测事件，包括：基于所述监控视频，确定所述至少一个监控点位对应的监测区域内的目标对象个数是否超过过密阈值；在所述目标对象个数超过所述过密阈值的情况下，确定所述至少一个监控点位对应的监测区域发生过密事件。如此，在基于监控视频，确定至少一个监控点位对应的监测区域内的目标对象个数超过过密阈值时，确定至少一个监控点位对应的监测区域发生过密事件，能够实现对过密事件的实时监测，提高过密事件监测的准确性。
在本公开的一些实施例中，在所述监测事件为过密事件的情况下，获取预设时间段内与所述监测事件匹配的人数监测数据，包括：统计所述预设时间段内不同采集时间点的所述目标对象的个数。如此，在监测事件为过密事件时，可以统计预设时间段内不同采集时间点的所述目标对象的个数，能够为后续确定过密事件对应的人流状态数据提供数据支持。
在本公开的一些实施例中，在所述监控点位为一个的情况下，基于所述人数监测数据，确定所述至少一个监控设备的人流状态数据，包括：基于所述目标对象的个数，确定所述监控点位对应的监测区域在所述预设时间段内的平均人数；所述在所述人流状态数据满足告警条件的情况下，生成人流状态告警信息，包括：在所述平均人数大于设置的第一人数阈值的情况下，生成所述人流状态告警信息。如此，在监控点位为一个时，基于预设时间段内不同采集时间点的目标对象的个数，确定监控点位对应的监测区域在预设时间段内的平均人数；并在监控点位对应的监测区域在预设时间段内的平均人数大于设置的第一人数阈值的情况下，生成人流状态告警信息，能够实现对该监控视频的检测区域中的平均人数的监控，以便基于生成的人流状态告警信息对检测区域进行人流疏导，降低检测区域内的人 员较为密集时，造成的安全事故的发生率。
在本公开的一些实施例中，在所述监控点位为多个的情况下，基于所述人数监测数据，确定所述至少一个监控设备的人流状态数据，包括：针对每个所述监控点位，基于所述目标对象的个数，确定所述监控点位对应的监测区域在所述预设时间段内的平均人数；基于所述平均人数，确定目标监控区域中的总实时人数；所述在所述人流状态数据满足告警条件的情况下，生成人流状态告警信息，包括：在所述总实时人数大于设置的第二人数阈值的情况下，生成所述人流状态告警信息。如此，在确定了每个监控点位对应的监测区域在预设时间段内的平均人数之后，可以基于多个所述监控点位分别对应的平均人数，确定目标监控区域中的总实时人数；并在确定目标监控区域中的总实时人数大于设置的第二人数阈值的情况下，生成人流状态告警信息，能够实现对目标监控区域中的总实时人数的预警，以便在总实时人数较多时，基于生成的人流状态告警信息对目标监控区域中的人员进行疏导，降低目标监控区域中总实时人数较多时造成安全事故的发生率。
在本公开的一些实施例中，所述方法还包括：将最近多个历史日期内同一采集时间点的人流状态数据求平均，得到每个采集时间点对应的预测人流状态数据；基于所述预测人流状态数据，构成人流状态数据在未来日期内的预测数据；其中，所述预测数据用于生成人流疏导计划。如此，能够基于人流状态数据在未来日期内的预测数据，生成人流疏导计划。
以下装置、电子设备等的效果描述参见上述方法的说明，这里不再赘述。
本公开实施例还提供了一种场景监测装置，包括：
第一获取模块，配置为获取设置于至少一个监控点位的监控设备采集的监控视频；
检测模块，配置为基于所述监控视频，确定所述至少一个监控点位对应的监测区域是否发生监测事件；
第二获取模块，配置为在所述至少一个监控点位对应的监测区域发生所述监测事件的情况下，获取预设时间段内与所述监测事件匹配的人数监测数据；
确定模块，配置为基于所述人数监测数据，确定所述至少一个监控设备的人流状态数据。
在本公开的一些实施例中，在确定所述至少一个监控设备的人流状态数据之后，所述场景监测装置，还包括：告警模块，配置为在所述人流状态数据满足告警条件的情况下，生成人流状态告警信息。
在本公开的一些实施例中，在所述监测事件为跨线事件的情况下，检测模块，配置为基于所述监控视频确定所述至少一个监控点位对应的监测区域内，是否存在跨越与预先绘制的进出界线匹配的目标位置的目标对象；在存在所述目标对象的情况下，确定所述至少一个监控点位对应的监测区域发生所述跨线事件。
在本公开的一些实施例中，在所述监测事件为跨线事件的情况下，第二获取模块，配置为获取所述预设时间段内不同采集时间点的进人流数量和出人流数量；其中，所述进人流数量是指在所述不同采集时间点，沿预先绘制的进方向跨越预先绘制的进出界线的人数；所述出人流数量是指在所述不同采集时间点，沿预先绘制的出方向跨越预先绘制的进出界线的人数。
在本公开的一些实施例中，在所述监控点位为一个的情况下，确定模块，配置为基于所述进人流数量和所述出人流数量，确定所述监控点位对应的监测区域中所述预设时间段内的总进人流数量和总出人流数量；
告警模块，配置为在满足以下条件之一的情况下，生成所述人流状态告警信息：所述总进人流数量大于设置的第一人流阈值，且所述总出人流数量大于设置的第二人流阈值；所述总进人流数量大于所述第一人流阈值，或，所述总出人流数量大于所述第二人流阈值。
在本公开的一些实施例中，在所述监控点位为一个的情况下，确定模块，配置为基于所述进人流数量和所述出人流数量，确定所述监控点位对应的监测区域中的进人流速度和出人流速度。
在本公开的一些实施例中，在所述监控点位为多个的情况下，确定模块，配置为针对每个所述监控点位，基于所述进人流数量和所述出人流数量，确定每个所述监控点位分别对应的监测区域中所述预设时间段内的总进人流数量和总出人流数量；基于所述预设时间段内目标监控区域的历史人数，以及每个所述监控点位分别对应的所述预设时间段内的总进人流数量和总出人流数量，确定所述目标监 控区域内的人员净存量；
告警模块，配置为在所述人员净存量大于设置的净存量阈值的情况下，生成所述人流状态告警信息。
在本公开的一些实施例中，在所述监测事件为过密事件的情况下，检测模块，配置为基于所述监控视频，确定所述至少一个监控点位对应的监测区域内的目标对象个数是否超过过密阈值；在所述目标对象个数超过所述过密阈值的情况下，确定所述至少一个监控点位对应的监测区域发生过密事件。
在本公开的一些实施例中，在所述监测事件为过密事件的情况下，第二获取模块，配置为统计所述预设时间段内不同采集时间点的所述目标对象的个数。
在本公开的一些实施例中，在所述监控点位为一个的情况下，确定模块，配置为基于所述目标对象的个数，确定所述监控点位对应的监测区域在所述预设时间段内的平均人数；
告警模块，配置为在所述平均人数大于设置的第一人数阈值的情况下，生成所述人流状态告警信息。
在本公开的一些实施例中，在所述监控点位为多个的情况下，确定模块，配置为针对每个所述监控点位，基于所述目标对象的个数，确定所述监控点位对应的监测区域在所述预设时间段内的平均人数；基于所述平均人数，确定目标监控区域中的总实时人数；
告警模块，配置为在所述总实时人数大于设置的第二人数阈值的情况下，生成所述人流状态告警信息。
在本公开的一些实施例中，所述场景检测装置还包括：预警模块，配置为将最近多个历史日期内同一采集时间点的人流状态数据求平均，得到每个采集时间点对应的预测人流状态数据；基于所述预测人流状态数据，构成人流状态数据在未来日期内的预测数据；其中，所述预测数据用于生成人流疏导计划。
本公开实施例还提供一种电子设备，包括：处理器、存储器和总线，所述存储器存储有所述处理器可执行的机器可读指令，当所述电子设备运行时，所述处理器与所述存储器之间通过总线通信，所述机器可读指令被所述处理器执行时执行如上述任一实施例所述的场景监测方法。
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如上述任一实施例所述的场景监测方法。
本公开实施例还提供一种计算机程序，所述计算机程序包括计算机可读代码，在所述计算机可读代码在电子设备中运行的情况下，所述电子设备的处理器执行如上述任一实施例所述的场景监测方法。
本公开实施例至少提供一种场景监测方法、装置、电子设备、存储介质及程序，通过获取监控设备采集的监控视频，在基于采集的监控视频，检测到至少一个监控点位对应的监测区域发生监测事件时，获取预设时间段内与监测时间匹配的人数监测数据，并基于人数监测数据，确定至少一个监控设备的人流状态数据，通过确定的人流状态数据表征监测事件的状态，实现对监控视频的监测。
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实施例，并与说明书一起用于说明本公开实施例的技术方案。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场景监测方法的流程示意图；
图2示出可以应用本公开实施例的场景监测方法的一种系统架构示意图；
图3A示出了本公开实施例所提供的一种展示绘制有监测标识的视频画面截图的界面示意图的界 面示意图；
图3B示出了本公开实施例所提供的另一种展示绘制有监测标识的视频画面截图的界面示意图的界面示意图；
图3C示出了本公开实施例展示的一种人流数据对应的预警信息的界面示意图；
图4示出了本公开实施例所提供的一种展示人流状态告警的详细信息的界面示意图；
图5A示出了本公开实施例所提供的一种展示人流状态告警的详细信息的界面示意图；
图5B示出了本公开实施例所提供的另一种展示人流状态告警的详细信息的界面示意图；
图5C示出了本公开实施例所提供的一种展示告警详情的界面示意图；
图6示出了本公开实施例所提供的一种展示人流状态告警的详细信息的界面示意图；
图7A示出了本公开实施例所提供的一种展示人流状态告警的详细信息的界面示意图；
图7B示出了本公开实施例所提供的一种展示告警详情的界面示意图；
图8A示出了本公开实施例提供的一种实时总人数和人流总存量随时间变化的界面示意图；
图8B示出了本公开实施例提供的一种预测未来时间段的人流数据的界面示意图；
图9示出了本公开实施例所提供的一种场景监测装置900的架构示意图；
图10示出了本公开实施例所提供的一种电子设备1900的结构示意图。
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的附图，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。通常在此处附图中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，以下对在附图中提供的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。
随着人们生活水平的提高，越来越多的大型活动在各地、各场所内举办。由于举办大型活动时，人流较为密集，使得举办大型活动的地方、场所容易发生事故，同时，人流密集区域对于通过视频实时分析人群密度、预测人流量、区域人员净存量的需求越来越迫切。人群分析在生活中的很多领域都有重要应用，能够及时的通过人群分析，了解区域内实时人数，减少踩踏等重大事件，保障人民安全；同时也能运用在商业场景，帮助商家分析客户行为。
但目前已有的人群分析相关应用更多关注的是估算视频区域内的人数或进行区域入侵等行为预警，这种机制只能针对实时情况进行查看和处理，无法进行预测预防安全问题，也无法进行数据分析进而总结人流规律。
为了解决上述问题，提高地方、场所的安全度，本公开实施例提供了一种场景监测方法、装置、电子设备、存储介质及程序。
为便于对本公开实施例进行理解，首先对本公开实施例所公开的一种场景检测方法进行详细介绍。本公开实施例所提供的场景监测方法的执行主体一般为具有一定计算能力的计算机设备，该计算机设备例如包括：终端设备或服务器或其它处理设备，终端设备可以为用户设备(User Equipment，UE)、移动设备、用户终端、终端、蜂窝电话、无绳电话、个人数字处理(Personal Digital Assistant，PDA)、手持设备、计算设备、车载设备、可穿戴设备等。在本公开的一些实施例中，该场景监测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。
参见图1所示，为本公开实施例所提供的一种场景监测方法的流程示意图，该方法包括S101至S104，具体包括：
S101，获取设置于至少一个监控点位的监控设备采集的监控视频。
S102，基于监控视频，确定至少一个监控点位对应的监测区域是否发生监测事件。
S103，在至少一个监控点位对应的监测区域发生监测事件的情况下，获取预设时间段内与监测事件匹配的人数监测数据。
S104，基于人数监测数据，确定至少一个监控设备的人流状态数据。
上述方法中，通过获取监控设备采集的监控视频，在基于采集的监控视频，检测到至少一个监控点位对应的监测区域发生监测事件时，获取预设时间段内与监测时间匹配的人数监测数据，并基于人数监测数据，确定至少一个监控设备的人流状态数据，通过确定的人流状态数据表征监测事件的状态，实现对监控视频的监测，比如，人流状态数据可以为总进人流数量，在总进人流数量较大时，表征监测事件的发生较为频繁。
图2示出可以应用本公开实施例的场景监测方法的一种系统架构示意图；如图2所示，该系统架构中包括：监控视频获取终端201、网络202和场景监测终端203。为实现支撑一个示例性应用，监控视频获取终端201和场景监测终端203通过网络202建立通信连接，监控视频获取终端201通过网络202向场景监测终端203上报监控视频，场景监测终端203响应于监控视频，并基于监控视频确定至少一个监控点位对应的监测区域是否发生监测事件，在至少一个监控点位对应的监测区域发生监测事件的情况下，获取预设时间段内与监测事件匹配的人数监测数据，以及基于人数监测数据，确定至少一个监控设备的人流状态数据。最后，场景监测终端203将监控视频的人流状态数据上传至网络202，并通过网络202发送给监控视频获取终端201。
作为示例，监控视频获取终端201可以包括视频采集设备，场景监测终端203可以包括具有视觉信息处理能力的视觉处理设备或远程服务器。网络202可以采用有线或无线连接方式。其中，当场景监测终端203为视觉处理设备时，监控视频获取终端201可以通过有线连接的方式与视觉处理设备通信连接，例如通过总线进行数据通信；当场景监测终端203为远程服务器时，监控视频获取终端201可以通过无线网络与远程服务器进行数据交互。
或者，在一些场景中，监控视频获取终端201可以是带有视频采集模组的视觉处理设备，具体实现为带有摄像头的主机。这时，本公开实施例的场景监测方法可以由监控视频获取终端201执行，上述系统架构可以不包含网络202和场景监测终端203。
下述对S101至S104进行说明。
针对S101：
在本公开的一些实施例中，可以使用该方法对目标监控区域进行检测，该目标监控区域可以为现实场景中的任一区域，比如，该目标监控区域可以为商场、沙滩、公园以及地铁站等。
示例性的，可以在目标监控区域处设置多个监控点位，在每个监控点位处安装一个监控设备，以便监控设备可以对对应的监测区域进行监测，实现对目标监控区域的监测。其中，监控点位的设置可以根据实际需要进行确定，比如，在目标监控区域为商场时，可以在商场的每个门处设置一个监控点位，和每个电梯口至少之一处设置一个监控点位等。
在本公开的一些实施例中，监控设备可以为监控摄像头等设备。通过在每个监控点位处设置一个监控设备，通过监控设备采集对应监测区域内的监控视频，以便可以获取每个监控设备采集的监控视频，即获取设置于至少一个监控点位的监控设备采集的监控视频。
针对S102：
这里，可以针对每个监控设备采集的监控视频，基于该监控视频，确定该监控点位对应的监测区域是否发生监测事件，进而可以确定至少一个监控点位中每个监控点位对应的监测区域是否发生监测事件。其中，监测事件可以包括过密事件和跨线事件至少之一；过密事件是指区域内的人数密度大于设置的值，即区域内的人员密度较大；跨线是指区域内有行人跨越了设置的基准线。
结合具体的场景对跨线事件和过密事件进行说明，比如，针对跨线事件，在地铁站内，可以在站台上，距离地铁预设距离(例如1米)的位置处设置一条基准线，监测是否有人跨越该基准线(即是否有人跨越基准线进入地铁或出地铁)，若是，则发生了跨越事件。比如，针对过密事件，在沙滩上，可以设置一个目标监控区域，在该目标监控区域内的人数大于设置的人数值时，确定发生了过密事件。
在本公开的一些实施例中，还可以针对每个监控视频，设置功能按钮，通过功能按钮的触发，确定对监控点位的监测区域进行监测事件的监测。比如，可以设置跨线事件对应的第一功能按钮(针对单个监控视频进行跨线事件的监测)，在监控视频A对应的第一功能按钮被触发后，则确定对监控视频 A进行跨线事件的监测；或者，还可以设置跨线事件对应的第二功能按钮(针对监控视频组进行跨线事件的监测)，在监控视频A、监控视频B等构成的监控视频组A对应的第二功能按钮被触发后，则确定对该监控视频组A进行跨线事件的监测。
在本公开的一些实施例中，还可以设置过密事件对应的第三功能按钮(针对单个监控视频进行过密事件的监测)，在监控视频A对应的第三功能按钮被触发后，则确定对监控视频A进行过密事件的监测；或者，还可以设置过密事件对应的第四功能按钮(针对监控视频组进行过密事件的监测)，在监控视频A、监控视频B等构成的监控视频组A对应的第四功能按钮被触发后，则确定对该监控视频组A进行过密事件的监测。
在本公开的一些实施例中，在基于监控视频，确定至少一个监控点位对应的监测区域是否发生监测事件之前，可以绘制该监控视频对应的监测标识。针对跨线事件，该监测标识可以为预先绘制的进出界线、进方向和出方向；针对过密事件，该监测标识可以为预先绘制的任一多边形，或者，针对过密事件，可以不设置对应的监测标识。其中，不同监控视频对应的监测标识不同，即可以针对每个监控视频，为该监控视频绘制对应的监测标识(跨线事件对应的监测标识和过密事件对应的监测标识至少之一)。
在本公开的一些实施例中，针对每个监控视频，可以从该监控视频中采集一帧视频画面截图，展示该视频画面截图，使得用户可以根据实际需要在视频画面截图上绘制监测标识。再可以通过获取预先绘制有监测标识的视频画面截图，确定视频画面截图中的监测标识在视频画面截图中的位置信息，其中，该位置信息可以为监测标识在视频画面截图对应的像素坐标系下的坐标集合，比如，可以为进出界线的位置信息等。进而，可以在该监控视频的视频画面中确定与监测标识匹配的目标位置信息。其中，在监控设备的位置、朝向等安装信息不发生改变时，监测标识在视频画面截图中的位置信息，可以为监测标识在监控视频的视频画面中的目标位置信息。进而，可以基于针对该监控设备采集的监控视频和确定的目标位置信息，确定该监控点位对应的监测区域是否发生监测事件。
在本公开的一些实施例中，在监测标识包括跨线事件对应的监测标识时，参见图3A所示的一种展示绘制有监测标识的视频画面截图的界面示意图，该图3A中包括预先绘制的监测标识31，监测标识包括绘制的进出界线和指示进出方向的箭头标识。在绘制监测标识时，还可以在显示的界面上设置入人流阈值(即第一人流阈值)和出人流阈值(即第二人流阈值)至少之一，以便基于设置的入人流阈值和出人流阈值至少之一，对监控视频进行监测。该图3A中还包括位于视频画面截图上方的越线事件设置的提示信息，以便用户根据显示的越线事件设置的提示信息，绘制监测标识。在绘制监测标识时，还可以触发“重新绘制”的按钮，将已绘制的监测标识删除，重新绘制新的监测标识。
在本公开的一些实施例中，在监测标识包括过密事件对应的监测标识时，参见图3B所示的另一种绘制有监测标识的视频画面截图的界面示意图，该图3B中包括预先绘制的监测标识32，监测标识包括指示检测区域的多边形，其中，检测区域的数量可以为多个。在绘制监测标识时，还可以在显示的界面上设置分级预警人数，即一般风险对应的预警人数、较大风险对应的预警人数以及重大风险对应的预警人数，以便基于设置的分级预警人数，对监控视频进行监测。该图3B中还包括位于视频画面截图上方的过密事件设置的提示信息，以便用户根据显示的越线事件设置的提示信息，绘制指示检测区域的监测标识。在绘制监测标识时，还可以触发“重新绘制”的按钮，将已绘制的监测标识删除，重新绘制新的监测标识。在绘制了监控视频对应的监测标识后，可以将绘制的监测标识存储在复用区域内，以便下次确定监测标识时，可以直接触发复用区域的功能按钮，实现监测标识的再次利用。
图3B中的人体标注的功能按钮，用于展示人体标注的设置信息。考虑到，监控视频画面中人体的面积大小、与监控设备的高度和角度有关，且同一人体与监控设备的距离不同，在监控视频画面中的面积大小不同，即与监控设备的距离较近时，人体的面积较大，故人体标注是跨线事件、过密事件的基础设置。
在本公开的一些实施例中，可以在图3A和图3B示出的视频画面截图中，从位于不同深度位置处标记多个行人的人体框，估计每个行人的人体框的面积和所处的深度信息；便于算法(比如，用于识别人体的图像识别算法)利用人体标注结果，对不同情况下的不同监控设备进行人体识别，提高识别 精度，其中，人体框越多，精度越高，在本公开的一些实施例中，标记的行人框的数量可以根据需要进行设置，比如，可以设置标记的行人框的数量范围为3至10个。进而可以利用多个行人的人体框的面积以及每个行人所处的深度信息，对监控视频中每秒视频画面的检测区域中包括的实时人数进行检测。
在过密事件中，当绘制了检测区域(监测标识)后，可以基于人体标注中标记的人体样本，计算出绘制区域在现实场景中的预测面积，并在图3B下方的“区域面积预估”处显示预测面积，并可以在后续的点位过密告警、视频组过密告警等中，计算出检测区域内的人员密度。以及，该图3B中还包括“更正面积”的功能按钮，可以在触发了“更正面积”的功能按钮后，对区域面积预估处显示的预测面积进行更正。
在本公开的一些实施例中，在视频画面的监测标识同时包括过密事件和跨线时间对应的监测标识时，参见图3C所示的为本公开实施例展示的一种人流数据对应的预警信息的界面示意图，该图3C中分别包括过密事件和跨线事件对应的预警级别以及人数。即过密事件中一般风险对应的预警人数、较大风险对应的预警人数以及重大风险对应的预警人数，跨线事件中存量越多对应的预警人数、存量警示对应的预警人数以及存量过热对应的预警人数，以便基于设置的分级预警人数，对监控视频进行监测。该图3C中还包括在视频画面启动过密事件和跨线事件的有效时间的两个按钮，即“长期有效”和“自定义”，以便用户灵活设置相关监测参数以及时间信息。
在本公开的一些实施例中，在监测事件为跨线事件的情况下，基于监控视频确定至少一个监控点位对应的监测区域是否发生监测事件，包括：基于监控视频确定至少一个监控点位对应的监测区域内，是否存在跨越与预先绘制的进出界线匹配的目标位置的目标对象；在存在目标对象的情况下，确定至少一个监控点位对应的监测区域发生跨线事件。
在本公开的一些实施例中，在监测事件为跨线事件时，针对每个监控点位采集的监控视频，可以基于监控视频，确定该监控点位对应的监测区域内，是否存在跨越与进出界线匹配的目标位置的目标对象，比如在监控视频中检测是否有行人跨越了绘制的进出界线，若存在，则确定该监控点位对应的监测区域发生跨线事件；若不存在，则确定该监控点位对应的监测区域未发生跨线事件。
其中，监控点位对应的监测区域可以为该监控点位处设置的监控设备可以监控的检测区域；监控点位对应的监测区域与监控设备的安装位置、安装角度有关，不同的安装位置和安装角度对应不同的监测区域。
在本公开的一些实施例中，在基于监控视频确定至少一个监控点位对应的监测区域内，存在跨越与进出界线匹配的目标位置的目标对象时，确定至少一个监控点位对应的监测区域发生跨线事件，能够实现对跨线事件的实时监测，提高跨线事件监测的准确性。
在本公开的一些实施例中，在监测事件为过密事件的情况下，基于监控视频确定至少一个监控点位对应的监测区域是否发生监测事件，包括：基于监控视频，确定至少一个监控点位对应的监测区域内的目标对象个数是否超过过密阈值；在目标对象个数超过过密阈值的情况下，确定至少一个监控点位对应的监测区域发生过密事件。
在本公开的一些实施例中，在监测事件为过密事件时，针对每个监控点位采集的监控视频，可以基于监控视频，确定该监控点位对应的监测区域内目标对象的数量是否超过过密阈值，比如在监控视频中确定监测区域内的人类数量，判断该人类数量是否大于预先设置的过密阈值，若是，则确定该监控点位对应的监测区域发生过密事件；若否，则确定该监控点位对应的监测区域未发生过密事件。这里，在监测事件为过密事件时，监控点位对应的监测区域可以为与绘制的多边形匹配的检测区域；在未绘制监测标识时，则监控点位对应的监测区域为该监控点位处设置的监控设备可以监控的检测区域(即监控视频的监控界面对应的区域均为监测区域)。
上述方法中，在基于监控视频，确定至少一个监控点位对应的监测区域内的目标对象个数超过过密阈值时，确定至少一个监控点位对应的监测区域发生过密事件，能够实现对过密事件的实时监测，提高过密事件监测的准确性。
针对S103和S104：
在本公开的一些实施例中，在确定至少一个监控点位对应的监测区域发生监测事件时，可以获取预设时间段内与监测事件匹配的人数监测数据；人数监测数据包括跨线事件对应的人数监测数据和过密事件对应的人数监测数据至少之一。进而，基于预设时间段内与监测事件匹配的人数监测数据，确定至少一个监控设备的人流状态数据；人流状态数据包括跨线事件对应的人流状态数据和过密事件对应的人流状态数据至少之一。其中，预设时间段可以根据需要进行设置，比如，预设时间段可以为在确定发生监测事件的时刻开始至一个小时后的时间段，若确定发生监测事件的时刻为13时10分00秒，则预设时间段为从13时10分00秒至14时10分00秒内的时间段。比如，预设时间段可以为在确定发生监测事件的时刻开始至一分钟后的时间段，若确定发生监测事件的时刻为13时10分00秒，则预设时间段为从13时10分00秒至13时11分00秒内的时间段。
针对跨线事件，可以获取预设时间段内与跨线事件匹配的人数监测数据；并基于预设时间段内与跨线事件匹配的人数监测数据，确定至少一个监控设备的、与跨线事件匹配的人流状态数据。
针对过密事件，可以获取预设时间段内与过密事件匹配的人数监测数据；并基于预设时间段内与过密事件匹配的人数监测数据，确定至少一个监控设备的、与过密事件匹配的人流状态数据。
在本公开的一些实施例中，在确定至少一个监控设备的人流状态数据之后，还包括：在人流状态数据满足告警条件的情况下，生成人流状态告警信息。
这里，判断至少一个监控设备的人流状态数据是否满足告警条件，若满足，则生成人流状态告警信息，以便用户可以基于人流状态告警信息，生成疏导计划，降低目标监控区域发生踩踏、拥堵等事件的概率。
这里，在确定的人流状态数据满足告警条件时，生成人流状态告警信息，基于生成的人流状态告警信息，可以对目标监控区域进行调控，降低安全事故的发生率，提高目标监控区域下人流的安全性。
下述分别对跨线事件的告警过程和过密事件的告警过程进行详细说明。
首先对跨线事件的告警过程进行说明。
在监测事件为跨线事件的情况下，获取预设时间段内与监测事件匹配的人数监测数据，包括：获取预设时间段内不同采集时间点的进人流数量和出人流数量，其中，进人流数量是指在不同采集时间点，沿预先绘制的进方向跨越预先绘制的进出界线的人数；出人流数量是指在不同采集时间点，沿预先绘制的出方向跨越预先绘制的进出界线的人数。
这里，跨线事件对应的监测标识中可以包括预先设置的进出界线和进出方向(进方向和出方向至少之一，出方向为进方向的反方向)，进出界线可以将监控视频对应的监测区域划分为进区域和出区域，进出方向中的进方向可以为从出区域进入进区域的方向，进出方向中的出方向可以为从进区域进入出区域的方向。
进而可以基于设置的进出界线、进出方向和监控视频，确定监控视频中预设时间段内每个采集时间点的进人流数量(即入人流数量)和出人流数量，不同采集时间点的进人流数量是指在不同采集时间点，沿进方向跨越进出界线的人数；不同采集时间点的出人流数量是指在不同采集时间点，沿出方向跨越进出界线的人数。
在本公开的一些实施例中，可以利用训练好的目标追踪算法，基于设置的监测标识对监控视频进行检测，在预设时间段内，每间隔预设时间输出一次检测结果，预设时间段内的多次检测结果可以为预设时间段内不同采集时间点的进人流数量和出人流数量，每个检测结果关联有输出时间(该输出时间为采集时间点)，进而可以获取预设时间段内不同采集时间点的进人流数量和出人流数量。
上述方法中，在监测事件为跨线事件时，可以获取预设时间段内不同采集时间点的进人流数量和出人流数量，能够为后续确定跨线事件对应的人流状态数据提供数据支持。
在本公开的一些实施例中，在监控点位为一个的情况下，基于人数监测数据，确定至少一个监控设备的人流状态数据，包括：基于进人流数量和出人流数量，确定监控点位对应的监测区域中预设时间段内的总进人流数量和总出人流数量。
在人流状态数据满足告警条件的情况下，生成人流状态告警信息，包括：在满足以下条件之一的情况下，生成人流状态告警信息：总进人流数量大于设置的第一人流阈值，且总出人流数量大于设置 的第二人流阈值；总进人流数量大于设置的第一人流阈值，或，总出人流数量大于设置的第二人流阈值。
在获取了预设时间段内与监测事件(跨线事件)匹配的人数监测数据之后，即针对跨线事件，在获取了预设时间段内不同采集时间点的进人流数量和出人流数量之后，可以基于预设时间段内不同采集时间点的进人流数量和出人流数量，确定该监控点位对应的监测区域中预设时间段内的总进人流数量和总出人流数量。
承接上述实施例继续说明，训练好的目标追踪算法可以每3秒输出一次检测结果(每3秒确定一个采集时间点)，该检测结果可以为该3秒内的进人流数量和出人流数量，例如，检测结果可以为：08时10分01秒至08时10分03秒(包括10分01秒和10分03秒)之间的进人流数量为20、出人流数量为50，关联的输出时间(采集时间点)为08时10分03秒；进而可以得到预设时间段内的多次检测结果，即得到预设时间段内不同采集时间点的进人流数量和出人流数量。
在得到了预设时间段内不同采集时间点的进人流数量和出人流数量之后，可以将不同采集时间点的进人流数量进行相加，得到预设时间段内的总进人流数量；以及可以将不同采集时间点的出人流数量进行相加，得到预设时间段内的总出人流数量。
这里，第一人流阈值、第二人流阈值为预先设置的，第一人流阈值和第二人流阈值可以根据实际需要进行设置。在得到预设时间段内的总进人流数量和总出人流数量之后，可以判断该预设时间段内的总进人流数量是否大于设置的第一人流阈值，和/或，判断该预设时间段内的总出人流数量是否大于设置的第二人流阈值。
在判断该预设时间段内的总进人流数量是否大于设置的第一人流阈值，以及判断该预设时间段内的总出人流数量是否大于设置的第二人流阈值的情况下，若预设时间段内的总进人流数量大于设置的第一人流阈值，和/或，若在预设时间段内的总出人流数量大于设置的第二人流阈值时，生成人流状态告警信息。生成的人流状态告警信息可以为文字、语音、视频等格式的信息，比如，生成的人流状态告警信息可以为“注意，进人流数量较大”。这种情况下，人流状态告警信息的告警事件类型为：点位越线告警。
在本公开的一些实施例中，可以在触发生成的人流状态告警信息之后，可以显示该人流状态告警的详细信息，详细信息包括但不限于告警点位(即告警的监控设备的名称等)、告警时间、告警事件类型，在告警事件类型为点位越线告警时，详细信息还包括该单位时间内的进人流数量、出人流数量等。
这里，在监控点位为一个时，基于预设时间段内不同采集时间点的进人流数量和出人流数量，确定监控点位对应的监测区域中预设时间段内的总进人流数量和总出人流数量。在预设时间段内的总进人流数量大于设置的第一人流阈值，和/或，在预设时间段内的总出人流数量大于设置的第二人流阈值的情况下，生成人流状态告警信息，能够实现对该监控视频的进人流数量和出人流数量的预警，以便基于生成的人流状态告警信息进行人流的疏导，降低短时间内进人流数量较多，或者出人流数量较多造成的安全事故的发生率。
在本公开的一些实施例中，在监控点位为一个的情况下，基于人数监测数据，确定至少一个监控设备的人流状态数据，包括：基于进人流数量和出人流数量，确定监控点位对应的监测区域中的进人流速度和出人流速度。
在本公开的一些实施例中，还可以基于预设时间段内不同采集时间点的进人流数量和出人流数量，确定监控点位对应的监测区域中的进人流速度和出人流速度。
在本公开的一些实施例中，在得到了预设时间段内不同采集时间点的进人流数量和出人流数量之后，可以对多次检测结果按照输出时间(采集时间点)进行归类、并整合，得到单位时间内(比如一分钟)的进人流数量和出人流数量，可以得到进人流速度和出人流速度。
比如，可以将输出时间为08时10分00秒至08时11分00秒(不包括08时10分00秒、包括08时11分00秒)之内的输出结果进行归类并整合，即将输出时间为08时10分03秒、08时10分06秒、……、08时10分57秒、08时11分00秒得到的输出结果划分为一类，并将该类内的检测结果进行整合，得到08时10分 00秒至08时11分00秒之间的1分钟内(单位时间内)的进人流数量和出人流数量，即得到了08时10分对应的进人流速度(单位：人/分)和出人流速度(单位：人/分)。
上述方法中，可以基于预设时间段内不同采集时间点的进人流数量和出人流数量，确定监控点位对应的监测区域中的进人流速度和出人流速度，实现对进人流速度和出人流速度的监测，降低进人流速度较大，或者出人流速度较大造成的安全事故的发生率。
参见图4所示的一种展示人流状态告警的详细信息的界面示意图，该图4中包括告警详情、当天跨线事件时段统计，告警详情包括告警点位、事件类型、告警时间、持续时长(跨线事件持续的时间)、入流峰值、出流峰值等，当前跨线事件时段统计包括从当天的零点至统计的当前时间之间的越线事件告警。该图中还包括视频画面截图，该视频画面截图上显示有当前时刻对应的出人流信息(出人流数量和出人流速度)和入人流信息(入人流数量和入人流速度)；视频画面截图下方显示有多帧告警图片，其中，告警图片的数量与告警的持续时长相关，比如，跨线事件的持续时长为17分钟时，可以每间隔一分钟提取一帧告警图片，作为告警记录，即可以在视频画面截图下方显示17帧告警图片。
在本公开的一些实施例中，还可以将监控设备的名称、安装位置、采集的监控视频等点位信息、以及单位时间内的出人流数量、进人流数量等信息持久化存储在搜索服务器(比如，elasticsearch)中，以便后续搜索查询。
在本公开的一些实施例中，在监控点位为多个的情况下，基于人数监测数据，确定至少一个监控设备的人流状态数据，包括：
步骤一、针对每个监控点位，基于进人流数量和出人流数量，确定每个监控点位分别对应的监测区域中预设时间段内的总进人流数量和总出人流数量。
步骤二、基于预设时间段内目标监控区域的历史人数，以及每个监控点位分别对应的预设时间段内的总进人流数量和总出人流数量，确定目标监控区域内的人员净存量。
在人流状态数据满足告警条件的情况下，生成人流状态告警信息，包括：在人员净存量大于设置的净存量阈值的情况下，生成人流状态告警信息。
这里，考虑到一个场地或场所可能设置多个监控设备，故可以对多个监控设备分别采集的监控视频进行人流分析，得到多个监控视频的人流状态数据。其中，多个监控设备分别采集的监控视频构成了视频组，即可以对视频组进行人流分析，得到该视频组对应的人流状态数据。在本公开的一些实施例中，可以通过触发展示界面上设置的视频组对应的跨线事件的开启按钮，开启视频组内每个监控视频的跨线分析功能。同时，还可以在展示界面上设置人流总存量分级预警的具体信息，比如，填写存量趋多对应的一级预警人数、存量警示对应的二级预警人数、和存量过热对应的三级预警人数。
在本公开的一些实施例中，针对每个监控点位，在获取了该监控点位对应的预设时间段内不同采集时间点的进人流数量和出人流数量之后，可以将不同采集时间点的进人流数量相加，得到该监控点位对应的预设时间段内的总进人流数量；以及可以将不同采集时间点的出人流数量相加，得到该监控点位对应的预设时间段内的总出人流数量。在本公开的一些实施例中可以得到各个监控点位分别对应的预设时间段内的总进人流数量和总出人流数量。
在本公开的一些实施例中，可以得到每个监控设备对应的08时11分00秒的时间点至08时12分00秒的时间点内的总进人流数量和总出人流数量，08时11分00秒的时间点至08时12分00秒的时间点之间的时间段即为预设时间段。
在本公开的一些实施例中，基于预设时间段内目标监控区域的历史人数，以及每个监控点位分别对应的预设时间段内的总进人流数量和总出人流数量，确定目标监控区域内的人员净存量。比如，针对视频组内的每个监控视频，可以将该监控视频分别对应的预设时间段的总进人流数量与总出人流数量相减，得到该监控视频在该预设时间段内的人流变化量，将各个监控视频分别对应的预设时间段内的人流变化量相加，得到视频组对应的总人流变化量(即视频组对应的场地或场所对应的总人流变化量)，再将视频组对应的总人流变化量与预设时间段内目标监控区域的历史人数相加，得到目标监控区域内的人员净存量(即得到视频组对应的场地或场所对应的当前时间点的当前人数)。
在本公开的一些实施例中，预设时间段可以为08时11分00秒的时间点至08时12分00秒的时间点之 间的时间段，再可以得到08时11分00秒的时间点对应的当前人数(即预设时间段内目标监控区域的历史人数)，并可以得到视频组内每个监控视频对应的08时11分00秒至08时12分00秒(预设时间段)内的总进人流数量和总出人流数量，在基于预设时间段内目标监控区域的历史人数(即得到的08时11分00秒的时间点的人员净存量)、以及视频组内每个监控视频对应的08时11分00秒至08时12分00秒内的总进人流数量和总出人流数量，确定目标监控区域中08时12分00秒的人员净存量。
在得到目标监控区域内的人员净存量之后，可以对目标监控区域内的人员净存量进行监测，在监测到目标监控区域内的人员净存量大于预先设置的净存量阈值的时，生成人流状态告警信息。比如，生成的人流状态告警信息可以为“注意，当前时间xx场地人员净存量较多”。这种情况下，人流状态告警信息的告警事件类型为：视频组跨线告警。
在本公开的一些实施例中，针对视频组跨线告警，可以设置多级告警风险，比如，多级告警风险包括：存量趋多、存量警示、存量过热，针对不同的告风险设置不同的净存量阈值，例如，存量趋多对应的净存量阈值可以为100，存量警示对应的净存量阈值可以为200，存量过热对应的净存量阈值可以为500。针对不同的告警风险，可以设置不同的人流状态告警信息。比如，存量趋多对应的人流状态告警信息可以为：文字格式的告警信息；存量警示对应的人流状态告警信息可以为：语音格式的告警信息；存量过热对应的人流状态告警信息可以为：视频格式的告警信息。
在本公开的一些实施例中，可以在触发生成的人流状态告警信息之后，可以显示该人流状态告警的详细信息，详细信息包括但不限于告警点位(即告警的监控设备的名称等)、告警时间、告警事件类型，在告警时间类型为视频组跨线告警时，详细信息还可以包括：当前时间点的人员净存量。
这里，在确定每个监控点位对应的监测区域中预设时间段内的总进人流数量和总出人流数量之后，可以基于预设时间段内目标监控区域的历史人数，以及多个监控点位分别对应的预设时间段内的总进人流数量和总出人流数量，确定目标监控区域内的人员净存量，并在目标监控区域内的人员净存量大于设置的净存量阈值的情况下，生成人流状态告警信息，能够实现对目标监控区域中的人员净存量的预警，以便在人员净存量较多时，基于生成的人流状态告警信息进行人员的疏导，降低目标监控区域中人员较多时造成安全事故的发生率。
参见图5A所示的一种展示人流状态告警的详细信息的界面示意图，图5A中以地图模式展示了人流状态告警的详细信息；以及参见图5B所示的另一种展示人流状态告警的详细信息的界面示意图，图5B中以列表模式展示了人流状态告警的详细信息，其中，图5B中展示的列表中包括过密事件和越线事件。在本公开的一些实施例中，在触发图5A中显示的越线事件的信息之后，或者，在触发图5B中显示的越线事件的信息之后，可以展示图5C中显示的告警详情，图5C中显示的告警详情包括分组名称(即视频组对应的名称)、事件类型、告警时间、持续时长、人流总存量峰值、当天人流总存量统计。
其次，可以对过密事件的告警过程进行详细说明。
在本公开的一些实施例中，在监测事件为过密事件的情况下，获取预设时间段内与监测事件匹配的人数监测数据，包括：统计预设时间段内不同采集时间点的目标对象的个数。
这里，在监控视频中存在过密事件对应的监测标识时，可以基于过密事件对应的监测标识和监控视频，对监测标识对应的检测区域中的目标对象(人类)进行检测，得到各个采集时间点时检测区域内的目标对象的数量。在监控视频中不存在过密时间对应的监测标识时，则认为监控视频的整个监控画面均为检测区域，可以对监控视频进行检测，得到各个采集时间点时检测区域中的目标对象的数量。
在本公开的一些实施例中，可以利用训练好的用于识别目标对象的深度学习算法，对监控视频中的检测区域进行检测，实时的输出检测结果，检测结果可以为监控视频中每个采集时间点时检测区域内的人数。其中，深度学习算法可以周期性的输出检测结果，比如，深度学习算法可以每秒输出一次检测结果，或者，还可以每两秒输出一次检测结果等。比如，检测结果可以为：08时10分00秒(采集时间点)时检测区域内的人数为50；08时10分01秒时检测区域内的人数为54等。
在本公开的一些实施例中，可以统计预设时间段内不同采集时间点的目标对象的个数，比如，预设时间段为08时10分00秒至08时11分00秒，将预设时间点内的每秒时间点作为一次采集时间点，即可 以统计08时10分00秒(采集时间点1)的目标对象的个数、08时10分01秒(采集时间点2)的目标对象的个数、……、08时10分59秒(采集时间点60)的目标对象的个数等。
上述方法中，在监测事件为过密事件时，可以统计预设时间段内不同采集时间点的目标对象的个数，能够为后续确定过密事件对应的人流状态数据提供数据支持。
在本公开的一些实施例中，在监控点位为一个的情况下，基于人数监测数据，确定至少一个监控设备的人流状态数据，包括：基于目标对象的个数，确定监控点位对应的监测区域在预设时间段内的平均人数。
在人流状态数据满足告警条件的情况下，生成人流状态告警信息，包括：在平均人数大于设置的第一人数阈值的情况下，生成人流状态告警信息。
在本公开的一些实施例，可以将预设时间段内不同采集时间点的目标对象的个数求平均，得到监控点位对应的监测区域在预设时间段内的平均人数。并将该预设时间段内的平均人数进行监测，在该平均人数大于设置的第一人数阈值时，生成人流状态告警信息。其中，预设时间段的长度可以根据需要进行设置，比如，预设时间段的长度可以为5秒、10秒、60秒、5分钟等。跨线事件对应的预设时间段与过密事件对应的预设时间段的长度可以相同，也可以不同。
在本公开的一些实施例中，预设时间段内不同采集时间点的目标对象的个数包括：08时10分01秒时目标对象的个数为50、08时10分02秒时目标对象的个数为53、08时10分03秒时目标对象的个数为52、08时10分04秒时目标对象的个数为51、08时10分05秒时目标对象的个数为54，则可以将5次检测结果取平均值，得到平均值为52，确定08时10分01秒至08时10分05秒内，监控点位对应的监测区域的平均人数为52。
在本公开的一些实施例中，可以对该监控点位对应的监测区域在预设时间段内的平均人数进行监测，在该平均人数大于设置的第一人数阈值时，生成人流状态告警信息。比如，生成的人流状态告警信息可以为“注意，当前时间xx区域人数较多”。这种情况下，人流状态告警信息的告警事件类型为：点位过密告警。
在本公开的一些实施例中，可以在触发生成的人流状态告警信息之后，可以显示该人流状态告警的详细信息，详细信息包括但不限于告警点位(即告警的监控设备的名称等)、告警时间、告警事件类型，在告警事件类型为点位过密告警时，详细信息还可以包括：当前时间点的检测区域内的实时人数。
参见图6所示的一种展示人流状态告警的详细信息的界面示意图，该图6中包括告警详情、当天过密事件时段统计，告警详情包括告警点位、事件类型、告警时间、过密时长、人数峰值、密度峰值等，当前过密事件时段统计包括从当天的零点至统计的当前时间之间的过密事件告警。该图6中还包括视频画面截图，以及在视频画面截图下方显示有多帧告警图片，其中，告警图片的数量与过密事件的持续时长相关，比如，过密事件的持续时长为17分钟时，可以每间隔一分钟提取一帧告警图片，作为告警记录，即可以在视频画面截图下方显示17帧告警图片。
在本公开的一些实施例中，还可以将监控设备的名称、安装位置、采集的监控视频等点位信息、以及该监控设备每分钟内的实时人数、实时人数的最大值、实时人数的最小值等信息持久化关联存储在搜索服务器(比如，elasticsearch)中，以便后续搜索查询。
上述方法中，在监控点位为一个时，基于预设时间段内不同采集时间点的目标对象的个数，确定监控点位对应的监测区域在预设时间段内的平均人数；并在监控点位对应的监测区域在预设时间段内的平均人数大于设置的第一人数阈值的情况下，生成人流状态告警信息，能够实现对该监控视频的检测区域中的平均人数的监控，以便基于生成的人流状态告警信息对检测区域进行人流疏导，降低检测区域内的人员较为密集时，造成的安全事故的发生率。
在本公开的一些实施例中，在监控点位为多个的情况下，基于人数监测数据，确定至少一个监控设备的人流状态数据，包括：
步骤一、针对每个监控点位，基于目标对象的个数，确定监控点位对应的监测区域在预设时间段内的平均人数。
步骤二、基于平均人数，确定目标监控区域中的总实时人数。
在人流状态数据满足告警条件的情况下，生成人流状态告警信息，包括：在总实时人数大于设置的第二人数阈值的情况下，生成人流状态告警信息。
这里，针对每个监控点位，可以基于预设时间段内不同采集时间点的目标对象的个数，确定该监控点位对应的监测区域在预设时间段内的平均人数；再可以将多个监控点位分别对应的平均人数相加，确定目标监控区域中的总实时人数。
在确定了目标监控区域中的总实时人数之后，可以对该总实时人数进行监测，在确定目标监控区域中的总实时人数大于设置的第二人数阈值时，生成人流状态告警信息。
在本公开的一些实施例中，可以通过触发展示界面上设置的视频组对应的过密事件的开启按钮，开启视频组内每个监控视频的人流过密分析功能。同时，还可以在展示界面上设置实时总人数分级预警的具体信息，比如，填写一般风险对应的一级预警人数、较大风险对应的二级预警人数、和重大风险对应的三级预警人数。
这里，在监控设备包括多个时，多个监控设备分别采集的监控视频构成了视频组。针对每个监控设备采集的监控视频(即针对视频组内的每个监控视频)，可以利用训练好的用于识别目标对象的深度学习算法，对监控视频中监测标识指示的检测区域进行检测，实时的输出检测结果，检测结果可以为监控视频中采集时间点的检测区域内目标对象的个数。在本公开的一些实施例中，可以基于周期性得到的检测结果，确定监控点位对应的监测区域在预设时间段内的平均人数。
在得到视频组内每个监控视频对应的平均人数之后，可以将视频组内包括的各个监控视频对应的平均人数相加，确定目标监控区域中的总实时人数。进而可以对目标监控区域中的总实时人数进行监测，在目标监控区域中的总实时人数大于设置的第二人数阈值的情况下，生成人流状态告警信息。比如，生成的人流状态告警信息可以为“注意，当前时间xx场景总人数较多”。这种情况下，人流状态告警信息的告警事件类型为：视频组过密告警。
在本公开的一些实施例中，针对视频组过密告警，可以设置多级告警风险，比如，多级告警风险包括：一级风险、较大风险、重大风险，针对不同的告警风险设置不同的第二人数阈值，例如，一级风险对应的第二人数阈值可以为100，二级风险对应的第二人数阈值可以为200，重大风险对应的第二人数阈值可以为500。针对不同的告警风险，可以设置不同的人流状态告警信息。比如，一级风险对应的人流状态告警信息可以为：文字格式的告警信息；二级风险对应的人流状态告警信息可以为：语音格式的告警信息；三级风险对应的人流状态告警信息可以为：视频格式的告警信息。
在本公开的一些实施例中，可以在触发生成的人流状态告警信息之后，可以显示该人流状态告警的详细信息，详细信息包括但不限于告警点位(即告警的监控设备的名称等)、告警时间、告警事件类型，在告警事件类型为视频组过密告警时，详细信息还可以包括：现实场景的总实时人数。
参见图7A所示的一种展示人流状态告警的详细信息的界面示意图，图7A中以地图模式展示了人流状态告警的详细信息；以及参见图5B所示的另一种展示人流状态告警的详细信息的界面示意图，图5B中以列表模式展示了人流状态告警的详细信息，其中，图5B中展示的列表中包括过密事件和越线事件。在本公开的一些实施例中，在触发图7A中显示的过密事件的信息之后，或者，在触发图5B中显示的过密事件的信息之后，可以展示图7B中显示的告警详情，图7B中显示的告警详情包括分组名称(即视频组对应的名称)、事件类型、告警时间、持续时长、人数峰值、密度峰值、当天实时总人数统计、视频源统计。
这里，在基于至少一个监控设备采集的监控视频，以及预先绘制的与视频画面中的目标位置匹配的监测标识，确定现实场景中的人流状态数据之后，还可以生成人流状态数据随着时间的变化示意图，以便对当天的人流状态数据进行直观展示。在本公开的一些实施例中，人流状态数据随时间的变化示意图包括实时总人数随时间的第一变化示意图，第一变化示意图中包括人数峰值随时间的变化关系、和人数谷值随时间的变化关系；和/或，人流总存量随时间的第二变化示意图；第二变化示意图中包括出总人流随时间的变化关系、入总人流随时间的变化关系、和人流总存量随时间的变化关系。其中，第一变化图、第二变化图设置的时间间隔可以为5分钟、10分钟、30分钟、1小时等。
同时，图8A示出了本公开实施例提供的一种实时总人数和人流总存量随时间变化的界面示意图；其中，图8A中分别给出实时总人数随时间变化的界面示意图、以及人流总存量随时间变化的界面示意图。同时图8A中可以查看到过密事件中的人数峰值和人数谷值，其中，人数峰值即某个时间段内，实时人数的最高值，以及某个时间段内，实时人数的最底值等，能够实现对监控视频的监测，以便用户能够实时监控人流状态数据。
同时图8B示出了本公开实施例提供的一种预测未来时间段的人流数据的界面示意图；图8B中给出时间段为20年04月16日1时至20年04月16日11时中每一小时的入总人流、出总人流以及人流总存量，并基于每个小时分别对应的入总人流、出总人流以及人流总存量，构成在未来时间段即图8B中预测区域内的人流状态数据，如此，能够基于人流状态数据在未来日期内的预测数据，生成人流疏导计划。
上述方法中，在确定了每个监控点位对应的监测区域在预设时间段内的平均人数之后，可以基于多个监控点位分别对应的平均人数，确定目标监控区域中的总实时人数；并在确定目标监控区域中的总实时人数大于设置的第二人数阈值的情况下，生成人流状态告警信息，能够实现对目标监控区域中的总实时人数的预警，以便在总实时人数较多时，基于生成的人流状态告警信息对目标监控区域中的人员进行疏导，降低目标监控区域中总实时人数较多时造成安全事故的发生率。
在本公开的一些实施例中，所述方法还包括：将最近多个历史日期内同一采集时间点的人流状态数据求平均，得到每个采集时间点对应的预测人流状态数据；基于预测人流状态数据，构成人流状态数据在未来日期内的预测数据；其中，预测数据用于生成人流疏导计划。
这里，多个历史周期可以根据需要进行设置，比如，多个历史周期可以为最近7天(一个历史周期对应一天)内的人流状态数据，即在10月8日00点00分时，可以获取10月1日至10月7日(7个历史周期)的人流状态数据，将最近7个历史日期内同一采集时间点的人流状态数据求平均，得到每个采集时间点对应的预测人流状态数据；各个采集时间点分别对应的预测人流状态数据，构成人流状态数据在未来日期内的预测数据。
比如，可以将10月1日至10月7日的人流状态数据，在同一采集时间点求平均值，得到各个采集时间点的平均值，该平均值为该采集时间点对应的预测人流状态数据；各个采集时间点分别对应的预测人流状态数据，构成人流状态数据在未来日期内的预测数据。比如，生成总进人流数量在未来日期(未来一天)内的预测数据、生成总出人流数量在未来日期(未来一天)内的预测数据、和生成人员净存量在未来日期(未来一天)内的预测数据。
进而可以基于人流状态数据在未来日期内的预测数据，生成人流疏导计划，比如，若在预测数据中可知，15点时总实时人数最多，则可以在15点时控制进入目标监控区域的人流数量。
在实际的应用场景中，该方法可以应用于商场、大厅等场景中。下述以商场为例分别对一个监控视频的越线事件和视频组的越线事件进行说明，假设商场有两个门，则可以在每个门口位置处(监控点位)设置一个监控设备，即监控设备一(监控点位一处设置的监控设备一)采集大门A的监控视频，监控设备二(监控点位二处设置的监控设备二)采集大门B的监控视频，该监控设备一、监控设备二可以对该进出该门的行人进行监测。
在本公开的一些实施例中，可以获取监控设备一采集的监控视频一，以及获取监控设备二采集的监控视频二。针对监控视频一，在监控视频一的视频画面截图上绘制进出界线和进出方向，确定监控视频一中监控点位对应的监测区域中预设时间段内的总进人流数量和总出人流数量；进而在预设时间段内的总进人流数量大于设置的第一人流阈值，和预设时间段内的总出人流数量大于设置的第二人流阈值至少之一的情况下，生成人流状态告警信息。以及针对监控视频二，在监控视频二的视频画面截图上设置进出界线和进出方向，确定监控视频二中监控点位对应的监测区域中预设时间段内的总进人流数量和总出人流数量。进而在预设时间段内的总进人流数量大于设置的第一人流阈值，和在预设时间段内的总出人流数量大于设置的第二人流阈值至少之一的情况下，生成人流状态告警信息。
同时，监控视频一和监控视频二构成了视频组，可以对视频组进行分析，确定监控设备一和监控设备二对应的目标监控区域中的人流状态数据。在本公开的一些实施例中，针对监控视频一，确定监控点位一对应的监测区域中预设时间段内的总进人流数量和总出人流数量；针对监控视频二，确定监 控点位二对应的监测区域中预设时间段内的总进人流数量和总出人流数量。进而，基于预设时间段内目标监控区域的历史人数，以及多个监控点位分别对应的预设时间段内的总进人流数量和总出人流数量，确定目标监控区域内的人员净存量。即确定了该商场内的人员净存量。在人员净存量大于设置的净存量阈值的情况下，生成人流状态告警信息，以便在接收到人流状态告警信息之后，可以对商场中的行人进行调控，降低拥堵事件的发生率。同时还可以获取当前场所当天每个时间段的人员净存量和实时总人数曲线图，从而能够动态的总结规律合理控制人数，调整商场工作人员工作时间以及调整不同时间段工作人员数量等。
下述以大厅为例分别对一个监控视频的过密事件和视频组的过密事件进行说明。假设在大厅的四个角落(四个监控点位)分别设置有监控设备，即在四个监控设备对大厅的四个检测区域进行检测，四个监控设备中各个监控设备采集的监控视频构成了视频组。
在本公开的一些实施例中，针对视频组中的每个监控视频，基于预设时间段内不同采集时间点的目标对象的个数，确定监控点位对应的监测区域在预设时间段内的平均人数，在该监控视频对应的平均人数大于设置的第一人数阈值时，生成该监控视频对应的人流状态告警信息。即能够实现针对视频组中每个监控视频的过密事件监测。
同时，可以在视频画面截图上绘制监测标识，即监测标识对应的区域为检测区域；也可以不在视频画面截图上绘制监测标识，即监控视频不存在对应的基准面标识，此时，默认整个视频画面均为检测区域。
同时，可以对视频组进行过密事件监测，确定视频组对应的目标监控区域中的总实时人数。在本公开的一些实施例中，针对视频组内的每个监控视频，确定监控点位对应的监测区域在预设时间段内的平均人数；并基于四个监控点位分别对应的平均人数，确定目标监控区域中的总实时人数。即确定了该大厅内的多个检测区域的总实时人数。在目标监控区域中的总实时人数大于设置的第二人数阈值的情况下，生成人流状态告警信息，以便在接收到人流状态告警信息之后，可以对大厅中的密集区域进行疏导，降低人员密集造成的事故发生率。
本领域技术人员可以理解，在本公开实施例提供的的上述方法中，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。
基于相同的构思，本公开实施例还提供了一种场景监测装置，参见图9所示，为本公开实施例提供的场景监测装置900的架构示意图，包括第一获取模块901、检测模块902、第二获取模块903、确定模块904，具体包括：
第一获取模块901，配置为获取设置于至少一个监控点位的监控设备采集的监控视频；
检测模块702，配置为基于监控视频，确定至少一个监控点位对应的监测区域是否发生监测事件；
第二获取模块703，配置为在至少一个监控点位对应的监测区域发生监测事件的情况下，获取预设时间段内与监测事件匹配的人数监测数据；
确定模块704，配置为基于人数监测数据，确定至少一个监控设备的人流状态数据。
在本公开的一些实施例中，在确定至少一个监控设备的人流状态数据之后，场景监测装置900还包括：告警模块905，配置为在人流状态数据满足告警条件的情况下，生成人流状态告警信息。
在本公开的一些实施例中，在监测事件为跨线事件的情况下，检测模块902，配置为基于监控视频确定至少一个监控点位对应的监测区域内，是否存在跨越与预先绘制的进出界线匹配的目标位置的目标对象；在存在目标对象的情况下，确定至少一个监控点位对应的监测区域发生跨线事件。
在本公开的一些实施例中，在监测事件为跨线事件的情况下，第二获取模块903，配置为获取预设时间段内不同采集时间点的进人流数量和出人流数量；其中，进人流数量是指在不同采集时间点，沿预先绘制的进方向跨越预先绘制的进出界线的人数；出人流数量是指在不同采集时间点，沿预先绘制的出方向跨越预先绘制的进出界线的人数。
在本公开的一些实施例中，在监控点位为一个的情况下，确定模块904，配置为基于进人流数量和出人流数量，确定监控点位对应的监测区域中预设时间段内的总进人流数量和总出人流数量；
告警模块905，配置为在满足以下条件之一的情况下，生成人流状态告警信息：总进人流数量大于设置的第一人流阈值，且总出人流数量大于设置的第二人流阈值；总进人流数量大于第一人流阈值，或，总出人流数量大于第二人流阈值。
在本公开的一些实施例中，在监控点位为一个的情况下，确定模块904，配置为基于进人流数量和出人流数量，确定监控点位对应的监测区域中的进人流速度和出人流速度。
在本公开的一些实施例中，在监控点位为多个的情况下，确定模块904，配置为针对每个监控点位，基于进人流数量和出人流数量，确定监每个控点位对应的监测区域中预设时间段内的总进人流数量和总出人流数量；基于预设时间段内目标监控区域的历史人数，以及每个监控点位分别对应的预设时间段内的总进人流数量和总出人流数量，确定目标监控区域内的人员净存量；
告警模块905，配置为在人员净存量大于设置的净存量阈值的情况下，生成人流状态告警信息。
在本公开的一些实施例中，在监测事件为过密事件的情况下，检测模块902，配置为基于监控视频，确定至少一个监控点位对应的监测区域内的目标对象个数是否超过过密阈值；在目标对象个数超过过密阈值的情况下，确定至少一个监控点位对应的监测区域发生过密事件。
在本公开的一些实施例中，在监测事件为过密事件的情况下，第二获取模块903，配置为统计预设时间段内不同采集时间点的目标对象的个数。
在本公开的一些实施例中，在监控点位为一个的情况下，确定模块904，配置为基于目标对象的个数，确定监控点位对应的监测区域在预设时间段内的平均人数；
告警模块905，配置为在平均人数大于设置的第一人数阈值的情况下，生成人流状态告警信息。
在本公开的一些实施例中，在监控点位为多个的情况下，确定模块904，配置为针对每个监控点位，基于目标对象的个数，确定监控点位对应的监测区域在预设时间段内的平均人数；基于平均人数，确定目标监控区域中的总实时人数；
告警模块905，配置为在总实时人数大于设置的第二人数阈值的情况下，生成人流状态告警信息。
在本公开的一些实施例中，场景监测装置900还包括：预警模块906，配置为将最近多个历史日期内同一采集时间点的人流状态数据求平均，得到每个采集时间点对应的预测人流状态数据；基于预测人流状态数据，构成人流状态数据在未来日期内的预测数据；其中，预测数据用于生成人流疏导计划。
在本公开的一些实施例中，本公开实施例提供的装置具有的功能或包含的模板可以配置为执行上文方法实施例描述的方法，其具体实现可以参照上文方法实施例的描述，为了简洁，这里不再赘述。
基于同一技术构思，本公开实施例还提供了一种电子设备1900。参照图10所示，为本公开实施例提供的电子设备1900的结构示意图，包括处理器1901、存储器1902、和总线1903。其中，存储器1902用于存储执行指令，包括内存1921和外部存储器1922；这里的内存1921也称内存储器，用于暂时存放处理器1901中的运算数据，以及与硬盘等外部存储器1922交换的数据，处理器1901通过内存1921与外部存储器1922进行数据交换，当电子设备1900运行时，处理器1901与存储器1902之间通过总线1903通信，使得处理器1901执行以下指令：
获取设置于至少一个监控点位的监控设备采集的监控视频；
基于监控视频确定，至少一个监控点位对应的监测区域是否发生监测事件；
在至少一个监控点位对应的监测区域发生监测事件的情况下，获取预设时间段内与监测事件匹配的人数监测数据；
基于人数监测数据，确定至少一个监控设备的人流状态数据。
此外，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场景监测方法。
本公开实施例还提供一种计算机程序，该计算机程序包括计算机可读代码，在计算机可读代码在电子设备中运行的情况下，电子设备中的处理器执行用于实现上述任一所述的场景监测方法。
本公开实施例还提供给另一种计算机程序产品，包括存储了程序代码的计算机可读存储介质，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场景监测方法，具体可参见上述方法实施例，在此不再赘述。
本公开实施例中涉及的设备可以是系统、方法和/或计算机程序产品。计算机程序产品可以包括计算机可读存储介质，其上载有用于使处理器实现本公开的各个方面的计算机可读程序指令。
计算机可读存储介质可以是可以保持和存储由指令执行设备使用的指令的有形设备。计算机可读存储介质例如可以是但不限于电存储设备、磁存储设备、光存储设备、电磁存储设备、半导体存储设备或者上述的任意合适的组合。计算机可读存储介质的更具体的例子(非穷举的列表)包括：便携式计算机盘、硬盘、随机存取存储器(Random Access Memory，RAM)、只读存储器(Read-Only Memory，ROM)、可擦除可编程只读存储器(Electrical Programmable Read Only Memory，EPROM)或闪存、静态随机存取存储器(Static Random-Access Memory，SRAM)、便携式压缩盘只读存储器(Compact Disc Read-Only Memory，CD-ROM)、数字多功能盘(Digital Video Disc，DVD)、记忆棒、软盘、机械编码设备、例如其上存储有指令的打孔卡或凹槽内凸起结构、以及上述的任意合适的组合。这里所使用的计算机可读存储介质不被解释为瞬时信号本身，诸如无线电波或者其他自由传播的电磁波、通过波导或其他传输媒介传播的电磁波(例如，通过光纤电缆的光脉冲)、或者通过电线传输的电信号。
这里所描述的计算机可读程序指令可以从计算机可读存储介质下载到各个计算/处理设备，或者通过网络、例如因特网、局域网、广域网和/或无线网下载到外部计算机或外部存储设备。网络可以包括铜传输电缆、光纤传输、无线传输、路由器、防火墙、交换机、网关计算机和/或边缘服务器。每个计算/处理设备中的网络适配卡或者网络接口从网络接收计算机可读程序指令，并转发该计算机可读程序指令，以供存储在各个计算/处理设备中的计算机可读存储介质中。
用于执行本公开操作的计算机程序指令可以是汇编指令、指令集架构(Industry Standard Architecture，ISA)指令、机器指令、机器相关指令、微代码、固件指令、状态设置数据、或者以一种或多种编程语言的任意组合编写的源代码或目标代码，所述编程语言包括面向对象的编程语言—诸如Smalltalk、C++等，以及常规的过程式编程语言，诸如“C”语言或类似的编程语言。计算机可读程序指令可以完全地在用户计算机上执行、部分地在用户计算机上执行、作为一个独立的软件包执行、部分在用户计算机上部分在远程计算机上执行、或者完全在远程计算机或服务器上执行。在涉及远程计算机的情形中，远程计算机可以通过任意种类的网络，包括局域网(Local Area Network，LAN)或广域网(Wide Area Network，WAN)连接到用户计算机，或者，可以连接到外部计算机(例如利用因特网服务提供商来通过因特网连接)。在一些实施例中，通过利用计算机可读程序指令的状态信息来个性化定制电子电路，例如可编程逻辑电路、FPGA或可编程逻辑阵列(Programmable Logic Arrays，PLA)，该电子电路可以执行计算机可读程序指令，从而实现本公开的各个方面。
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，上述描述的系统和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，所揭露的系统、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者可以集成到另一个系统，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、ROM、RAM、磁碟或者光盘等各种可以存储程序代码的介质。
以上仅为本公开的具体实施方式，但本公开的保护范围并不局限于此，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，可轻易想到变化或替换，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以权利要求的保护范围为准。
工业实用性
本公开实施例提供了一种场景监测方法、装置、电子设备、存储介质及程序，该方法包括：获取设置于至少一个监控点位的监控设备采集的监控视频；基于所述监控视频，确定所述至少一个监控点位对应的监测区域是否发生监测事件；在所述至少一个监控点位对应的监测区域发生所述监测事件的情况下，获取预设时间段内与所述监测事件匹配的人数监测数据；基于所述人数监测数据，确定所述至少一个监控设备的人流状态数据。

Claims (16)

  1. 一种场景监测方法，所述方法由电子设备执行，所述方法包括：
    获取设置于至少一个监控点位的监控设备采集的监控视频；
    基于所述监控视频，确定所述至少一个监控点位对应的监测区域是否发生监测事件；
    在所述至少一个监控点位对应的监测区域发生所述监测事件的情况下，获取预设时间段内与所述监测事件匹配的人数监测数据；
    基于所述人数监测数据，确定所述至少一个监控设备的人流状态数据。
  2. 根据权利要求1所述的方法，其中，在确定所述至少一个监控设备的人流状态数据之后，还包括：
    在所述人流状态数据满足告警条件的情况下，生成人流状态告警信息。
  3. 根据权利要求1或2所述的方法，其中，在所述监测事件为跨线事件的情况下，基于所述监控视频，确定所述至少一个监控点位对应的监测区域是否发生监测事件，包括：
    基于所述监控视频确定所述至少一个监控点位对应的监测区域内，是否存在跨越与预先绘制的进出界线匹配的目标位置的目标对象；
    在存在所述目标对象的情况下，确定所述至少一个监控点位对应的监测区域发生所述跨线事件。
  4. 根据权利要求2所述的方法，其中，在所述监测事件为跨线事件的情况下，获取预设时间段内与所述监测事件匹配的人数监测数据，包括：
    获取所述预设时间段内不同采集时间点的进人流数量和出人流数量；其中，所述进人流数量是指在所述不同采集时间点，沿预先绘制的进方向跨越预先绘制的进出界线的人数；所述出人流数量是指在所述不同采集时间点，沿预先绘制的出方向跨越预先绘制的进出界线的人数。
  5. 根据权利要求4所述的方法，其中，在所述监控点位为一个的情况下，基于所述人数监测数据，确定所述至少一个监控设备的人流状态数据，包括：
    基于所述进人流数量和所述出人流数量，确定所述监控点位对应的监测区域中所述预设时间段内的总进人流数量和总出人流数量；
    所述在所述人流状态数据满足告警条件的情况下，生成人流状态告警信息，包括：
    在满足以下条件之一的情况下，生成所述人流状态告警信息：
    所述总进人流数量大于设置的第一人流阈值，且所述总出人流数量大于设置的第二人流阈值；
    所述总进人流数量大于所述第一人流阈值，或，所述总出人流数量大于所述第二人流阈值。
  6. 根据权利要求4所述的方法，其中，在所述监控点位为一个的情况下，基于所述人数监测数据，确定所述至少一个监控设备的人流状态数据，包括：
    基于所述进人流数量和所述出人流数量，确定所述监控点位对应的监测区域中的进人流速度和出人流速度。
  7. 根据权利要求4所述的方法，其中，在所述监控点位为多个的情况下，基于所述人数监测数据，确定所述至少一个监控设备的人流状态数据，包括：
    针对每个所述监控点位，基于所述进人流数量和所述出人流数量，确定每个所述监控点位分别对应的监测区域中所述预设时间段内的总进人流数量和总出人流数量；
    基于所述预设时间段内目标监控区域的历史人数，以及每个所述监控点位分别对应的所述预设时间段内的总进人流数量和总出人流数量，确定所述目标监控区域内的人员净存量；
    所述在所述人流状态数据满足告警条件的情况下，生成人流状态告警信息，包括：
    在所述人员净存量大于设置的净存量阈值的情况下，生成所述人流状态告警信息。
  8. 根据权利要求1或2所述的方法，其中，在所述监测事件为过密事件的情况下，基于所述监控视频，确定所述至少一个监控点位对应的监测区域是否发生监测事件，包括：
    基于所述监控视频，确定所述至少一个监控点位对应的监测区域内的目标对象个数是否超过过密阈值；
    在所述目标对象个数超过所述过密阈值的情况下，确定所述至少一个监控点位对应的监测区域发生过密事件。
  9. 根据权利要求2所述的方法，其中，在所述监测事件为过密事件的情况下，获取预设时间段内与所述监测事件匹配的人数监测数据，包括：
    统计所述预设时间段内不同采集时间点的所述目标对象的个数。
  10. 根据权利要求9所述的方法，其中，在所述监控点位为一个的情况下，基于所述人数监测数据，确定所述至少一个监控设备的人流状态数据，包括：
    基于所述目标对象的个数，确定所述监控点位对应的监测区域在所述预设时间段内的平均人数；
    所述在所述人流状态数据满足告警条件的情况下，生成人流状态告警信息，包括：
    在所述平均人数大于设置的第一人数阈值的情况下，生成所述人流状态告警信息。
  11. 根据权利要求9所述的方法，其中，在所述监控点位为多个的情况下，基于所述人数监测数据，确定所述至少一个监控设备的人流状态数据，包括：
    针对每个所述监控点位，基于所述目标对象的个数，确定所述监控点位对应的监测区域在所述预设时间段内的平均人数；
    基于所述平均人数，确定目标监控区域中的总实时人数；
    所述在所述人流状态数据满足告警条件的情况下，生成人流状态告警信息，包括：
    在所述总实时人数大于设置的第二人数阈值的情况下，生成所述人流状态告警信息。
  12. 根据权利要求1至11任一所述的方法，其中，所述方法还包括：
    将最近多个历史日期内同一采集时间点的人流状态数据求平均，得到每个采集时间点对应的预测人流状态数据；
    基于所述预测人流状态数据，构成人流状态数据在未来日期内的预测数据；其中，所述预测数据用于生成人流疏导计划。
  13. 一种场景监测装置，包括：
    第一获取模块，配置为获取设置于至少一个监控点位的监控设备采集的监控视频；
    检测模块，配置为基于所述监控视频，确定所述至少一个监控点位对应的监测区域是否发生监测事件；
    第二获取模块，配置为在所述至少一个监控点位对应的监测区域发生所述监测事件的情况下，获取预设时间段内与所述监测事件匹配的人数监测数据；
    确定模块，配置为基于所述人数监测数据，确定所述至少一个监控设备的人流状态数据。
  14. 一种电子设备，包括：处理器、存储器和总线，所述存储器存储有所述处理器可执行的机器可读指令，当所述电子设备运行时，所述处理器与所述存储器之间通过总线通信，所述机器可读指令被所述处理器执行时执行如权利要求1至12任一所述的场景监测方法。
  15. 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如权利要求1至12任一所述的场景监测方法。
  16. 一种计算机程序，所述计算机程序包括计算机可读代码，在所述计算机可读代码在电子设备中运行的情况下，所述电子设备的处理器执行用于实现如权利要求1至12任一所述的场景监测方法。
PCT/CN2021/094699 2020-10-30 2021-05-19 场景监测方法、装置、电子设备、存储介质及程序 WO2022088653A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
JP2021576933A JP7305808B2 (ja) 2020-10-30 2021-05-19 現場監視方法及び装置、電子機器、記憶媒体並びにプログラム
KR1020217042832A KR20220058859A (ko) 2020-10-30 2021-05-19 시나리오 모니터링 방법, 장치, 전자 기기, 저장 매체 및 프로그램

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011190695.6A CN112333431B (zh) 2020-10-30 2020-10-30 场景监测方法、装置、电子设备及存储介质
CN202011190695.6 2020-10-30

Publications (1)

Publication Number Publication Date
WO2022088653A1 true WO2022088653A1 (zh) 2022-05-05

Family

ID=74297411

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/094699 WO2022088653A1 (zh) 2020-10-30 2021-05-19 场景监测方法、装置、电子设备、存储介质及程序

Country Status (4)

Country Link
JP (1) JP7305808B2 (zh)
KR (1) KR20220058859A (zh)
CN (2) CN114900669A (zh)
WO (1) WO2022088653A1 (zh)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115471978A (zh) * 2022-08-18 2022-12-13 北京声智科技有限公司 一种游泳场所的监控方法及装置
CN115909667A (zh) * 2022-12-07 2023-04-04 宁波云弧科技有限公司 一种油罐区的监控报警系统
CN116012776A (zh) * 2022-12-09 2023-04-25 北京数原数字化城市研究中心 一种人数监测方法、装置、电子设备及可读存储介质
CN116188357A (zh) * 2022-09-27 2023-05-30 珠海视熙科技有限公司 一种出入口人体检测方法、摄像设备、装置及存储介质
CN117238092A (zh) * 2023-11-16 2023-12-15 建龙西林钢铁有限公司 基于倾斜摄影和人车定位的工业厂区风险预警方法

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
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
CN113507588A (zh) * 2021-06-03 2021-10-15 山西三友和智慧信息技术股份有限公司 一种基于人工智能的智慧校园人流量监测系统
CN113536932A (zh) * 2021-06-16 2021-10-22 中科曙光国际信息产业有限公司 人群聚集预测方法、装置、计算机设备和存储介质
CN113762169A (zh) * 2021-09-09 2021-12-07 北京市商汤科技开发有限公司 人流量统计方法及装置、电子设备和存储介质
TWI796033B (zh) * 2021-12-07 2023-03-11 巨鷗科技股份有限公司 人流分析辨識系統
CN114724360A (zh) * 2022-03-14 2022-07-08 江上(上海)软件科技有限公司 一种基于智慧城市的应用预警系统及预警方法
CN114694285B (zh) * 2022-03-29 2023-09-01 重庆紫光华山智安科技有限公司 人流量告警方法、装置、电子设备和存储介质
CN115474005A (zh) * 2022-10-28 2022-12-13 通号通信信息集团有限公司 数据处理方法及数据处理装置、电子设备、存储介质

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
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
CN109087478A (zh) * 2018-08-22 2018-12-25 徐自远 一种智能防拥挤踩踏的预警与导流方法及系统
CN109272153A (zh) * 2018-09-10 2019-01-25 合肥巨清信息科技有限公司 一种旅游景区人流预警系统
CN109428938A (zh) * 2017-09-04 2019-03-05 上海仪电（集团）有限公司中央研究院 一种基于视频分析的联动控制智能系统
JP2019117425A (ja) * 2017-12-26 2019-07-18 キヤノンマーケティングジャパン株式会社 情報処理装置、及びその制御方法、プログラム
CN110708518A (zh) * 2019-11-05 2020-01-17 北京深测科技有限公司 一种人流分析预警疏导方法及系统
CN110929648A (zh) * 2019-11-22 2020-03-27 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
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
CN112333431A (zh) * 2020-10-30 2021-02-05 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

Family Cites Families (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007243342A (ja) 2006-03-06 2007-09-20 Yokogawa Electric Corp 画像監視装置及び画像監視システム
US9197861B2 (en) * 2012-11-15 2015-11-24 Avo Usa Holding 2 Corporation Multi-dimensional virtual beam detection for video analytics
WO2014174737A1 (ja) 2013-04-26 2014-10-30 日本電気株式会社 監視装置、監視方法および監視用プログラム
CN104239908A (zh) * 2014-07-28 2014-12-24 中国科学院自动化研究所 基于自适应阈值的智能乘客流量自动统计方法
JP6631619B2 (ja) * 2015-03-27 2020-01-15 日本電気株式会社 映像監視システム及び映像監視方法
US9840166B2 (en) * 2015-04-13 2017-12-12 Verizon Patent And Licensing Inc. Determining the number of people in a vehicle
CN105139425B (zh) * 2015-08-28 2018-12-07 浙江宇视科技有限公司 一种人数统计方法及装置
CN105447458B (zh) * 2015-11-17 2018-02-27 深圳市商汤科技有限公司 一种大规模人群视频分析系统和方法
CN205354276U (zh) * 2015-12-24 2016-06-29 上海市水利工程设计研究院有限公司 一种感压式人流密度报警装置
SG11201805830TA (en) 2016-01-12 2018-08-30 Hitachi Int Electric Inc Congestion-state-monitoring system
CN106211065A (zh) * 2016-06-30 2016-12-07 北京奇虎科技有限公司 人员流量数据的监控方法及装置
US10936882B2 (en) 2016-08-04 2021-03-02 Nec Corporation People flow estimation device, display control device, people flow estimation method, and recording medium
CN107844848B (zh) * 2016-09-20 2020-12-29 中国移动通信集团湖北有限公司 一种区域人流量预测方法及系统
CN106407946B (zh) * 2016-09-29 2020-03-03 北京市商汤科技开发有限公司 跨线计数方法和深度神经网络训练方法、装置和电子设备
CN106778688B (zh) * 2017-01-13 2020-03-31 辽宁工程技术大学 一种拥挤场景监控视频中人群流异常事件的检测方法
CN107133607B (zh) * 2017-05-27 2019-10-11 上海应用技术大学 基于视频监控的人群统计方法及系统
CN107911653B (zh) * 2017-11-16 2021-01-01 王磊 驻所智能视频监控模组、系统、方法以及存储介质
CN109685009A (zh) * 2018-12-20 2019-04-26 天和防务技术（北京）有限公司 一种区域人群密度视频检测的方法
CN111274340B (zh) * 2020-01-15 2023-06-30 中国联合网络通信集团有限公司 人流密度的监控处理方法、设备及存储介质

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
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
CN109428938A (zh) * 2017-09-04 2019-03-05 上海仪电（集团）有限公司中央研究院 一种基于视频分析的联动控制智能系统
JP2019117425A (ja) * 2017-12-26 2019-07-18 キヤノンマーケティングジャパン株式会社 情報処理装置、及びその制御方法、プログラム
CN109087478A (zh) * 2018-08-22 2018-12-25 徐自远 一种智能防拥挤踩踏的预警与导流方法及系统
CN109272153A (zh) * 2018-09-10 2019-01-25 合肥巨清信息科技有限公司 一种旅游景区人流预警系统
CN110708518A (zh) * 2019-11-05 2020-01-17 北京深测科技有限公司 一种人流分析预警疏导方法及系统
CN110929648A (zh) * 2019-11-22 2020-03-27 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
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
CN112333431A (zh) * 2020-10-30 2021-02-05 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

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115471978A (zh) * 2022-08-18 2022-12-13 北京声智科技有限公司 一种游泳场所的监控方法及装置
CN116188357A (zh) * 2022-09-27 2023-05-30 珠海视熙科技有限公司 一种出入口人体检测方法、摄像设备、装置及存储介质
CN115909667A (zh) * 2022-12-07 2023-04-04 宁波云弧科技有限公司 一种油罐区的监控报警系统
CN115909667B (zh) * 2022-12-07 2024-05-17 宁波云弧科技有限公司 一种油罐区的监控报警系统
CN116012776A (zh) * 2022-12-09 2023-04-25 北京数原数字化城市研究中心 一种人数监测方法、装置、电子设备及可读存储介质
CN116012776B (zh) * 2022-12-09 2024-02-23 北京数原数字化城市研究中心 一种人数监测方法、装置、电子设备及可读存储介质
CN117238092A (zh) * 2023-11-16 2023-12-15 建龙西林钢铁有限公司 基于倾斜摄影和人车定位的工业厂区风险预警方法
CN117238092B (zh) * 2023-11-16 2024-01-30 建龙西林钢铁有限公司 基于倾斜摄影和人车定位的工业厂区风险预警方法

Also Published As

Publication number Publication date
JP7305808B2 (ja) 2023-07-10
CN112333431B (zh) 2022-06-07
JP2023502816A (ja) 2023-01-26
CN112333431A (zh) 2021-02-05
CN114900669A (zh) 2022-08-12
KR20220058859A (ko) 2022-05-10

Similar Documents

Publication Publication Date Title
WO2022088653A1 (zh) 场景监测方法、装置、电子设备、存储介质及程序
CN104750768B (zh) 用于从社交媒体中识别、监控和排名事件的方法和系统
US10832416B2 (en) Crowd flow rate estimation
CN113330491B (zh) 电子闸门的开启方法、装置及服务器
Zuo et al. Reference-free video-to-real distance approximation-based urban social distancing analytics amid COVID-19 pandemic
CN109615256A (zh) 智慧园区安防风险控制方法、存储介质、电子设备及系统
KR102511626B1 (ko) 인공지능 기반 엣지 컴퓨팅을 활용한 유형별 객체 계수를 위한 장치 및 방법
US20220300727A1 (en) Systems and methods of detecting mask usage
CN114358980A (zh) 一种基于物联网的智慧社区物业管理系统及方法
CN116614717B (zh) 用于智慧社区的视频监控方法和系统
Davies et al. A progress review of intelligent CCTV surveillance systems
JP6621092B1 (ja) 危険度判別プログラム及びシステム
WO2023173616A1 (zh) 一种人群统计方法及装置、电子设备和存储介质
CN113901946A (zh) 一种异常行为检测的方法、装置、电子设备及存储介质
KR102614856B1 (ko) 군중 난류 위험 예측 시스템 및 방법
CN112508626A (zh) 一种信息处理方法、装置、电子设备及存储介质
CN116863580A (zh) 基于物联网的智慧门禁系统
US20200020110A1 (en) Image-based object tracking systems and methods
Mongia et al. Detecting activities at metro stations using smartphone sensors
CN111862430A (zh) 一种人脸识别无闸机的门禁系统及其方法
JP6739115B1 (ja) 危険度判別プログラム及びシステム
US20220269878A1 (en) Systems and methods of detecting incorrect mask usage
CN209993020U (zh) 一种人脸识别无闸机的门禁系统
JP2021097308A (ja) 危険度判別プログラム及びシステム
Wang et al. Early warning of city-scale unusual social event on public transportation smartcard data

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2021576933

Country of ref document: JP

Kind code of ref document: A

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

Ref document number: 21884396

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 11/08/2023)

122 Ep: pct application non-entry in european phase

Ref document number: 21884396

Country of ref document: EP

Kind code of ref document: A1