WO2020048156A1 - Method and apparatus for counting vehicle flow, computing device, and computer readable storage medium - Google Patents

Method and apparatus for counting vehicle flow, computing device, and computer readable storage medium Download PDF

Info

Publication number
WO2020048156A1
WO2020048156A1 PCT/CN2019/087229 CN2019087229W WO2020048156A1 WO 2020048156 A1 WO2020048156 A1 WO 2020048156A1 CN 2019087229 W CN2019087229 W CN 2019087229W WO 2020048156 A1 WO2020048156 A1 WO 2020048156A1
Authority
WO
WIPO (PCT)
Prior art keywords
vehicle
video
reference line
video stream
distance
Prior art date
Application number
PCT/CN2019/087229
Other languages
French (fr)
Chinese (zh)
Inventor
李涛
林伟彬
李健
Original Assignee
华为技术有限公司
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 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2020048156A1 publication Critical patent/WO2020048156A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/065—Traffic control systems for road vehicles by counting the vehicles in a section of the road or in a parking area, i.e. comparing incoming count with outgoing count
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125—Traffic data processing

Definitions

  • This application relates to the field of video, and in particular, to a method and device for counting vehicle traffic, a computing device, and a computer-readable storage medium.
  • Vehicles have become a common means of transportation for people to travel. With the popularization of vehicles, roads (such as urban traffic) are often congested, so traffic flow needs to be monitored.
  • the current monitoring method is to use radar signals to monitor passing vehicles, or to use gravity sensors to monitor passing vehicles.
  • the present application provides a method, a device, and a computing device for counting vehicle traffic, which can implement monitoring of vehicle traffic through video streaming.
  • the present application provides a method for counting vehicle traffic.
  • the computing device obtains a video stream that monitors vehicle traffic; for example, the video stream may be a video stream generated by a shooting device that monitors vehicle traffic on a road, and the computing device obtains the video stream from the shooting device.
  • the video stream includes a plurality of video frames.
  • the computing device identifies a vehicle in each video frame from the video stream.
  • the computing device calculates the vehicle flow monitored by the video stream according to the time sequence and the direction of the vehicle flow according to the vehicle in each video frame in the video stream.
  • the present application can monitor the traffic flow on the road by means of video streaming.
  • the computing device may calculate the vehicle flow in the following manner.
  • the computing device sets a reference line at the same position in each video frame in the video stream, the reference line being perpendicular to the direction of the vehicle flow. In this way, the computing device can count vehicles passing the reference line.
  • the computing device identifies the number of vehicles passing the reference line from the video stream according to the time sequence and the direction of the vehicle flow.
  • the number of vehicles passing the reference line is the traffic volume.
  • the computing device may identify the number of vehicles passing the reference line from the video stream in the following manner.
  • the computing device determines a target vehicle in each video frame, and the target vehicle is a vehicle that is closest to the reference line in the video frame in the direction of the traffic flow.
  • the computing device calculates the distance between the target vehicle in the video frame and the reference line, and uses the calculated distance as the target distance in the video frame.
  • the computing device can calculate the target distance in each video frame in the video stream.
  • the computing device identifies each vehicle passing the reference line in chronological order, based on a strategy and a target distance in each of the video frames in the video stream.
  • the strategy is: identifying the currently passing vehicle when the target distance in the first video frame is less than the target distance in the second video frame, and the first video frame and the second video frame are in the video stream Two video frames next to each other in chronological order.
  • the present application can recognize the number of vehicles passing the reference line through the event that the target distance of adjacent video frames in the video stream becomes larger, and simultaneously identify the number of vehicles passing the reference line.
  • vehicles that pass through the reference line each time can be identified; in this way, vehicles that pass through the reference line all times, that is, traffic flow can be calculated.
  • the computing device calculates a distance from a head of a target vehicle in the video frame to a reference line, and the calculated distance is the distance in the video frame. Target distance.
  • the computing device calculates a distance from a tail of a target vehicle in the video frame to a reference line, and the calculated distance is in the video frame. Target distance.
  • the computing device calculates a distance from a reference point on a vehicle body of a target vehicle in the video frame to a reference line, and the calculated distance is The target distance in the video frame.
  • the present application provides a device for counting vehicle traffic, including a plurality of functional units.
  • the device executes the steps in the first aspect or the method for counting traffic flow provided by any possible design of the first aspect through the multiple functional units.
  • the present application provides a computing device including a processor and a memory.
  • the memory stores computer instructions; the processor executes the computer instructions stored in the memory, so that the computing device executes steps in the first aspect or any possible design provided by the first aspect of the method for counting vehicle traffic.
  • the computer instructions stored in the memory are used to implement a functional unit in any one of the devices for counting vehicle traffic provided by the second aspect.
  • the computing device executes the steps in the method for counting vehicle traffic through the functional unit.
  • a computer-readable storage medium stores computer instructions.
  • the processor of the computing device executes the computer instructions
  • the computing device executes the first aspect or any possibility of the first aspect. Steps in designing a provided method for counting vehicle traffic.
  • the computer instructions stored in the computer-readable storage medium are used to implement a functional unit in any one of the devices for counting vehicle traffic provided by the second aspect.
  • the computing device executes the steps in the method for counting vehicle traffic through the functional unit.
  • a computer program product includes computer instructions stored in a computer-readable storage medium.
  • the processor of the computing device may read and execute the computer instructions from the computer-readable storage medium, so that the computing device executes the steps in the method of counting vehicle traffic provided by the first aspect or any possible design provided by the first aspect.
  • the computer instructions in the computer program product are used to implement a functional unit in any one of the devices for counting vehicle traffic provided by the second aspect.
  • the computing device executes the steps in the method for counting vehicle traffic through the functional unit.
  • a system for counting vehicle traffic includes a computing device and a photographing device in the first aspect or any possible design of the first aspect.
  • the computing device executes the steps in the first aspect or any possible design method provided by the first aspect of the method for counting vehicle traffic, for example, the computing device deploys the device for counting vehicle traffic provided by the second aspect, and the computing device passes the device in the device.
  • FIG. 1 is a schematic diagram of traffic flow
  • FIG. 2 is a schematic diagram of a scenario to which the method provided by this application is applicable;
  • FIG. 3 is a schematic flowchart of a method for counting traffic flow provided by the present application.
  • FIG. 4A and FIG. 4B are respectively a schematic diagram for identifying the passage of a vehicle provided by the present application.
  • FIG. 5 is a schematic diagram of a logical structure of a device 50 for counting vehicle traffic provided by the present application
  • FIG. 6 is a schematic diagram of a hardware structure of a computing device 12 provided in this application.
  • Vehicle means the vehicle that moves the body by turning the wheels.
  • the vehicle may be a non-motor vehicle or a motor vehicle.
  • Traffic flow direction the direction in which multiple vehicles move on the road in the same direction. Taking Figure 1 as an example, three vehicles (A, B, C) are driving in the lane. These three vehicles are driving in the same direction, and this direction is the direction of traffic flow.
  • Traffic refers to vehicles that pass through the same reference line on the road in the same direction in unit time.
  • Figure 1 illustrates the reference line.
  • Vehicles (A, B) have not reached the reference line, and vehicle C is passing through the reference line.
  • the reference line may be a straight line or a line composed of a plurality of spaced points.
  • the reference line may be a straight line or other types of lines.
  • the reference line is perpendicular to the direction of the traffic flow.
  • a method for calculating the traffic flow When the number of vehicles passing the reference line within a preset time period is M, the traffic flow is a ratio of M to the preset time period, and M is a positive integer.
  • the system includes a photographing device 21 and a computing device 22. It can be understood that, in another embodiment, the system may further include other processing devices, such as a network forwarding device, a video processing device, and the like.
  • the photographing device 11 is configured to acquire a traffic flow on a monitored road and photograph a vehicle passing through a fixed road segment.
  • the photographing device 11 forms a video stream based on the photos obtained by continuous photographing, and the video stream includes multiple video frames, that is, each frame in the video stream is a video frame.
  • This application does not limit the manner of forming a video stream based on multiple photos; for example, one photo obtained from each photo may be a video frame, and multiple photos obtained from the photo are combined into a video stream in chronological order; for example, A plurality of photos obtained by taking pictures are processed to form a video stream.
  • One photo corresponds to one video frame, but the characteristics of the vehicle in the picture are retained in the video frame.
  • the photographing device 11 sends the generated video stream to a computing device 12 in the system.
  • the computing device 12 may use the method provided in the present application to count the vehicle traffic.
  • This application provides an embodiment of a method for counting vehicle traffic.
  • the computing device 12 in the embodiment shown in FIG. 2 described above can be used to count vehicle traffic.
  • the method includes steps S31, S32, and S33, as shown in FIG.
  • step S31 the computing device 12 obtains a video stream from the photographing device 11.
  • the video stream is a video stream obtained by the shooting device 11 monitoring the traffic flow of a preset road segment, and the video stream includes a plurality of video frames.
  • the computing device 12 may request a video stream from the shooting device 11, and the shooting device 11 sends the video stream to the computing device 12.
  • the shooting device 11 may actively send the video stream to the computing device 12, for example, periodically send the video stream to the computing device.
  • the computing device 12 receives the video stream sent by the photographing device 11 and thus completes the acquisition of the video stream.
  • step S32 the computing device 12 identifies a vehicle in each video frame from the video stream.
  • the video stream includes a plurality of video frames, and the computing device 12 makes a vehicle identification for each video frame in the video stream.
  • the computing device 12 deploys a Convolutional Neural Network (CNN) algorithm for identifying vehicles.
  • CNN Convolutional Neural Network
  • the computing device 12 uses the CNN algorithm to identify vehicles from the video frame, for example, to identify vehicles closest to the reference line, or to identify all vehicles.
  • the computing device 12 deploys a candidate region (RP) algorithm for identifying a vehicle.
  • the computing device 12 uses the RP algorithm to identify vehicles from the video frame, for example, to identify the vehicle closest to the above reference line, or to identify all vehicles.
  • step S33 the computing device 12 calculates the traffic flow monitored by the video stream according to the time sequence and the direction of the traffic flow, based on the vehicles in each of the video frames identified from the video stream.
  • the computing device 12 can calculate the preset according to the chronological order and the direction of the traffic flow. Traffic on road segments. The following examples provide several statistical methods.
  • the computing device 12 can count the number of vehicles entering the preset road segment according to the time sequence and the direction of the traffic flow.
  • the computing device 12 may calculate the traffic volume of the preset road segment monitored by the video stream according to the counted number and the time period monitored by the video stream.
  • the range of scenes ie, preset road segments
  • the range of scenes that can be monitored by a video stream is limited.
  • it can be identified from the pair of adjacent video frames of the video stream whether a new vehicle has entered the preset road segment; if a new vehicle has entered , Then the number of newly driven vehicles is identified.
  • the newly entered vehicles can be identified from the video stream, and the total number of newly entered vehicles can be counted each time. Therefore, the traffic volume of the preset road segment monitored by the video stream can be calculated according to the sum of statistics and the time period monitored by the video stream.
  • the computing device 12 may count the number of vehicles driving out of the preset road segment according to the time sequence and the direction of the traffic flow.
  • the computing device 12 may calculate the traffic volume of the preset road segment monitored by the video stream according to the counted number and the time period monitored by the video stream.
  • the range of scenes (ie, preset road segments) that can be monitored by a video stream is limited.
  • it can be identified from the pair of adjacent video frames of the video stream whether a new vehicle is driving out of the preset road segment; if a new vehicle is driving out , The number of newly driven vehicles is identified.
  • the traffic volume of the preset road segment monitored by the video stream can be calculated according to the sum of statistics and the time period monitored by the video stream.
  • the computing device 12 may set a reference line for each video frame in the video stream at the same position in each video frame, as shown in FIG. 1, the reference line and the vehicle flow Direction is vertical.
  • this method is not limited to the specific position of the reference line in the video frame. For example, it can be set at the entrance or exit of the traffic direction in the video frame, and can also be set at any other position in the video frame. position.
  • the computing device 12 After setting the reference line, the computing device 12 identifies vehicles passing the reference line from the video stream in chronological order and the direction of the traffic flow. The computing device 12 counts the number of vehicles passing the reference line, and may calculate the traffic volume of the preset road segment monitored by the video stream according to the counted number and the time period monitored by the video stream.
  • An example is optionally provided to identify a vehicle passing the reference line from the video stream.
  • the computing device 12 determines a target vehicle in each video frame in the video stream, and the target vehicle is a vehicle that is closest to the reference line in the direction of the flow in the video frame. In the video frame, the distance between the target vehicle of the video frame and the reference line is calculated, and the calculated distance is used as the target distance in the video frame.
  • the computing device 12 identifies each vehicle passing the reference line in chronological order according to a policy and a target distance in each of the video frames in the video stream.
  • the strategy is: the first video frame and the second video frame are two video frames adjacent to each other in time sequence in the video stream; the target distance in the first video frame is smaller than the target in the second video frame
  • the number of currently passing vehicles can be identified.
  • the number of passing vehicles can be one or more. If multiple vehicles pass at the same time, the number of vehicles passing at the same time is identified. If only one vehicle passes, the number of vehicles identified as passing is one.
  • step S32 Since the vehicle in each video frame has been identified in step S32, that is, the position of the vehicle in the video frame has been identified, so when a vehicle is passed through the reference line according to the strategy, the currently passing vehicle and the identification can be further identified. The number of vehicles currently passing.
  • An example is optionally provided to identify a vehicle passing the reference line from the video stream.
  • the computing device 12 calculates the distance between each vehicle and the reference line in each video frame in the video stream, and selects the shortest distance from the calculated distance between each vehicle and the reference line. Distance, using the selected shortest distance as the target distance in the video frame, and the vehicle corresponding to the shortest distance as the target vehicle in the video frame. In the same video frame, if multiple vehicles have the target distance from the reference line, the multiple vehicles simultaneously serve as multiple target vehicles, that is, each of the multiple vehicles is the target vehicle; if only If a vehicle has the target distance from the reference line, the vehicle is regarded as the target vehicle.
  • the computing device 12 identifies each vehicle passing the reference line in chronological order according to a policy and a target distance in each of the video frames in the video stream.
  • the computing device 12 can identify all vehicles passing the reference line in the video stream through the strategy. According to the number of all vehicles passing the reference line, the traffic volume of the preset road segment monitored by the video stream can be calculated.
  • the strategy further includes: identifying that no vehicle passes when the target distance in the first video frame is greater than or equal to the target distance in the second video frame, the first video frame and the second video frame
  • the video frames are two video frames adjacent to each other in the video stream in chronological order. In this way, the computing device 12 can accurately identify whether a vehicle passes the reference line based on the target distance in each of the video frames in the video stream in chronological order through the strategy.
  • the I frame, the J frame, and the K frame are three video frames in the video stream arranged in time sequence.
  • the head of the vehicle C is closest to the reference line compared to the vehicle A and the vehicle B, and the distance between the head of the vehicle C and the reference line is 10 meters (meter, m).
  • the head of the vehicle C is closest to the reference line compared to the vehicle A and the vehicle B, and the distance between the head of the vehicle C and the reference line is 5 meters.
  • the head of vehicle C has passed the reference line, so the distance between vehicle C and the reference line is no longer considered, and only vehicle A and vehicle B are compared; at this time, compared with vehicle A, vehicle B
  • vehicle B The front of the vehicle is closest to the reference line, and the distance between the front of vehicle B and the reference line is 15 meters. Because the distance between vehicle B and the reference line (15 meters) in frame K is greater than the distance between vehicle C and the reference line (5 meters) in frame J, it is recognized that a vehicle has passed, and the vehicle that has passed is identified as vehicle C , That is, the number of passing vehicles is one.
  • the I frame, the J frame, and the K frame are three video frames in the video stream arranged in time sequence.
  • the heads of vehicles B and C are closest to the reference line compared to vehicle A.
  • the distance between the head of vehicle B and the reference line is 10 meters.
  • the distance is 10 meters.
  • the heads of vehicles B and C are closest to the reference line compared to vehicle A.
  • the distance between the head of vehicle B and the reference line is 5 meters, and the distance between the head of vehicle C and the reference line The distance is 5 meters.
  • the head of vehicle B and the head of vehicle C pass the reference line at the same time according to the direction of the traffic, so the distances between vehicle B and vehicle C from the reference line are no longer considered; at this time, the head of vehicle A is away from the reference line
  • the distance between the front of the vehicle A and the reference line was 10 meters. Since the distance between vehicle A and the reference line (10 meters) in frame K is greater than the distance between vehicle C and the reference line (5 meters) in frame J, it is recognized that a vehicle has passed, and the vehicle that has passed is identified as vehicle B. And vehicle C, that is, the number of passing vehicles is two.
  • the computing device 12 calculates the distance between the head of the vehicle and the reference line, and uses the calculated distance as the distance of the vehicle from the reference line. For example, in each video frame, the distance from the head of the target vehicle to the reference line is calculated, and the calculated distance is the target distance in the video frame.
  • the computing device 12 calculates the distance between the tail of the vehicle and the reference line, and uses the calculated distance as the distance of the vehicle from the reference line.
  • a reference point is selected from each vehicle (may be located at the middle position of the vehicle or any other position), and the computing device 12 calculates the reference point of the vehicle and the The distance of the reference line. The calculated distance is taken as the distance of the vehicle from the reference line.
  • This application provides a device for counting vehicle traffic.
  • the device is deployed on the computing device 12 in the embodiment shown in FIG. 2.
  • the device includes functional units for implementing steps in the method for counting vehicle traffic.
  • the embodiment does not limit how to divide the functional units in the device, and a division of the functional units is provided as an example below, as shown in FIG. 5.
  • the device 50 for counting vehicle traffic shown in FIG. 5 includes:
  • An obtaining unit 51 configured to obtain a video stream of monitored vehicle traffic, where the video stream includes multiple video frames;
  • An identification unit 52 configured to identify a vehicle in each of the video frames from the video stream
  • the calculation unit 53 is configured to calculate, according to the time sequence and the direction of the traffic flow, the traffic volume monitored by the video flow according to the vehicles in each of the video frames identified from the video flow.
  • calculation unit 53 is configured to:
  • the number of vehicles passing the reference line is identified from the video stream in chronological order and the direction of the vehicle flow.
  • calculation unit 53 is configured to:
  • the target distance being a distance between a target vehicle in the video frame and the reference line, and the target vehicle being in the video frame The vehicle closest to the reference line in the direction of the traffic flow;
  • each vehicle passing through the reference line is identified according to a strategy and a target distance in each of the video frames in the video stream, the strategy is: the target distance in the first video frame is less than the first When the target distance in the two video frames is recognized, the currently passing vehicle is identified, and the first video frame and the second video frame are two video frames adjacent to each other in time sequence in the video stream.
  • the calculation unit 53 is configured to calculate a distance from a head of the target vehicle to the reference line in each of the video frames, and the calculated distance is a target distance in the video frame.
  • Each function in the acquisition unit 51, the recognition unit 52, and the calculation unit 53 has corresponding steps in the above method. Therefore, for the implementation details of the functions in the acquisition unit 51, the identification unit 52, and the calculation unit 53, refer to the description of the corresponding steps in the foregoing method.
  • a possible basic hardware architecture of the computing device 12 is provided as an example below, as shown in FIG. 6.
  • the computing device 12 includes a processor 121, a memory 122, a communication interface 123, and a bus 124.
  • the number of processors 121 may be one or more, and FIG. 1 illustrates only one of the processors 121.
  • the processor 121 may be a central processing unit (CPU) or an ARM processor. If the computing device 12 has multiple processors 121, the types of the multiple processors 121 may be different or may be the same. Optionally, the multiple processors 121 of the computing device 12 may also be integrated into a multi-core processor.
  • the memory 122 stores computer instructions; for example, the computer instructions include a chain code; for example, the computer instructions are used to implement each step in the method provided by the present application; for example, the computer instructions are used to implement each of the steps included in the apparatus 50 provided in the present application. A functional unit, or used to implement each step in the method provided in this application.
  • the memory 122 may be any one or any combination of the following storage media: non-volatile memory (NVM) (such as read-only memory (ROM), solid state drives (Solid State Drives, SSD), mechanical hard disks, magnetic disks, and entire arrays of disks), volatile memory (volatile memory).
  • NVM non-volatile memory
  • ROM read-only memory
  • SSD Solid State Drives
  • volatile memory volatile memory
  • the communication interface 123 may be any one or any combination of the following devices: a network interface (such as an Ethernet interface), a wireless network card, and other devices having a network access function.
  • the communication interface 123 is used for data communication between the computing device 12 and other devices (for example, a computing device).
  • FIG. 6 shows the bus 124 with a thick line.
  • the processor 121, the memory 122, and the communication interface 123 are connected through a bus 124.
  • the processor 121 can access the memory 122 through the bus 124 and use the communication interface 123 to perform data interaction with other devices (for example, computing devices) through the bus 124.
  • the computing device 12 executes the computer instructions in the memory 122 to implement the method for counting vehicle traffic provided in the present application on the computing device 12 or implement the device 50 provided in the present application on the computing device 12.
  • the computing device 12 is a server in a public cloud, a private cloud, or a hybrid cloud. After the resources of the computing device 12 are virtualized, a device is deployed on the virtualized resources. This device is used to implement the method for counting vehicle traffic provided in this application, or this device is the device 50 provided in this application.
  • the present application provides a computer-readable storage medium.
  • the computer-readable storage medium stores computer instructions.
  • the processor 121 of the computing device 12 executes the computer instructions, the computing device 12 executes the statistics of vehicle traffic provided by the application. Steps in the method.
  • the application provides a computer-readable storage medium.
  • the computer-readable storage medium stores computer instructions, and the computer instructions are used to implement the device 50.
  • the application provides a computer program product including computer instructions, the computer instructions being stored in a computer-readable storage medium.
  • the processor of the computing device may read and execute the computer instructions from the computer-readable storage medium, so that the computing device executes the steps of the method for counting vehicle traffic provided by the present application.
  • the application provides a computer program product.
  • the computer program product includes computer instructions for implementing the device 50.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Traffic Control Systems (AREA)
  • Image Analysis (AREA)

Abstract

A method and apparatus for counting a vehicle flow, and a computing device (12). The method comprises: the computing device (12) obtains a video stream monitoring the vehicle flow (S31), the video stream comprising multiple video frames; the computing device (12) identifies vehicles in each video frame from the video stream (S32); the computing device (12) calculates the vehicle flow monitored by means of the video stream according to the vehicles, which are identified from the video stream, in each video frame on the basis of a chronological order and a vehicle flow direction. According to the method, the vehicle flow on a road can be monitored by means of the video stream (S33).

Description

统计车流量的方法和装置、计算设备和计算机可读存储介质Method and device for counting traffic flow, computing device and computer-readable storage medium 技术领域Technical field
本申请涉及视频领域，尤其涉及统计车流量的方法和装置、计算设备和计算机可读存储介质。This application relates to the field of video, and in particular, to a method and device for counting vehicle traffic, a computing device, and a computer-readable storage medium.
背景技术Background technique
车辆已成为人们出行的常用交通工具。随着车辆的普及，道路(例如城市交通)经常拥塞，所以需要对车流量进行监控。目前的监控方式是雷达信号监测通过的车辆，或者通过重力传感器监测通过的车辆。Vehicles have become a common means of transportation for people to travel. With the popularization of vehicles, roads (such as urban traffic) are often congested, so traffic flow needs to be monitored. The current monitoring method is to use radar signals to monitor passing vehicles, or to use gravity sensors to monitor passing vehicles.
发明内容Summary of the Invention
有鉴于此，本申请提供了一种统计车流量的方法和装置、计算设备,可以通过视频流实现对车流量的监控。In view of this, the present application provides a method, a device, and a computing device for counting vehicle traffic, which can implement monitoring of vehicle traffic through video streaming.
第一方面，本申请提供一种统计车流量的方法。计算设备获取监控车流量的视频流；例如视频流可以是拍摄设备监控道路的车流量所生成的视频流，该计算设备从该拍摄设备获取该视频流。In a first aspect, the present application provides a method for counting vehicle traffic. The computing device obtains a video stream that monitors vehicle traffic; for example, the video stream may be a video stream generated by a shooting device that monitors vehicle traffic on a road, and the computing device obtains the video stream from the shooting device.
该视频流包括多个视频帧。该计算设备从该视频流中识别每个视频帧中的车辆。The video stream includes a plurality of video frames. The computing device identifies a vehicle in each video frame from the video stream.
该计算设备根据该视频流中每个视频帧中的车辆，按照时间顺序和车流方向计算出所述视频流监控到的车流量。The computing device calculates the vehicle flow monitored by the video stream according to the time sequence and the direction of the vehicle flow according to the vehicle in each video frame in the video stream.
可见，本申请可以通过视频流的方式监控道路上的车流量。It can be seen that the present application can monitor the traffic flow on the road by means of video streaming.
第一方面的一种可能设计中，计算设备可以通过如下方式计算车流量。In a possible design of the first aspect, the computing device may calculate the vehicle flow in the following manner.
在该方式中，计算设备在该视频流中每个视频帧中的相同位置设定参考线，该参考线与该车流方向垂直。这样，计算设备可以统计通过该参考线的车辆。In this manner, the computing device sets a reference line at the same position in each video frame in the video stream, the reference line being perpendicular to the direction of the vehicle flow. In this way, the computing device can count vehicles passing the reference line.
具体地，计算设备按照时间顺序和车流方向，从所述视频流中识别通过所述参考线的车辆的个数。通过该参考线的车辆的个数为该车流量。Specifically, the computing device identifies the number of vehicles passing the reference line from the video stream according to the time sequence and the direction of the vehicle flow. The number of vehicles passing the reference line is the traffic volume.
第一方面的一种可能设计中，计算设备可以通过如下方式从所述视频流中识别通过所述参考线的车辆的个数。In a possible design of the first aspect, the computing device may identify the number of vehicles passing the reference line from the video stream in the following manner.
在该方式中，计算设备确定每个视频帧中的目标车辆，该目标车辆为在该视频帧中按照所述车流方向距离与所述参考线最近的车辆。计算设备计算该视频帧中的目标车辆与所述参考线的距离，并将计算出的距离作为该视频帧中的目标距离。以此类推，计算设备可以计算出该视频流中每个视频帧中的目标距离。In this manner, the computing device determines a target vehicle in each video frame, and the target vehicle is a vehicle that is closest to the reference line in the video frame in the direction of the traffic flow. The computing device calculates the distance between the target vehicle in the video frame and the reference line, and uses the calculated distance as the target distance in the video frame. By analogy, the computing device can calculate the target distance in each video frame in the video stream.
计算设备按照时间顺序，根据策略和所述视频流中每个所述视频帧中的目标距离，识别通过所述参考线的每个车辆。所述策略为：在第一视频帧中的目标距离小于第二视频帧中的目标距离时识别当前通过的车辆，所述第一视频帧和所述第二视频帧为在所述视频流中按照时间顺序前后相邻的两个视频帧。The computing device identifies each vehicle passing the reference line in chronological order, based on a strategy and a target distance in each of the video frames in the video stream. The strategy is: identifying the currently passing vehicle when the target distance in the first video frame is less than the target distance in the second video frame, and the first video frame and the second video frame are in the video stream Two video frames next to each other in chronological order.
可见，本申请可以通过视频流中相邻视频帧的目标距离变大这个事件，识别出有车辆通过该参考线，同时识别通过该参考线的车辆的个数。以此类推，从该视频流中可以 识别出每次通过该参考线的车辆；这样，可以计算所有次数通过该参考线的车辆，即车流量。It can be seen that the present application can recognize the number of vehicles passing the reference line through the event that the target distance of adjacent video frames in the video stream becomes larger, and simultaneously identify the number of vehicles passing the reference line. By analogy, from this video stream, vehicles that pass through the reference line each time can be identified; in this way, vehicles that pass through the reference line all times, that is, traffic flow can be calculated.
第一方面的一种可能设计中，在该视频流中的每个视频帧中，计算设备计算该视频帧中的目标车辆的车头到参考线的距离，计算出的距离为该视频帧中的目标距离。In a possible design of the first aspect, in each video frame in the video stream, the computing device calculates a distance from a head of a target vehicle in the video frame to a reference line, and the calculated distance is the distance in the video frame. Target distance.
第一方面的一种可能设计中，在该视频流中的每个视频帧中，计算设备计算该视频帧中的目标车辆的车尾到参考线的距离，计算出的距离为该视频帧中的目标距离。In a possible design of the first aspect, in each video frame in the video stream, the computing device calculates a distance from a tail of a target vehicle in the video frame to a reference line, and the calculated distance is in the video frame. Target distance.
第一方面的一种可能设计中，在该视频流中的每个视频帧中，计算设备计算该视频帧中的目标车辆的车身上的基准点到参考线的距离，计算出的距离为该视频帧中的目标距离。In a possible design of the first aspect, in each video frame in the video stream, the computing device calculates a distance from a reference point on a vehicle body of a target vehicle in the video frame to a reference line, and the calculated distance is The target distance in the video frame.
第二方面，本申请提供一种统计车流量的装置，包括多个功能单元。该装置通过该多个功能单元执行第一方面或者第一方面的任意可能设计提供的统计车流量的方法中的步骤。In a second aspect, the present application provides a device for counting vehicle traffic, including a plurality of functional units. The device executes the steps in the first aspect or the method for counting traffic flow provided by any possible design of the first aspect through the multiple functional units.
第三方面，本申请提供一种计算设备，该计算设备包括处理器和存储器。该存储器存储计算机指令；该处理器执行该存储器存储的计算机指令，使得该计算设备执行第一方面或者第一方面的任意可能设计提供的统计车流量的方法中的步骤。In a third aspect, the present application provides a computing device including a processor and a memory. The memory stores computer instructions; the processor executes the computer instructions stored in the memory, so that the computing device executes steps in the first aspect or any possible design provided by the first aspect of the method for counting vehicle traffic.
第三方面的一种可能设计中，该存储器中存储的计算机指令用于实现第二方面提供的任一种统计车流量的装置中的功能单元。该计算设备通过该功能单元来执行统计车流量的方法中的步骤。In a possible design of the third aspect, the computer instructions stored in the memory are used to implement a functional unit in any one of the devices for counting vehicle traffic provided by the second aspect. The computing device executes the steps in the method for counting vehicle traffic through the functional unit.
第四方面，提供一种计算机可读存储介质，计算机可读存储介质中存储有计算机指令，当计算设备的处理器执行该计算机指令时，该计算设备执行第一方面或者第一方面的任意可能设计提供的统计车流量的方法中的步骤。According to a fourth aspect, a computer-readable storage medium is provided. The computer-readable storage medium stores computer instructions. When the processor of the computing device executes the computer instructions, the computing device executes the first aspect or any possibility of the first aspect. Steps in designing a provided method for counting vehicle traffic.
第四方面的一种可能设计中，该计算机可读存储介质中存储的计算机指令用于实现第二方面提供的任一种统计车流量的装置中的功能单元。该计算设备通过该功能单元来执行统计车流量的方法中的步骤。In a possible design of the fourth aspect, the computer instructions stored in the computer-readable storage medium are used to implement a functional unit in any one of the devices for counting vehicle traffic provided by the second aspect. The computing device executes the steps in the method for counting vehicle traffic through the functional unit.
第五方面，提供一种计算机程序产品，该计算机程序产品包括计算机指令，该计算机指令存储在计算机可读存储介质中。计算设备的处理器可以从计算机可读存储介质读取并执行该计算机指令，使得该计算设备执行第一方面或者第一方面的任意可能设计提供的统计车流量的方法中的步骤。In a fifth aspect, a computer program product is provided. The computer program product includes computer instructions stored in a computer-readable storage medium. The processor of the computing device may read and execute the computer instructions from the computer-readable storage medium, so that the computing device executes the steps in the method of counting vehicle traffic provided by the first aspect or any possible design provided by the first aspect.
第五方面的一种可能设计中，该计算机程序产品中的计算机指令用于实现第二方面提供的任一种统计车流量的装置中的功能单元。该计算设备通过该功能单元来执行统计车流量的方法中的步骤。In a possible design of the fifth aspect, the computer instructions in the computer program product are used to implement a functional unit in any one of the devices for counting vehicle traffic provided by the second aspect. The computing device executes the steps in the method for counting vehicle traffic through the functional unit.
第六方面，提供一种统计车流量的***,该***包括计算设备和第一方面或者第一方面的任意可能设计中的拍摄设备。该计算设备执行第一方面或者第一方面的任意可能设计提供的统计车流量的方法中的步骤，例如该计算设备部署第二方面提供的统计车流量的装置，该计算设备通过该装置中的功能单元来执行该方法中的步骤。According to a sixth aspect, a system for counting vehicle traffic is provided. The system includes a computing device and a photographing device in the first aspect or any possible design of the first aspect. The computing device executes the steps in the first aspect or any possible design method provided by the first aspect of the method for counting vehicle traffic, for example, the computing device deploys the device for counting vehicle traffic provided by the second aspect, and the computing device passes the device in the device. A functional unit to perform the steps in the method.
附图说明BRIEF DESCRIPTION OF THE DRAWINGS
图1为车流的一种示意图；FIG. 1 is a schematic diagram of traffic flow;
图2为本申请提供的方法所适用的场景的一种示意图；FIG. 2 is a schematic diagram of a scenario to which the method provided by this application is applicable;
图3为本申请提供的统计车流量的方法的一种流程示意图；FIG. 3 is a schematic flowchart of a method for counting traffic flow provided by the present application; FIG.
图4A和图4B分别为本申请提供的识别车辆通过的一种示意图；FIG. 4A and FIG. 4B are respectively a schematic diagram for identifying the passage of a vehicle provided by the present application;
图5为本申请提供的统计车流量的装置50的一种逻辑结构示意图；FIG. 5 is a schematic diagram of a logical structure of a device 50 for counting vehicle traffic provided by the present application;
图6为本申请提供的计算设备12的一种硬件结构示意图。FIG. 6 is a schematic diagram of a hardware structure of a computing device 12 provided in this application.
具体实施方式detailed description
下面将结合本申请中的附图，对本申请提供的技术方案进行描述。The technical solutions provided in the present application will be described below with reference to the drawings in the present application.
术语简介Terminology introduction
车辆：是指通过轮子转动来移动车身的交通工具。例如车辆可以是非机动车或者机动车。Vehicle: means the vehicle that moves the body by turning the wheels. For example, the vehicle may be a non-motor vehicle or a motor vehicle.
车流方向：在道路上多个车辆的同向移动方向。以图1为例，在车道上行驶三辆车(A、B、C)，这三辆车是按照同方向行驶的，该方向为车流方向。Traffic flow direction: the direction in which multiple vehicles move on the road in the same direction. Taking Figure 1 as an example, three vehicles (A, B, C) are driving in the lane. These three vehicles are driving in the same direction, and this direction is the direction of traffic flow.
车流量：是指在单位时间内同方向车流通过道路上相同参考线的车辆。图1示意了该参考线，辆车(A、B)未到达该参考线，辆车C正在通过该参考线。可选地,该参考线可以是直线,或者是由多个间隔点构成的线。可选地,该参考线可以是直线，或者是其它类型的线。可选地，该参考线与车流方向垂直。Traffic: refers to vehicles that pass through the same reference line on the road in the same direction in unit time. Figure 1 illustrates the reference line. Vehicles (A, B) have not reached the reference line, and vehicle C is passing through the reference line. Optionally, the reference line may be a straight line or a line composed of a plurality of spaced points. Alternatively, the reference line may be a straight line or other types of lines. Optionally, the reference line is perpendicular to the direction of the traffic flow.
一种计算车流量的方式，在预设时间段内通过该参考线的车辆的个数为M，则车流量为M与该预设时间段的比值，M为正整数。A method for calculating the traffic flow. When the number of vehicles passing the reference line within a preset time period is M, the traffic flow is a ratio of M to the preset time period, and M is a positive integer.
本申请实施例提供了一种统计车流量的***。参见图2，在该实施例中，该***包括拍摄设备21和计算设备22。可以理解的是，在另外的实施例中，该***还可以进一步包括其他处理设备，如网络转发设备，视频处理设备等。The embodiment of the present application provides a system for counting vehicle traffic. Referring to FIG. 2, in this embodiment, the system includes a photographing device 21 and a computing device 22. It can be understood that, in another embodiment, the system may further include other processing devices, such as a network forwarding device, a video processing device, and the like.
图2所示的实施例中，该拍摄设备11用于获取监控道路上的车流量，并对通过固定道路段的车辆进行拍照。拍摄设备11基于连续拍照所得的照片形成视频流，该视频流包括多个视频帧，即该视频流中的每一帧为一个视频帧。本申请对基于多张照片形成视频流的方式不做限定；例如可以是每次拍照所得的一张照片为一个视频帧，按照时间顺序将拍照所得的多张照片合成视频流；例如，可以对拍照所得的多张照片做形成视频流的处理，一张照片对应一个视频帧，但在视频帧中保留该张图片中的车辆的特性。In the embodiment shown in FIG. 2, the photographing device 11 is configured to acquire a traffic flow on a monitored road and photograph a vehicle passing through a fixed road segment. The photographing device 11 forms a video stream based on the photos obtained by continuous photographing, and the video stream includes multiple video frames, that is, each frame in the video stream is a video frame. This application does not limit the manner of forming a video stream based on multiple photos; for example, one photo obtained from each photo may be a video frame, and multiple photos obtained from the photo are combined into a video stream in chronological order; for example, A plurality of photos obtained by taking pictures are processed to form a video stream. One photo corresponds to one video frame, but the characteristics of the vehicle in the picture are retained in the video frame.
拍摄设备11将生成的视频流发送至该***中的计算设备12。计算设备12可以使用本申请提供的方法来统计车流量。The photographing device 11 sends the generated video stream to a computing device 12 in the system. The computing device 12 may use the method provided in the present application to count the vehicle traffic.
本申请提供了一种统计车流量的方法的实施例，在该方法中可以用上述图2中所示实施例中的计算设备12统计车流量。该方法包括步骤S31、步骤S32和步骤S33，如图3所示。This application provides an embodiment of a method for counting vehicle traffic. In this method, the computing device 12 in the embodiment shown in FIG. 2 described above can be used to count vehicle traffic. The method includes steps S31, S32, and S33, as shown in FIG.
步骤S31，计算设备12从拍摄设备11获取视频流。In step S31, the computing device 12 obtains a video stream from the photographing device 11.
如上所述，该视频流是该拍摄设备11监控预设道路段的车流量所得的视频流，该视频流包括多个视频帧。As described above, the video stream is a video stream obtained by the shooting device 11 monitoring the traffic flow of a preset road segment, and the video stream includes a plurality of video frames.
可选地，计算设备12可以从拍摄设备11请求视频流，拍摄设备11将该视频流发送至计算设备12。Optionally, the computing device 12 may request a video stream from the shooting device 11, and the shooting device 11 sends the video stream to the computing device 12.
可选地，拍摄设备11可以主动将该视频流发送至计算设备12，例如定期向计算设备发送该视频流。相应地，计算设备12接收拍摄设备11发送的视频流，这样完成对该视频流的获取。Optionally, the shooting device 11 may actively send the video stream to the computing device 12, for example, periodically send the video stream to the computing device. Correspondingly, the computing device 12 receives the video stream sent by the photographing device 11 and thus completes the acquisition of the video stream.
步骤S32，计算设备12从该视频流中识别每个该视频帧中的车辆。In step S32, the computing device 12 identifies a vehicle in each video frame from the video stream.
该视频流包括多个视频帧，计算设备12针对该视频流中的每个视频帧分别作车辆识别。The video stream includes a plurality of video frames, and the computing device 12 makes a vehicle identification for each video frame in the video stream.
可选地，计算设备12部署用于识别车辆的卷积神经网络(convolution neural network，CNN)算法。计算设备12通过该CNN算法从视频帧中识别出车辆，例如识别出距离参考线最近的车辆，或者识别出所有车辆。Optionally, the computing device 12 deploys a Convolutional Neural Network (CNN) algorithm for identifying vehicles. The computing device 12 uses the CNN algorithm to identify vehicles from the video frame, for example, to identify vehicles closest to the reference line, or to identify all vehicles.
可选地，计算设备12部署用于识别车辆的候选区域(region proposal，RP)算法。计算设备12通过该RP算法从视频帧中识别出车辆，例如识别出距离上述参考线最近的车辆，或者识别出所有车辆。Optionally, the computing device 12 deploys a candidate region (RP) algorithm for identifying a vehicle. The computing device 12 uses the RP algorithm to identify vehicles from the video frame, for example, to identify the vehicle closest to the above reference line, or to identify all vehicles.
步骤S33，计算设备12按照时间顺序和车流方向，根据从该视频流中识别出的每个该视频帧中的车辆，计算通过该视频流监控到的车流量。In step S33, the computing device 12 calculates the traffic flow monitored by the video stream according to the time sequence and the direction of the traffic flow, based on the vehicles in each of the video frames identified from the video stream.
由于该视频流是拍摄设备11监控预设道路段所得的，并且在该视频流中已经识别出每个视频帧中的车辆，因此计算设备12可以按照时间顺序和车流方向，统计出该预设道路段的车流量。下面举例提供几种可以统计方式。Since the video stream is obtained by the shooting device 11 monitoring the preset road segment, and the vehicle in each video frame has been identified in the video stream, the computing device 12 can calculate the preset according to the chronological order and the direction of the traffic flow. Traffic on road segments. The following examples provide several statistical methods.
第一种可选统计方式，计算设备12可以按照时间顺序和车流方向，统计出驶入该预设道路段的车辆的个数。计算设备12可以根据统计出的个数和该视频流监控的时间段，计算该视频流监控到的该预设道路段的车流量。In a first optional statistical manner, the computing device 12 can count the number of vehicles entering the preset road segment according to the time sequence and the direction of the traffic flow. The computing device 12 may calculate the traffic volume of the preset road segment monitored by the video stream according to the counted number and the time period monitored by the video stream.
举例说明，视频流可以监控的场景范围(即预设道路段)是有限的。在确定该视频流中每个视频帧中的车辆方向时，可以从该视频流的两两相邻视频帧中识别是否有新的车辆驶入该预设道路段；如果有新的车辆驶入，则识别新驶入的车辆的个数。以此类推，可以从视频流中识别出每次新驶入的车辆，并统计每次新驶入的车辆的个数的总和。从而可以根据统计的总和和该视频流监控的时间段，计算该视频流监控到的该预设道路段的车流量。For example, the range of scenes (ie, preset road segments) that can be monitored by a video stream is limited. When determining the direction of the vehicle in each video frame in the video stream, it can be identified from the pair of adjacent video frames of the video stream whether a new vehicle has entered the preset road segment; if a new vehicle has entered , Then the number of newly driven vehicles is identified. By analogy, the newly entered vehicles can be identified from the video stream, and the total number of newly entered vehicles can be counted each time. Therefore, the traffic volume of the preset road segment monitored by the video stream can be calculated according to the sum of statistics and the time period monitored by the video stream.
第二种可选统计方式，计算设备12可以按照时间顺序和车流方向，统计出驶出该预设道路段的车辆的个数。计算设备12可以根据统计出的个数和该视频流监控的时间段，计算该视频流监控到的该预设道路段的车流量。In a second optional statistical manner, the computing device 12 may count the number of vehicles driving out of the preset road segment according to the time sequence and the direction of the traffic flow. The computing device 12 may calculate the traffic volume of the preset road segment monitored by the video stream according to the counted number and the time period monitored by the video stream.
举例说明，视频流可以监控的场景范围(即预设道路段)是有限的。在确定该视频流中每个视频帧中的车辆方向时，可以从该视频流的两两相邻视频帧中识别是否有新的车辆驶出该预设道路段；如果有新的车辆驶出，则识别新驶出的车辆的个数。以此类推，可以从视频流中识别出每次新驶出的车辆，并统计每次新驶出的车辆的个数的总和。从而可以根据统计的总和和该视频流监控的时间段，计算该视频流监控到的该预设道路段的车流量。For example, the range of scenes (ie, preset road segments) that can be monitored by a video stream is limited. When determining the direction of the vehicle in each video frame in the video stream, it can be identified from the pair of adjacent video frames of the video stream whether a new vehicle is driving out of the preset road segment; if a new vehicle is driving out , The number of newly driven vehicles is identified. By analogy, it is possible to identify each newly driven vehicle from the video stream and count the total number of newly driven vehicles each time. Therefore, the traffic volume of the preset road segment monitored by the video stream can be calculated according to the sum of statistics and the time period monitored by the video stream.
第三种可选统计方式，计算设备12可以针对该视频流中的每个视频帧，在每个该视频帧中的相同位置设定参考线，如图1所示，该参考线与该车流方向垂直。但本方式不限定该参考线在该视频帧中的具***置，例如可以设定在该视频帧中该车流方向的驶入处或者驶出处，还可以设定在该视频帧中的其它任一位置。In a third optional statistical manner, the computing device 12 may set a reference line for each video frame in the video stream at the same position in each video frame, as shown in FIG. 1, the reference line and the vehicle flow Direction is vertical. However, this method is not limited to the specific position of the reference line in the video frame. For example, it can be set at the entrance or exit of the traffic direction in the video frame, and can also be set at any other position in the video frame. position.
在设定该参考线后，计算设备12按照时间顺序和车流方向，从该视频流中识别通过该参考线的车辆。计算设备12统计通过该参考线的车辆的个数，并可以根据通过统计的个数和该视频流监控的时间段，计算该视频流监控到的该预设道路段的车流量。After setting the reference line, the computing device 12 identifies vehicles passing the reference line from the video stream in chronological order and the direction of the traffic flow. The computing device 12 counts the number of vehicles passing the reference line, and may calculate the traffic volume of the preset road segment monitored by the video stream according to the counted number and the time period monitored by the video stream.
可选地提供一种举例来实现从该视频流中识别通过该参考线的车辆。An example is optionally provided to identify a vehicle passing the reference line from the video stream.
具体地，计算设备12在该视频流中的每个视频帧中分别确定目标车辆，该目标车辆为在该视频帧中按照车流方向距离该参考线最近的车辆。在该视频帧中，计算该视频帧的目标车辆与该参考线的距离，计算出的距离作为该视频帧中的目标距离。Specifically, the computing device 12 determines a target vehicle in each video frame in the video stream, and the target vehicle is a vehicle that is closest to the reference line in the direction of the flow in the video frame. In the video frame, the distance between the target vehicle of the video frame and the reference line is calculated, and the calculated distance is used as the target distance in the video frame.
然后，计算设备12按照时间顺序，根据策略和该视频流中每个该视频帧中的目标距离，识别通过该参考线的每个车辆。该策略为：该第一视频帧和该第二视频帧为在该视频流中按照时间顺序前后相邻的两个视频帧；在第一视频帧中的目标距离小于第二视频帧中的目标距离时识别为当前有车辆通过,并且识别当前通过的车辆的个数，通过的车辆的个数可以是一个或多个，如果多个车辆同时通过，则识别同时通过的车辆的个数，如果只有一个车辆通过，则识别为通过的车辆的个数为一个。由于步骤S32已经识别出了每个视频帧中的车辆，也即识别出了车辆在视频帧中的位置，因此在根据该策略识别出有车辆通过参考线时可以进一步识别当前通过的车辆和识别当前通过的车辆的个数。Then, the computing device 12 identifies each vehicle passing the reference line in chronological order according to a policy and a target distance in each of the video frames in the video stream. The strategy is: the first video frame and the second video frame are two video frames adjacent to each other in time sequence in the video stream; the target distance in the first video frame is smaller than the target in the second video frame When the distance is recognized, there are currently passing vehicles, and the number of currently passing vehicles can be identified. The number of passing vehicles can be one or more. If multiple vehicles pass at the same time, the number of vehicles passing at the same time is identified. If only one vehicle passes, the number of vehicles identified as passing is one. Since the vehicle in each video frame has been identified in step S32, that is, the position of the vehicle in the video frame has been identified, so when a vehicle is passed through the reference line according to the strategy, the currently passing vehicle and the identification can be further identified. The number of vehicles currently passing.
可选地提供一种举例来实现从该视频流中识别通过该参考线的车辆。An example is optionally provided to identify a vehicle passing the reference line from the video stream.
具体地，计算设备12在该视频流中的每个视频帧中，分别计算该视频帧中每个车辆与该参考线的距离，从计算出的每个车辆与该参考线的距离中选择最短距离，将选择的最短距离作为该视频帧中的目标距离，将该最短距离对应的车辆作为该视频帧中的目标车辆。在同一视频帧中，如果多个车辆均具有距离该参考线的该目标距离，则该多个车辆同时作为多个目标车辆，即该多个车辆中的每个车辆均为目标车辆；如果只有一个车辆具有距离该参考线的该目标距离，则该个车辆作为目标车辆。Specifically, the computing device 12 calculates the distance between each vehicle and the reference line in each video frame in the video stream, and selects the shortest distance from the calculated distance between each vehicle and the reference line. Distance, using the selected shortest distance as the target distance in the video frame, and the vehicle corresponding to the shortest distance as the target vehicle in the video frame. In the same video frame, if multiple vehicles have the target distance from the reference line, the multiple vehicles simultaneously serve as multiple target vehicles, that is, each of the multiple vehicles is the target vehicle; if only If a vehicle has the target distance from the reference line, the vehicle is regarded as the target vehicle.
然后，计算设备12按照时间顺序，根据策略和该视频流中每个该视频帧中的目标距离，识别通过该参考线的每个车辆。Then, the computing device 12 identifies each vehicle passing the reference line in chronological order according to a policy and a target distance in each of the video frames in the video stream.
以此类推，计算设备12通过该策略可以识别出该视频流中通过该参考线的所有车辆。根据通过该参考线的所有车辆的个数，可以计算出该视频流监控到的该预设道路段的车流量。By analogy, the computing device 12 can identify all vehicles passing the reference line in the video stream through the strategy. According to the number of all vehicles passing the reference line, the traffic volume of the preset road segment monitored by the video stream can be calculated.
一种可选实施方式中，该策略还包括：在第一视频帧中的目标距离大于或等于第二视频帧中的目标距离时识别为没有一个车辆通过，该第一视频帧和该第二视频帧为在该视频流中按照时间顺序前后相邻的两个视频帧。这样，计算设备12通过该策略，按照时间顺序，基于该视频流中每个该视频帧中的目标距离，可以准确识别出是否有车辆通过该参考线。In an optional implementation manner, the strategy further includes: identifying that no vehicle passes when the target distance in the first video frame is greater than or equal to the target distance in the second video frame, the first video frame and the second video frame The video frames are two video frames adjacent to each other in the video stream in chronological order. In this way, the computing device 12 can accurately identify whether a vehicle passes the reference line based on the target distance in each of the video frames in the video stream in chronological order through the strategy.
以此类推，通过该策略可以更精确地识别出该视频流中通过该参考线的所有车辆的个数。By analogy, through this strategy, the number of all vehicles passing the reference line in the video stream can be more accurately identified.
举例说明，如图4A所示，第I帧、第J帧和第K帧为视频流中按照时间顺序依次排列的三个视频帧。在第I帧中，按照车流方向，相较于车辆A和车辆B，车辆C的车头距离参考线最近，车辆C的车头与参考线的距离为10米(meter，m)。在第J帧中，按照车流方向，相较于车辆A和车辆B，车辆C的车头距离参考线最近，车辆C的车头与参考线的距离为5米。在第K帧中，按照车流方向，车辆C的车头已经通过参考线，因此不 再考虑车辆C与参考线的距离，只比较车辆A和车辆B；这时，相较于车辆A，车辆B的车头距离参考线最近，车辆B的车头与参考线的距离为15米。由于在第K帧中车辆B与参考线的距离(15米)大于在第J帧中车辆C与参考线的距离(5米)，因此识别为有车辆通过，并识别通过的车辆为车辆C，即通过的车辆的个数为一个。For example, as shown in FIG. 4A, the I frame, the J frame, and the K frame are three video frames in the video stream arranged in time sequence. In the first frame, according to the direction of the traffic flow, the head of the vehicle C is closest to the reference line compared to the vehicle A and the vehicle B, and the distance between the head of the vehicle C and the reference line is 10 meters (meter, m). In the Jth frame, according to the direction of the traffic flow, the head of the vehicle C is closest to the reference line compared to the vehicle A and the vehicle B, and the distance between the head of the vehicle C and the reference line is 5 meters. In frame K, according to the direction of traffic flow, the head of vehicle C has passed the reference line, so the distance between vehicle C and the reference line is no longer considered, and only vehicle A and vehicle B are compared; at this time, compared with vehicle A, vehicle B The front of the vehicle is closest to the reference line, and the distance between the front of vehicle B and the reference line is 15 meters. Because the distance between vehicle B and the reference line (15 meters) in frame K is greater than the distance between vehicle C and the reference line (5 meters) in frame J, it is recognized that a vehicle has passed, and the vehicle that has passed is identified as vehicle C , That is, the number of passing vehicles is one.
举例说明，如图4B所示，第I帧、第J帧和第K帧为视频流中按照时间顺序依次排列的三个视频帧。在第I帧中，按照车流方向，相较于车辆A，车辆B和车辆C的车头距离参考线最近，车辆B的车头与参考线的距离为10米，同时车辆C的车头与参考线的距离为10米。在第J帧中，按照车流方向，相较于车辆A，车辆B和车辆C的车头距离参考线最近，车辆B的车头与参考线的距离为5米，同时车辆C的车头与参考线的距离为5米。在第K帧中，按照车流方向，车辆B的车头和车辆C的车头同时通过参考线，因此不再考虑车辆B和车辆C分别与参考线的距离；这时，车辆A的车头距离参考线最近，车辆A的车头与参考线的距离为10米。由于在第K帧中车辆A与参考线的距离(10米)大于在第J帧中车辆C与参考线的距离(5米)，因此识别为有车辆通过，并识别通过的车辆为车辆B和车辆C，即为通过的车辆的个数为两个。For example, as shown in FIG. 4B, the I frame, the J frame, and the K frame are three video frames in the video stream arranged in time sequence. In the first frame, according to the direction of the traffic flow, the heads of vehicles B and C are closest to the reference line compared to vehicle A. The distance between the head of vehicle B and the reference line is 10 meters. The distance is 10 meters. In frame J, according to the direction of traffic flow, the heads of vehicles B and C are closest to the reference line compared to vehicle A. The distance between the head of vehicle B and the reference line is 5 meters, and the distance between the head of vehicle C and the reference line The distance is 5 meters. In frame K, the head of vehicle B and the head of vehicle C pass the reference line at the same time according to the direction of the traffic, so the distances between vehicle B and vehicle C from the reference line are no longer considered; at this time, the head of vehicle A is away from the reference line The distance between the front of the vehicle A and the reference line was 10 meters. Since the distance between vehicle A and the reference line (10 meters) in frame K is greater than the distance between vehicle C and the reference line (5 meters) in frame J, it is recognized that a vehicle has passed, and the vehicle that has passed is identified as vehicle B. And vehicle C, that is, the number of passing vehicles is two.
可选地，在该视频流中的每个视频帧中，计算设备12计算车辆的车头与该参考线的距离，将计算所得的距离作为该车辆距离该参考线的距离。举例说明，在每个该视频帧中，计算该目标车辆的车头到该参考线的距离，计算出的距离为该视频帧中的目标距离。Optionally, in each video frame in the video stream, the computing device 12 calculates the distance between the head of the vehicle and the reference line, and uses the calculated distance as the distance of the vehicle from the reference line. For example, in each video frame, the distance from the head of the target vehicle to the reference line is calculated, and the calculated distance is the target distance in the video frame.
可选地，在该视频流中的每个视频帧中，计算设备12计算车辆的车尾与该参考线的距离，将计算所得的距离作为该车辆距离该参考线的距离。Optionally, in each video frame in the video stream, the computing device 12 calculates the distance between the tail of the vehicle and the reference line, and uses the calculated distance as the distance of the vehicle from the reference line.
可选地，在该视频流中的每个视频帧中，从每个车辆上选择一个基准点(可以位于该车辆的中间位置或者其它任一位置)，计算设备12计算车辆的基准点与该参考线的距离，将计算所得的距离作为该车辆距离该参考线的距离。Optionally, in each video frame in the video stream, a reference point is selected from each vehicle (may be located at the middle position of the vehicle or any other position), and the computing device 12 calculates the reference point of the vehicle and the The distance of the reference line. The calculated distance is taken as the distance of the vehicle from the reference line.
本申请提供一种统计车流量的装置，该装置部署在上述图2所示实施例中的计算设备12上，该装置包括的功能单元用于实现上述统计车流量的方法中的步骤；本申请实施例对在该装置中如何划分功能单元不做限定，下面实例性地提供一种功能单元的划分，如图5所示。This application provides a device for counting vehicle traffic. The device is deployed on the computing device 12 in the embodiment shown in FIG. 2. The device includes functional units for implementing steps in the method for counting vehicle traffic. The embodiment does not limit how to divide the functional units in the device, and a division of the functional units is provided as an example below, as shown in FIG. 5.
如图5所示的统计车流量的装置50，包括：The device 50 for counting vehicle traffic shown in FIG. 5 includes:
获取单元51，用于获取监控车流量的视频流，所述视频流包括多个视频帧；；An obtaining unit 51, configured to obtain a video stream of monitored vehicle traffic, where the video stream includes multiple video frames;
识别单元52，用于从所述视频流中识别每个所述视频帧中的车辆；An identification unit 52, configured to identify a vehicle in each of the video frames from the video stream;
计算单元53，用于按照时间顺序和车流方向，根据从所述视频流中识别出的每个所述视频帧中的车辆，计算通过所述视频流监控到的车流量。The calculation unit 53 is configured to calculate, according to the time sequence and the direction of the traffic flow, the traffic volume monitored by the video flow according to the vehicles in each of the video frames identified from the video flow.
可选地，所述计算单元53，用于：Optionally, the calculation unit 53 is configured to:
在每个所述视频帧中的相同位置设定参考线，所述参考线与所述车流方向垂直；Setting a reference line at the same position in each of the video frames, the reference line being perpendicular to the direction of the traffic flow;
按照时间顺序和车流方向，从所述视频流中识别通过所述参考线的车辆的个数。The number of vehicles passing the reference line is identified from the video stream in chronological order and the direction of the vehicle flow.
可选地，所述计算单元53，用于：Optionally, the calculation unit 53 is configured to:
计算在所述视频流中每个所述视频帧中的目标距离，所述目标距离为所述视频帧中的目标车辆与所述参考线的距离，所述目标车辆为在所述视频帧中按照所述车流方向距离与所述参考线最近的车辆；Calculating a target distance in each of the video frames in the video stream, the target distance being a distance between a target vehicle in the video frame and the reference line, and the target vehicle being in the video frame The vehicle closest to the reference line in the direction of the traffic flow;
按照时间顺序，根据策略和所述视频流中每个所述视频帧中的目标距离，识别通过所述参考线的每个车辆，所述策略为：在第一视频帧中的目标距离小于第二视频帧中的目标距离时识别当前通过的车辆，所述第一视频帧和所述第二视频帧为在所述视频流中按照时间顺序前后相邻的两个视频帧。According to the chronological order, each vehicle passing through the reference line is identified according to a strategy and a target distance in each of the video frames in the video stream, the strategy is: the target distance in the first video frame is less than the first When the target distance in the two video frames is recognized, the currently passing vehicle is identified, and the first video frame and the second video frame are two video frames adjacent to each other in time sequence in the video stream.
可选地，所述计算单元53，用于在每个所述视频帧中，计算所述目标车辆的车头到所述参考线的距离，计算出的距离为所述视频帧中的目标距离。Optionally, the calculation unit 53 is configured to calculate a distance from a head of the target vehicle to the reference line in each of the video frames, and the calculated distance is a target distance in the video frame.
获取单元51、识别单元52和计算单元53中各功能，在上述方法中有相应的步骤。因此，获取单元51、识别单元52和计算单元53中各功能的实现细节，可以参见上述方法中相应步骤的描述。Each function in the acquisition unit 51, the recognition unit 52, and the calculation unit 53 has corresponding steps in the above method. Therefore, for the implementation details of the functions in the acquisition unit 51, the identification unit 52, and the calculation unit 53, refer to the description of the corresponding steps in the foregoing method.
下面示例性地提供该计算设备12的一种可能的基本硬件架构，如图6所示。A possible basic hardware architecture of the computing device 12 is provided as an example below, as shown in FIG. 6.
参见图6，计算设备12包括处理器121、存储器122、通信接口123和总线124。6, the computing device 12 includes a processor 121, a memory 122, a communication interface 123, and a bus 124.
计算设备12中，处理器121的数量可以是一个或多个，图1仅示意了其中一个处理器121。可选地，处理器121可以是中央处理器(central processing unit，CPU)或者ARM处理器。如果计算设备12具有多个处理器121，多个处理器121的类型可以不同，或者可以相同。可选地，计算设备12的多个处理器121还可以集成为多核处理器。In the computing device 12, the number of processors 121 may be one or more, and FIG. 1 illustrates only one of the processors 121. Optionally, the processor 121 may be a central processing unit (CPU) or an ARM processor. If the computing device 12 has multiple processors 121, the types of the multiple processors 121 may be different or may be the same. Optionally, the multiple processors 121 of the computing device 12 may also be integrated into a multi-core processor.
存储器122存储计算机指令；例如，该计算机指令包括链代码；例如,该计算机指令用于实现本申请提供的方法中的各个步骤；例如,该计算机指令用于实现本申请提供的装置50包括的各功能单元，或者用于实现本申请提供的方法中的各个步骤。The memory 122 stores computer instructions; for example, the computer instructions include a chain code; for example, the computer instructions are used to implement each step in the method provided by the present application; for example, the computer instructions are used to implement each of the steps included in the apparatus 50 provided in the present application. A functional unit, or used to implement each step in the method provided in this application.
存储器122可以是以下存储介质的任一种或任一种组合：非易失性存储器(non-volatile memory，NVM)(例如只读存储器(read only memory，ROM)、固态硬盘(Solid State Drives，SSD)、机械硬盘、磁盘、磁盘整列)，易失性存储器(volatile memory)。The memory 122 may be any one or any combination of the following storage media: non-volatile memory (NVM) (such as read-only memory (ROM), solid state drives (Solid State Drives, SSD), mechanical hard disks, magnetic disks, and entire arrays of disks), volatile memory (volatile memory).
通信接口123可以是以下器件的任一种或任一种组合：网络接口(例如以太网接口)、无线网卡等具有网络接入功能的器件。The communication interface 123 may be any one or any combination of the following devices: a network interface (such as an Ethernet interface), a wireless network card, and other devices having a network access function.
通信接口123用于计算设备12与其他设备(例如计算设备)进行数据通信。The communication interface 123 is used for data communication between the computing device 12 and other devices (for example, a computing device).
图6用一条粗线表示总线124。处理器121、存储器122和通信接口123通过总线124连接。这样，处理器121可以通过总线124访问存储器122,以及通过总线124利用通信接口123与其他设备(例如计算设备)进行数据交互。FIG. 6 shows the bus 124 with a thick line. The processor 121, the memory 122, and the communication interface 123 are connected through a bus 124. In this way, the processor 121 can access the memory 122 through the bus 124 and use the communication interface 123 to perform data interaction with other devices (for example, computing devices) through the bus 124.
可选地，计算设备12执行存储器122中的计算机指令,在计算设备12上实现本申请提供的统计车流量的方法，或者在计算设备12上实现本申请提供的装置50。Optionally, the computing device 12 executes the computer instructions in the memory 122 to implement the method for counting vehicle traffic provided in the present application on the computing device 12 or implement the device 50 provided in the present application on the computing device 12.
可选地，计算设备12为公有云或者私有云或者混合云中的服务器。计算设备12的资源被虚拟化以后，在该虚拟化的资源上部署装置。该装置用于实现本申请提供的统计车流量的方法，或者该装置为本申请提供的装置50。Optionally, the computing device 12 is a server in a public cloud, a private cloud, or a hybrid cloud. After the resources of the computing device 12 are virtualized, a device is deployed on the virtualized resources. This device is used to implement the method for counting vehicle traffic provided in this application, or this device is the device 50 provided in this application.
本申请提供一种计算机可读存储介质，该计算机可读存储介质中存储有计算机指令，当计算设备12的处理器121执行该计算机指令时，该计算设备12执行本申请提供的统计车流量的方法中的步骤。The present application provides a computer-readable storage medium. The computer-readable storage medium stores computer instructions. When the processor 121 of the computing device 12 executes the computer instructions, the computing device 12 executes the statistics of vehicle traffic provided by the application. Steps in the method.
本申请提供一种计算机可读存储介质，该计算机可读存储介质中存储有计算机指令，该计算机指令用于实现装置50。The application provides a computer-readable storage medium. The computer-readable storage medium stores computer instructions, and the computer instructions are used to implement the device 50.
本申请提供一种计算机程序产品，该计算机程序产品包括计算机指令，该计算机指令存储在计算机可读存储介质中。计算设备的处理器可以从计算机可读存储介质读取并执行该计算机指令，使得该计算设备执行本申请提供的统计车流量的方法的步骤。The application provides a computer program product including computer instructions, the computer instructions being stored in a computer-readable storage medium. The processor of the computing device may read and execute the computer instructions from the computer-readable storage medium, so that the computing device executes the steps of the method for counting vehicle traffic provided by the present application.
本申请提供一种计算机程序产品，该计算机程序产品包括的计算机指令用于实现装置50。The application provides a computer program product. The computer program product includes computer instructions for implementing the device 50.
以上实施例仅用以说明本发明的技术方案，而非对其限制；尽管参照前述实施例对本发明进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本发明各实施例技术方案的保护范围。The above embodiments are only used to illustrate the technical solutions of the present invention, but not limited to them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still describe The recorded technical solutions are modified, or some of the technical features are equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions outside the protection scope of the technical solutions of the embodiments of the present invention.

Claims (10)

  1. 一种统计车流量的方法，其特征在于，所述方法包括：A method for counting vehicle traffic, characterized in that the method includes:
    获取监控车流量的视频流，所述视频流包括多个视频帧；Obtaining a video stream for monitoring vehicle traffic, the video stream including multiple video frames;
    从所述视频流中识别每个所述视频帧中的车辆；Identifying a vehicle in each of the video frames from the video stream;
    按照时间顺序和车流方向，根据从所述视频流中识别出的每个所述视频帧中的车辆，计算通过所述视频流监控到的车流量。According to the time sequence and the direction of the traffic flow, the traffic flow monitored through the video stream is calculated according to the vehicles in each of the video frames identified from the video stream.
  2. 根据权利要求1所述的方法，其特征在于，所述计算通过所述视频流监控到的车流量，包括：The method according to claim 1, wherein the calculating the vehicle flow monitored through the video stream comprises:
    在每个所述视频帧中的相同位置设定参考线，所述参考线与所述车流方向垂直；Setting a reference line at the same position in each of the video frames, the reference line being perpendicular to the direction of the traffic flow;
    按照时间顺序和车流方向，从所述视频流中识别通过所述参考线的车辆的个数。The number of vehicles passing the reference line is identified from the video stream in chronological order and the direction of the vehicle flow.
  3. 根据权利要求2所述的方法，其特征在于，所述按照时间顺序和车流方向从所述视频流中识别通过所述参考线的车辆的个数包括：The method according to claim 2, wherein the number of vehicles that pass through the reference line from the video stream according to the chronological order and the direction of the vehicle flow comprises:
    计算在所述视频流中每个所述视频帧中的目标距离，所述目标距离为所述视频帧中的目标车辆与所述参考线的距离，所述目标车辆为在所述视频帧中按照所述车流方向距离与所述参考线最近的车辆；Calculating a target distance in each of the video frames in the video stream, the target distance being a distance between a target vehicle in the video frame and the reference line, and the target vehicle being in the video frame The vehicle closest to the reference line in the direction of the traffic flow;
    按照时间顺序，根据策略和所述视频流中每个所述视频帧中的目标距离，识别通过所述参考线的每个车辆，所述策略为：在第一视频帧中的目标距离小于第二视频帧中的目标距离时识别当前通过的车辆，所述第一视频帧和所述第二视频帧为在所述视频流中按照时间顺序前后相邻的两个视频帧。According to the chronological order, each vehicle passing through the reference line is identified according to a strategy and a target distance in each of the video frames in the video stream, the strategy is: the target distance in the first video frame is less than the first When the target distance in the two video frames is recognized, the currently passing vehicle is identified, and the first video frame and the second video frame are two video frames adjacent to each other in time sequence in the video stream.
  4. 根据权利要求3所述的方法，其特征在于，所述计算在所述视频流中每个所述视频帧中的目标距离，包括：The method according to claim 3, wherein said calculating a target distance in each of said video frames in said video stream comprises:
    在每个所述视频帧中，计算所述目标车辆的车头到所述参考线的距离，计算出的距离为所述视频帧中的目标距离。In each of the video frames, the distance from the head of the target vehicle to the reference line is calculated, and the calculated distance is the target distance in the video frame.
  5. 一种统计车流量的装置，其特征在于，所述装置包括：A device for counting vehicle traffic, characterized in that the device includes:
    获取单元，用于获取监控车流量的视频流，所述视频流包括多个视频帧；An obtaining unit, configured to obtain a video stream of monitored vehicle traffic, where the video stream includes multiple video frames;
    识别单元，用于从所述视频流中识别每个所述视频帧中的车辆；A recognition unit, configured to identify a vehicle in each of the video frames from the video stream;
    计算单元，用于按照时间顺序和车流方向，根据从所述视频流中识别出的每个所述视频帧中的车辆，计算通过所述视频流监控到的车流量。A calculation unit is configured to calculate, according to the time sequence and the direction of the traffic flow, the traffic volume monitored by the video flow according to the vehicles in each of the video frames identified from the video flow.
  6. 根据权利要求5所述的装置，其特征在于，所述计算单元，用于：The device according to claim 5, wherein the calculation unit is configured to:
    在每个所述视频帧中的相同位置设定参考线，所述参考线与所述车流方向垂直；Setting a reference line at the same position in each of the video frames, the reference line being perpendicular to the direction of the traffic flow;
    按照时间顺序和车流方向，从所述视频流中识别通过所述参考线的车辆的个数。The number of vehicles passing the reference line is identified from the video stream in chronological order and the direction of the vehicle flow.
  7. 根据权利要求6所述的装置，其特征在于，所述计算单元，用于：The apparatus according to claim 6, wherein the calculation unit is configured to:
    计算在所述视频流中每个所述视频帧中的目标距离，所述目标距离为所述视频帧中的目标车辆与所述参考线的距离，所述目标车辆为在所述视频帧中按照所述车流方向距离与所述参考线最近的车辆；Calculating a target distance in each of the video frames in the video stream, the target distance being a distance between a target vehicle in the video frame and the reference line, and the target vehicle being in the video frame The vehicle closest to the reference line in the direction of the traffic flow;
    按照时间顺序，根据策略和所述视频流中每个所述视频帧中的目标距离，识别通过所述参考线的每个车辆，所述策略为：在第一视频帧中的目标距离小于第二视频帧中的目标距离时识别当前通过的车辆，所述第一视频帧和所述第二视频帧为在所述视频流中按照时间顺序前后相邻的两个视频帧。According to the chronological order, each vehicle passing through the reference line is identified according to a strategy and a target distance in each of the video frames in the video stream, the strategy is: the target distance in the first video frame is less than the first When the target distance in the two video frames is recognized, the currently passing vehicle is identified, and the first video frame and the second video frame are two video frames adjacent to each other in time sequence in the video stream.
  8. 根据权利要求7所述的装置，其特征在于，The device according to claim 7, wherein:
    所述计算单元，用于在每个所述视频帧中，计算所述目标车辆的车头到所述参考线的距离，计算出的距离为所述视频帧中的目标距离。The calculation unit is configured to calculate a distance from a head of the target vehicle to the reference line in each of the video frames, and the calculated distance is a target distance in the video frame.
  9. 一种计算设备，其特征在于，包括处理器和存储器；A computing device, comprising a processor and a memory;
    所述存储器，用于存储计算机指令；The memory is configured to store computer instructions;
    所述处理器，用于执行所述存储器存储的计算机指令，使得所述计算设备执行权利要求1至4任一项所述的统计车流量的方法。The processor is configured to execute computer instructions stored in the memory, so that the computing device executes the method for counting vehicle traffic according to any one of claims 1 to 4.
  10. 一种计算机可读存储介质，所述计算机可读存储介质存储计算机指令，所述计算机指令指示计算设备执行权利要求1至4任一项所述的统计车流量的方法。A computer-readable storage medium storing computer instructions, the computer instructions instructing a computing device to execute the method for counting vehicle traffic according to any one of claims 1 to 4.
PCT/CN2019/087229 2018-09-05 2019-05-16 Method and apparatus for counting vehicle flow, computing device, and computer readable storage medium WO2020048156A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811031635.2 2018-09-05
CN201811031635.2A CN109377770B (en) 2018-09-05 2018-09-05 Method and device for counting traffic flow, computing equipment and storage medium

Publications (1)

Publication Number Publication Date
WO2020048156A1 true WO2020048156A1 (en) 2020-03-12

Family

ID=65404581

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/087229 WO2020048156A1 (en) 2018-09-05 2019-05-16 Method and apparatus for counting vehicle flow, computing device, and computer readable storage medium

Country Status (2)

Country Link
CN (1) CN109377770B (en)
WO (1) WO2020048156A1 (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109377770B (en) * 2018-09-05 2021-06-22 华为技术有限公司 Method and device for counting traffic flow, computing equipment and storage medium
CN111046820A (en) * 2019-12-17 2020-04-21 上海船舶研究设计院（中国船舶工业集团公司第六0四研究院） Statistical method and device for vehicles in automobile roll-on-roll-off ship and intelligent terminal
CN111369795B (en) * 2020-03-09 2022-11-08 深圳大学 Traffic flow statistical method, device, equipment and storage medium
CN112950961B (en) * 2021-01-27 2022-07-08 苏州智芯控联信息科技有限公司 Traffic flow statistical method, device, equipment and storage medium

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101510358A (en) * 2009-03-20 2009-08-19 吉林大学 Method and apparatus for processing real time statistical vehicle flowrate using video image
JP2011198030A (en) * 2010-03-19 2011-10-06 Mitsubishi Electric Corp Traffic flow measuring apparatus
CN102708690A (en) * 2012-06-20 2012-10-03 中国计量学院 Method for measuring road noise on basis of road monitoring video
WO2013187748A1 (en) * 2012-06-12 2013-12-19 Institute Of Electronics And Computer Science System and method for video-based vehicle detection
CN109377770A (en) * 2018-09-05 2019-02-22 华为技术有限公司 The method and apparatus of statistical vehicle flowrate calculate equipment and storage medium

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001283373A (en) * 2000-03-30 2001-10-12 Toshiba Corp Traffic flow measuring system
JP5583067B2 (en) * 2011-03-31 2014-09-03 中日本高速道路株式会社 Vehicle detector
CN104021679A (en) * 2014-05-30 2014-09-03 宁波海视智能***有限公司 Traffic flow statistic method based on video
CN104599502B (en) * 2015-02-13 2017-01-25 重庆邮电大学 Method for traffic flow statistics based on video monitoring
EP3413018B1 (en) * 2016-02-26 2020-08-26 Huawei Technologies Co., Ltd. Traffic control method and device
CN106097726A (en) * 2016-08-23 2016-11-09 苏州科达科技股份有限公司 The detection determination in region, traffic information detection method and device
CN106710228B (en) * 2016-12-21 2019-06-21 南京大学镇江高新技术研究院 A kind of implementation method of passenger-cargo shunting traffic parameter monitoring system

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101510358A (en) * 2009-03-20 2009-08-19 吉林大学 Method and apparatus for processing real time statistical vehicle flowrate using video image
JP2011198030A (en) * 2010-03-19 2011-10-06 Mitsubishi Electric Corp Traffic flow measuring apparatus
WO2013187748A1 (en) * 2012-06-12 2013-12-19 Institute Of Electronics And Computer Science System and method for video-based vehicle detection
CN102708690A (en) * 2012-06-20 2012-10-03 中国计量学院 Method for measuring road noise on basis of road monitoring video
CN109377770A (en) * 2018-09-05 2019-02-22 华为技术有限公司 The method and apparatus of statistical vehicle flowrate calculate equipment and storage medium

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
MAO. XIN: "Research and Implementation of Traffic Detection Method Based on Vehicle Identification", ELECTRONIC TECHNOLOGY & LNJC1RMATION SCIENCE, CHINA MASTER'S THESES FULL-TEXT DATABASE, no. 2, 15 February 2018 (2018-02-15), pages 6, 10, 21-28, 39, 41, 51-56 and 69, ISSN: 1674-0246 *
QIAN, LIANG: "The Design and Implementation of Traffic Flow Detection System Based on DMCU", SCIENCE-ENGINEERING (B), CHINESE SELECTED DOCTORAL DISSERTATIONS AND MASTER'S THESES FULL-TEXT DATABASES (MASTER), no. 7, 15 July 2006 (2006-07-15), pages 9-11, 20-23, 34-35 and 44-48, ISSN: 1671-6779 *

Also Published As

Publication number Publication date
CN109377770B (en) 2021-06-22
CN109377770A (en) 2019-02-22

Similar Documents

Publication Publication Date Title
WO2020048156A1 (en) Method and apparatus for counting vehicle flow, computing device, and computer readable storage medium
US10289918B2 (en) Method and apparatus for detecting a speed of an object
WO2022143549A1 (en) Expressway road condition monitoring method and apparatus based on toll collection data
JP7081438B2 (en) Object speed estimation method and equipment and image processing equipment
US20130093895A1 (en) System for collision prediction and traffic violation detection
CN105761494B (en) Abnormal traffic information collecting method based on unmanned plane
CN104021682B (en) Oversaturated intersection selfreparing control method
KR20170035922A (en) Techniques for automatic real-time calculation of user wait times
KR101951534B1 (en) Method and apparatus of integrated enforcement for traffic cctv enforcement system
Sen et al. Accurate speed and density measurement for road traffic in India
JP2017525064A (en) System and method for activity monitoring using video data
US9760783B2 (en) Vehicle occupancy detection using passenger to driver feature distance
CN113380027B (en) Intersection traffic state parameter estimation method and system based on multi-source data
WO2023019936A1 (en) Vehicle driving behavior detection method and apparatus, and device and storage medium
EP4020428A1 (en) Method and apparatus for recognizing lane, and computing device
CN111613070A (en) Traffic signal lamp control method, traffic signal lamp control device, electronic equipment and computer storage medium
JP2022105203A (en) Information processing device, control method and program
CN113380021B (en) Vehicle state detection method, device, server and computer readable storage medium
JP2022545322A (en) Road operation situation determination method and device, device, storage medium and program
WO2024067210A1 (en) Traffic state determination method and apparatus, and device
CN111402615B (en) Variable lane control method based on navigation information
CN113673311A (en) Traffic abnormal event detection method, equipment and computer storage medium
US10429200B1 (en) Determining adjusted trip duration using route features
CN111435568B (en) Regional traffic control method and device
CN115440063B (en) Traffic signal lamp control method, device, computer equipment and storage medium

Legal Events

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

Ref document number: 19858203

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19858203

Country of ref document: EP

Kind code of ref document: A1