WO2020052465A1 - 路况信息处理系统及其信息处理方法及装置 - Google Patents

路况信息处理系统及其信息处理方法及装置 Download PDF

Info

Publication number
WO2020052465A1
WO2020052465A1 PCT/CN2019/104124 CN2019104124W WO2020052465A1 WO 2020052465 A1 WO2020052465 A1 WO 2020052465A1 CN 2019104124 W CN2019104124 W CN 2019104124W WO 2020052465 A1 WO2020052465 A1 WO 2020052465A1
Authority
WO
WIPO (PCT)
Prior art keywords
rsu
surveillance
target
road
camera
Prior art date
Application number
PCT/CN2019/104124
Other languages
English (en)
French (fr)
Inventor
蔡岭
陈颖
孙海洋
王刚
Original Assignee
阿里巴巴集团控股有限公司
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 阿里巴巴集团控股有限公司 filed Critical 阿里巴巴集团控股有限公司
Publication of WO2020052465A1 publication Critical patent/WO2020052465A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G08SIGNALLING
    • G08GTRAFFIC CONTROL SYSTEMS
    • G08G1/00Traffic control systems for road vehicles
    • G08G1/01Detecting movement of traffic to be counted or controlled
    • G08G1/0104Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0108Measuring and analyzing of parameters relative to traffic conditions based on the source of data
    • G08G1/0116Measuring and analyzing of parameters relative to traffic conditions based on the source of data from roadside infrastructure, e.g. beacons
    • GPHYSICS
    • G08SIGNALLING
    • G08GTRAFFIC CONTROL SYSTEMS
    • G08G1/00Traffic control systems for road vehicles
    • GPHYSICS
    • G08SIGNALLING
    • G08GTRAFFIC CONTROL SYSTEMS
    • G08G1/00Traffic control systems for road vehicles
    • G08G1/01Detecting movement of traffic to be counted or controlled
    • GPHYSICS
    • G08SIGNALLING
    • G08GTRAFFIC CONTROL SYSTEMS
    • G08G1/00Traffic control systems for road vehicles
    • G08G1/01Detecting movement of traffic to be counted or controlled
    • G08G1/04Detecting movement of traffic to be counted or controlled using optical or ultrasonic detectors
    • GPHYSICS
    • G08SIGNALLING
    • G08GTRAFFIC CONTROL SYSTEMS
    • G08G1/00Traffic control systems for road vehicles
    • G08G1/09Arrangements for giving variable traffic instructions
    • G08G1/0962Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
    • G08G1/0967Systems involving transmission of highway information, e.g. weather, speed limits
    • GPHYSICS
    • G08SIGNALLING
    • G08GTRAFFIC CONTROL SYSTEMS
    • G08G1/00Traffic control systems for road vehicles
    • G08G1/09Arrangements for giving variable traffic instructions
    • G08G1/0962Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
    • G08G1/0967Systems involving transmission of highway information, e.g. weather, speed limits
    • G08G1/096766Systems involving transmission of highway information, e.g. weather, speed limits where the system is characterised by the origin of the information transmission
    • G08G1/096783Systems involving transmission of highway information, e.g. weather, speed limits where the system is characterised by the origin of the information transmission where the origin of the information is a roadside individual element

Definitions

  • the present application relates to the technical field of road condition information processing, and particularly to a road condition information processing system and an information processing method and device thereof.
  • highly intelligent autonomous vehicles can effectively sense their surroundings, obtain information about the type, location, orientation, size, and speed of nearby entities, and make reasonable decisions based on their driving status.
  • the driving decision of a vehicle completes the task of driving a vehicle, thereby replacing manual driving labor and realizing a highly automated driving process.
  • vehicle sensor network sensing there are two common autonomous driving sensing systems: vehicle sensor network sensing and vehicle communication network sensing.
  • vehicle sensor network sensing In the way of in-vehicle sensor network sensing, autonomous vehicles need to be equipped with a large number of sensors (radar, camera, etc.) in order to obtain sufficient and diverse sensor information, coupled with powerful computing equipment to process data in real time to obtain complete perception results quickly .
  • sensors radar, camera, etc.
  • computing equipment to process data in real time to obtain complete perception results quickly .
  • V2X Vehicle-to-everything
  • V2X Vehicle-to-everything
  • V2X is a security system that implements two-way communication and multi-directional communication. Similar to the WIFI connection method, it can allow infrastructure between vehicles (V2V), between people (V2P), and vehicles and traffic lights. Between (V2I), send signals to send information like position, speed, obstacle, danger, etc. to the other party, thereby improving driving safety.
  • V2I is one of the important components. Because the infrastructure (for example, roadside equipment RSU, etc.) can be erected at a relatively high position, it can pass through a higher angle of view, a longer detection distance, and a dead space. Get more information about your surroundings.
  • traffic participants such as vehicles, broadcast their own location and speed information. RSUs or other traffic participants can realize the perception of vehicles by receiving broadcast messages from the vehicles. At the same time, the RSU can use the traffic it receives. Participants' information is broadcast in the form of data packets, etc.
  • the on-board unit OBU in the vehicle receives the data packets of the RSU to obtain information about the surrounding environment, and then makes driving decisions.
  • this method relies on the reporting of vehicle information. Once any vehicle cannot report its own position, it means that it cannot be perceived by other vehicles, and thus becomes a hidden safety hazard.
  • the present application provides a road condition information processing system and an information processing method and device thereof, which can achieve more accurate target perception and improve system security.
  • a road condition information processing system A road condition information processing system
  • the system includes a plurality of roadside equipment RSUs deployed along the road;
  • the RSU is equipped with an image sensor and a data processing module
  • the image sensor includes a plurality of surveillance cameras for collecting information on a target on a road; wherein the plurality of surveillance cameras include a first surveillance camera in a first surveillance direction and a second surveillance in a second surveillance direction.
  • a camera, the first monitoring camera is used to collect back image data of a target in a specified road section, and the second monitoring camera is used to collect forward image data of a target in a specified road section;
  • the data processing module is configured to sense a target in a designated road segment by collecting information collection results of a surveillance camera in the RSU.
  • the RSU is equipped with an image sensor
  • the image sensor includes a plurality of surveillance cameras for collecting information on a target on a road; wherein the plurality of surveillance cameras include a first surveillance camera in a first surveillance direction and a second surveillance in a second surveillance direction.
  • a camera, the first monitoring camera is configured to collect back image data of a target in a specified road section, and the second monitoring camera is configured to collect forward image data of a target in a specified road section.
  • a road condition information processing method is applied to a roadside equipment RSU of a road condition information processing system.
  • the road condition information processing system includes a plurality of the RSUs deployed along a road travel direction.
  • the RSUs are equipped with image sensors.
  • the image sensor includes a plurality of surveillance cameras for collecting information on targets on the road; wherein the plurality of surveillance cameras include a first surveillance camera in a first surveillance direction and a second surveillance camera in a second surveillance direction.
  • the first monitoring camera is used to collect the back image data of the target in the designated road section, and the second monitoring camera is used to collect the forward image data of the target in the designated road section;
  • the method includes:
  • a road condition information processing device is applied to a roadside equipment RSU of a road condition information processing system.
  • the road condition information processing system includes a plurality of the RSUs deployed along a road travel direction.
  • the RSUs are equipped with image sensors.
  • the image sensor includes a plurality of surveillance cameras for collecting information on targets on the road; wherein the plurality of surveillance cameras include a first surveillance camera in a first surveillance direction and a second surveillance camera in a second surveillance direction.
  • the first monitoring camera is used to collect the back image data of the target in the designated road section, and the second monitoring camera is used to collect the forward image data of the target in the designated road section;
  • the device includes:
  • a collection result collection unit configured to collect an information collection result of a surveillance camera in the RSU
  • a target sensing unit is configured to sense a target in a designated road segment according to the information collection result.
  • FIG. 1 is a schematic diagram of a working area of a dome surveillance camera
  • FIG. 2 is a schematic diagram of a working area of a gun-type surveillance camera
  • FIG. 3 is a schematic diagram of the working mode of a gun-type surveillance camera
  • FIG. 4 is a schematic diagram of a surveillance camera installation manner provided by an embodiment of the present application.
  • FIG. 5 is a schematic diagram of a system provided by an embodiment of the present application.
  • FIG. 6 is a schematic diagram of another system according to an embodiment of the present application.
  • FIG. 8 is a schematic diagram of a device provided by an embodiment of the present application.
  • the RSU in order to reduce the dependence of the roadside equipment RSU on the information reported by traffic participants such as vehicles, the RSU has realized the ability of autonomous perception.
  • the RSU can be equipped with sensors, including lidar, millimeter-wave radar, and cameras.
  • the RSU can autonomously sense the target through the sensor and can be generated by the data processing module.
  • Data packet which can save the information of multiple targets (including location, speed, length, width, height, etc.), and send the data packet to the outside (including sending by broadcast, or it can also send by other methods To relevant recipients, etc.).
  • the RSU can sense its existence and broadcast it to other vehicles on the road, thereby improving the safety of autonomous driving.
  • a surveillance camera also known as a CCTV camera
  • a surveillance camera converts optical image signals into electrical signals for storage or transmission.
  • the light reflected on the object is collected by the camera lens to focus it on the light-receiving surface of the imaging device, and then the light is converted into electrical energy by the imaging device to obtain the "video signal".
  • the photoelectric signal is very weak. It needs to be amplified by a preamp circuit, and then processed and adjusted by various circuits.
  • the final standard signal can be recorded on a recording medium such as a video recorder, or transmitted through a transmission system or sent to a monitor. show.
  • the existing surveillance cameras are mainly divided into two categories, namely ball surveillance cameras and gun surveillance cameras.
  • the ball-type surveillance camera uses a short focal length ultra-wide-angle fixed lens, which can cover the entire cone-shaped area below the camera (the area shown in Figure 1) without dead angles in all directions.
  • the short focal length lens equipped by the camera will cause the camera to have a shorter shooting distance, poor resolution and low resolution for distant objects, unable to obtain enough effective information, and difficult to meet the requirements of the monitoring system. Therefore, dome surveillance cameras are usually used in small-scale surveillance scenarios, which is difficult to meet the needs of long-distance surveillance in road scenarios.
  • Gun-type surveillance cameras have the flexibility to choose lenses with different focal lengths, making them suitable for a variety of applications.
  • a telephoto lens is needed, but the field of view captured by the camera is small and the effective working range is limited (when the focal length exceeds 50mm, the horizontal viewing angle is less than 6 degrees).
  • the gun camera can only monitor scenes in a single direction (the area shown in Fig. 2) in the traffic scene, and cannot cover the area behind the camera, resulting in a blind spot in monitoring.
  • the ball-type surveillance camera and the gun-type surveillance camera each have their own advantages and disadvantages, and when used in a monitoring system in a traffic road scene, they also have their own disadvantages.
  • the scene to be monitored is a narrow rectangular area along the road direction
  • the surveillance area of the ball-type surveillance camera is a small circular area
  • the effective area covered is small and it is difficult to meet the road scene. Long-distance and large-area surveillance needs.
  • gun-type surveillance cameras are more suitable for use in narrow rectangular areas, the scope of action of gun-type surveillance cameras on the road plane is approximately a trapezoidal area. As shown in Figure 3, the camera is installed at one end of the rectangular area of the surveillance target, and fully covers the near-end area.
  • the range of the far-end scene will be correspondingly very large (the long bottom side of the ladder area), and the road area to be monitored
  • the proportion in the entire image is low. Therefore, the sharpness and low resolution of the target in the scene make it difficult to detect and track smaller objects, and it is difficult to meet the application requirements of detection.
  • the natural defect of a gun-type surveillance camera is that it can only cover one direction. Therefore, for a road (assuming a straight line), an obvious problem is that the area closer to the monitoring point, the worse the detection result (if it is in the Back of the current monitoring point).
  • an implementation scheme is provided that is more suitable for use in traffic road scenes and obtains higher definition and higher resolution images.
  • a gun-type surveillance camera can be used, and the surveillance cameras can be installed at both ends of the scene and are responsible for covering a part of the scene.
  • the scene can be combined into a complete surveillance Area (as shown in Figure 4).
  • each camera covers a part of the area in the scene and ensures sufficient sharpness and resolution in its area.
  • comprehensive and high-resolution coverage of the road area can be achieved.
  • a specific surveillance camera exists as one of the sensors of the RSU, it is usually installed on the RSU to monitor targets (traffic participants such as vehicles) on a specific monitored road section.
  • the surveillance cameras are respectively installed at both ends of the same scene.
  • the embodiment of the present application first provides a surveillance camera installation method on the RSU. Specifically, multiple surveillance cameras can be installed on each RSU.
  • the plurality of surveillance cameras include a first surveillance camera in a first surveillance direction and a second surveillance camera in a second surveillance direction.
  • the first surveillance camera is configured to collect back-facing image data of a target in a specified road section. Two surveillance cameras are used to collect the forward image data of the target in the specified road section.
  • two surveillance cameras can be installed on each RSU, one of which is forward monitoring (that is, monitoring facing the direction of road travel) and the other is backward monitoring (that is, facing the road) Travel in the opposite direction). Then, deploying multiple such RSUs along the road travel direction can realize the deployment of surveillance cameras at the two ends of the same road segment, and the surveillance directions are opposite.
  • the surveillance camera at the first end uses the first end as the near field area it monitors.
  • the surveillance camera on the second end uses the second end as the near-field area for surveillance.
  • the central part of the road section is the far-field area of the two surveillance cameras, and the two can overlap.
  • the information collection results of the two surveillance cameras can be combined to obtain more Comprehensive data of the forward and backward directions of the target can also realize the handover of the target in the cross-base station, which helps to maintain the continuous tracking of the target on the road.
  • the image of the area collected by the surveillance camera at the first end of the road section may not be clear, but the image of the area collected by the surveillance camera at the second end of the road section may be used as a supplement.
  • a road condition information processing system is first provided.
  • the system includes a plurality of roadside equipment RSUs deployed along a road traveling direction;
  • the RSU is equipped with an image sensor 501 and a data processing module 502;
  • the image sensor 501 includes a plurality of surveillance cameras for collecting information on targets on a road.
  • the plurality of surveillance cameras includes a first surveillance camera 5011 in a first surveillance direction and a first surveillance camera in a second surveillance direction.
  • Two monitoring cameras 5012, the first monitoring camera 5011 is used to collect the back image data of the target in the specified road section, and the second monitoring camera 5012 is used to collect the forward image data of the target in the specified road section;
  • the data processing module 502 is configured to sense a target in a designated road segment by collecting information collection results of a surveillance camera in the RSU.
  • each RSU can be installed with the opposite monitoring direction (the first monitoring direction and the second monitoring direction can have an included angle, and use the RSU's pillar as the axis, on each side of the pillar. )
  • At least two surveillance cameras to monitor the direction of travel and its reverse at the same location.
  • the two surveillance cameras can be installed back-to-back.
  • the entire road segment can be covered, and for every two RSUs
  • Each section of the road has a surveillance camera at each end of the road section to monitor the road section from different directions.
  • multiple surveillance cameras can be put into the same hardware module and then installed on the RSU.
  • the same RSU can only have the same data processing module for collecting data collected by the surveillance camera (of course, in actual applications, data collected by other types of sensors may also be collected and comprehensively calculated) , And then perceive the target in the corresponding road segment.
  • the data processing module may collect information in a variety of ways.
  • the data processing module may specifically be based on the first surveillance camera and the second surveillance camera equipped on the current RSU.
  • Information acquisition results, target sensing is performed on the section from the effective acquisition area of the first surveillance camera of the RSU to the effective acquisition area of the second surveillance camera.
  • the monitored sections are based on the RSU as the midpoint, and both surveillance cameras use the position of the RSU as the near end to monitor in the opposite direction. Therefore, the same RSU In the section monitored by RSU, both ends of the section are far-field areas.
  • the data processing module can combine the forward direction of the target in the specified road segment collected by the current RSU's second surveillance camera.
  • the image data and the back-facing image data of the target in the specified road segment collected by the first surveillance camera of the current RSU are used to perceive information on the target in the specified road segment. Therefore, it is possible to obtain clearer and clearer images than when a surveillance camera is installed Higher resolution data.
  • the RSU may be further equipped with a communication module; through the communication module, signal transmission between different RSUs may also be implemented, so as to implement other collection results collection methods.
  • the communication module of the RSU can use a specific wireless communication method.
  • the distance between different RSUs is usually close, and the range of signal transmission is greater than the distance.
  • RSUs can actually receive each other ’s signals.
  • a feasible approach is that each RSU uses its own sensor to realize the perception of the target, and the perception result includes multiple perceived targets and their detailed information (position, speed, length, width, height, etc.) , And then generate a packet of this information, and then send it to the outside via broadcast.
  • vehicles driving on the road can receive such broadcasts, can also make driving decisions based on them, and so on.
  • RSUs near the RSU can also listen to this data packet.
  • the RSU that listens to this data packet generates its own data packet, in addition to referring to the data collected by its own sensor, it can also communicate with this
  • the data in the data packets from other RSUs are fused to obtain more target data, or more detailed information on the same dimension in more dimensions, and so on.
  • the data packets are uniformly packaged to generate data packets and broadcast to the outside. In other words, in the foregoing manner, the RSU always broadcasts data packets containing specific target perception results.
  • the communication module in addition to broadcasting a data packet with a target perception result to the outside, can also send the information collection result of its first surveillance camera to the next RSU adjacent to it.
  • the designated road segment may specifically refer to the road segment from the previous RSU to the current RSU, correspondingly
  • the data processing module is specifically configured to combine back-facing image data in the road segment collected by the first surveillance camera in the previous RSU and the target in the specified road segment collected by the second surveillance camera in the current RSU. Forward image data to perceive the target on the specified road segment.
  • the first surveillance camera A1 on RSU A first captures the image of the target.
  • Data for example, mainly the image information of the vehicle facing away
  • the communication module of RSU A can send the image data collected by the first surveillance camera A1 to RSU B; in this way, when the target enters RSU B
  • the second monitoring camera B2 also collects image data of the target (mainly image information of the forward direction of the vehicle, etc.).
  • the data processing module of RSU B combines the data captured by the first surveillance camera A1 of RSU A and the data captured by the second surveillance camera B2 of RSU B to more accurately perceive the target in the road segment.
  • the communication module may be further configured to receive a notification message of a previous RSU.
  • the notification message includes a target moving in the direction of the RSU, and the appearance, speed, and number information of the target.
  • the data processing module may also obtain a priori information about the target according to the notification message of the previous RSU, and use the a priori information to compare the data from the previous RSU to the current RSU. Targets on the road are perceived. This can improve the accuracy of system target detection and tracking and reduce the computational complexity.
  • the target information in the scene can be transmitted throughout the system to achieve the full tracking and positioning of the target.
  • one first surveillance camera and one second surveillance camera may be deployed in the same RSU, that is, one surveillance camera in the forward direction and one surveillance camera in the backward direction may be sufficient.
  • more surveillance cameras can be deployed in the same RSU.
  • two first surveillance cameras and two second surveillance cameras can be installed on the same RSU.
  • the two first surveillance cameras monitor the direction of travel along the road, but the angle of view and other parameters of the two can be different.
  • the monitoring area of the same RSU in the same direction can be divided into multiple sections, which are monitored by different monitoring cameras, and so on.
  • This second embodiment separately protects the RSU installed with multiple different surveillance cameras mentioned in the first embodiment. Specifically, this second embodiment specifically provides a roadside device in a road condition information processing system.
  • the RSU is equipped with an image sensor
  • the image sensor includes a plurality of surveillance cameras for collecting information on a target on a road; wherein the plurality of surveillance cameras include a first surveillance camera in a first surveillance direction and a second surveillance in a second surveillance direction.
  • a camera, the first monitoring camera is configured to collect back image data of a target in a specified road section, and the second monitoring camera is configured to collect forward image data of a target in a specified road section.
  • the multiple surveillance cameras of the RSU there are multiple surveillance cameras with the same surveillance direction, each having a different surveillance perspective, and used to cover different areas of the corresponding road segment in the same surveillance direction. .
  • This third embodiment corresponds to the first embodiment.
  • a road condition information processing method is provided. The method is applied to a roadside device RSU of a road condition information processing system.
  • the road condition information processing system includes A plurality of the RSUs deployed along a road travel direction, the RSUs being equipped with an image sensor, the image sensor including a plurality of surveillance cameras for collecting information on targets on the road; wherein the plurality of surveillance cameras
  • the first monitoring camera includes a first monitoring camera in a first monitoring direction and a second monitoring camera in a second monitoring direction.
  • the first monitoring camera is configured to collect back-facing image data of a target in a specified road segment, and the second monitoring camera is configured to: Collect forward image data of the target in the specified road section;
  • the method may specifically include:
  • S701 Collect information collection results of the surveillance cameras in the RSU;
  • S702 Perceive a target in a designated road segment according to the information collection result.
  • the designated road section includes: The section between the effective acquisition area of the second surveillance camera of the current RSU and the effective acquisition area of the first surveillance camera; specifically, by combining the forward image data of the target in the specified road segment collected by the second surveillance camera of the current RSU, And the back-facing image data of the target in the specified road segment collected by the first surveillance camera of the current RSU, and performing information perception on the target in the specified road segment.
  • the collection results of the first surveillance camera of the adjacent previous RSU and the collection results of the second surveillance camera in the current RSU may also be collected; at this time, the designated road section includes: from the previous The section between the RSU and the current RSU can be obtained by combining the back-side image data collected by the first surveillance camera in the previous RSU and the positive direction of the target in the specified section collected by the second surveillance camera in the current RSU. To the image data, the target on the specified road segment is realized.
  • a notification message of the previous RSU can also be obtained.
  • the notification message includes the target moving in the direction of the RSU, and the result of the target's perception, which includes at least the appearance, speed, and number information of the target. ; Then, obtain prior information about the target according to the notification message of the previous RSU, so as to use the prior information to perceive the target on the road section from the previous RSU to the current RSU.
  • an embodiment of the present application further provides a road condition information processing device, which is applied to a roadside device RSU of a road condition information processing system, and the road condition information processing system includes a deployment along a road traveling direction.
  • the apparatus may specifically include:
  • a collection result collection unit 801 configured to collect an information collection result of a surveillance camera in the RSU;
  • the target sensing unit 802 is configured to sense a target in a designated road segment according to the information collection result.
  • the collection result collection unit may be specifically configured to:
  • the target sensing unit can be specifically used for:
  • target sensing is performed on the road segment from the effective acquisition area of the first surveillance camera of the RSU to the effective acquisition area of the second surveillance camera.
  • the collection result collection unit may be specifically configured to:
  • the target sensing unit can be specifically used for:
  • the target on the road segment from the previous RSU to the current RSU is realized.
  • the collection result collection unit may be further configured to obtain a notification message of the previous RSU, where the notification message includes the target moving in the direction of the RSU, and the appearance, speed, and number information of the target;
  • the device may further include:
  • a priori information obtaining unit is configured to obtain a priori information about the target according to the notification message of the previous RSU, so as to use the priori information to the road segment from the previous RSU to the current RSU The goal is perceived.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Atmospheric Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Traffic Control Systems (AREA)

Abstract

一种路况信息处理系统及其信息处理方法及装置,该系统包括沿着道路行进方向部署的多个路侧设备RSU,RSU配备有图像传感器(501)以及数据处理模块(502);图像传感器(501)包括多个监控相机,用于对道路上的目标进行信息采集;其中,多个监控相机中包括第一监控方向的第一监控相机(5011),以及第二监控方向的第二监控相机(5012),第一监控相机(5011)用于采集指定路段内目标的背向图像数据,第二监控相机(5012)用于采集指定路段内目标的正向图像数据;数据处理模块(502)用于通过收集RSU中的监控相机的信息采集结果,对指定路段内的目标进行感知。可以实现更精确的目标感知,提升系统安全性。

Description

路况信息处理系统及其信息处理方法及装置
本申请要求2018年09月14日递交的申请号为201811076738.0、发明名称为“路况信息处理系统及其信息处理方法及装置”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本申请涉及路况信息处理技术领域,特别是涉及路况信息处理系统及其信息处理方法及装置。
背景技术
在自动驾驶、智能交通等领域中,高度智能的自动驾驶车辆可对其四周环境进行有效感知,获取附近实体的类型、位置、方位、尺寸、速度等信息,并结合自身的驾驶状态,作出合理的驾驶决策完成车辆驾驶的工作,从而替代人工的驾驶劳动、实现高度自动化的驾驶过程。
目前,常见的自动驾驶感知系统有两种:车载传感器网络感知和车载通讯网络感知。在车载传感器网络感知的方式下,自动驾驶汽车需要装配众多的传感器(雷达、摄像头等)以获取足够丰富多样的传感信息,并配以强大的计算设备实时快速地处理数据获取完整的感知结果。但由于车自身视角较低所带来的局限性,处于多样的交通驾驶环境之中,其感知过程中存在很多不足且难以完成全面的感知功能。例如,当前方车辆体积过大遮挡了传感器有效感知区域时,自动驾驶车无法获得前方的准确信息,一旦有突发事件时(如行人、车辆等,从遮挡物身后出现),自动驾驶车辆难以及时处理、容易造成交通事故。同时,由于自动驾驶车辆传感器发射功率、分辨率、方向角等因素的限制,造成其收集信息的范围比较有限,无法在更大的空间中去感知交通驾驶的环境。在类似这样的情况下,自动驾驶车辆通常难以有效地保证驾驶的安全性。另外,这种方式对车辆自身传感器系统的依赖性很大,一旦出现故障,则会对自动驾驶的安全性产生非常大的影响。
在车载通讯网络感知的方式下,需要在自动驾驶车辆上装备高速无线通讯设备,目前可以采用V2X(Vehicle-to-everything,在车辆和任何会被该车辆所影响的实体之间分享信息的技术)来实现。简单来说,V2X是一种实现双向通讯和多向通讯的安全系统, 类似于WIFI的连接方式,可以让车车之间(V2V)、车人之间(V2P)、车与红绿灯等基础设施之间(V2I),发送信号,把类似于位置、速度、障碍、危险等信息发送给对方,从而提升行车安全。
其中,V2I是其中一项重要的组成部分,由于基础设施(例如,路侧设备RSU等)可以架设在比较高的位置,因此,可以通过更高的视角、更远的探测距离、无死角地获得更多的周围环境信息。现有技术中,是由车辆等交通参与者广播自身的位置、速度等信息,RSU或者其他交通参与者等通过接收车辆的广播消息实现对车辆的感知,同时,RSU可以将其接收到的交通参与者的信息以数据包等形式进行广播,车辆中的车载单元OBU通过接收RSU的数据包,来获知周围环境信息,进而做出行驶决策。但是,这种方式依赖于车辆的信息上报,一旦其中任一车辆不能上报自身位置,则意味着其不能被其他车辆所感知,进而成为安全隐患。
总之,如何进一步升级现有系统以提升其安全性能,成为需要本领域技术人员解决的技术问题。
发明内容
本申请提供了路况信息处理系统及其信息处理方法及装置,可以实现更精确的目标感知,提升系统安全性。
本申请提供了如下方案:
一种路况信息处理系统,
所述系统中包括沿着道路部署的多个路侧设备RSU;
所述RSU配备有图像传感器以及数据处理模块;
所述图像传感器包括多个监控相机,用于对道路上的目标进行信息采集;其中,所述多个监控相机中包括第一监控方向的第一监控相机,以及第二监控方向的第二监控相机,所述第一监控相机用于采集指定路段内目标的背向图像数据,所述第二监控相机用于采集指定路段内目标的正向图像数据;
所述数据处理模块,用于通过收集所述RSU中的监控相机的信息采集结果,对指定路段内的目标进行感知。
一种路况信息处理系统中的路侧设备,
所述RSU配备有图像传感器;
所述图像传感器包括多个监控相机,用于对道路上的目标进行信息采集;其中,所 述多个监控相机中包括第一监控方向的第一监控相机,以及第二监控方向的第二监控相机,所述第一监控相机用于采集指定路段内目标的背向图像数据,所述第二监控相机用于采集指定路段内目标的正向图像数据。
一种路况信息处理方法,应用于路况信息处理系统的路侧设备RSU中,所述路况信息处理系统中包括沿着道路行进方向部署的多个所述RSU,所述RSU配备有图像传感器,所述图像传感器包括多个监控相机,用于对道路上的目标进行信息采集;其中,所述多个监控相机中包括第一监控方向的第一监控相机,以及第二监控方向的第二监控相机,所述第一监控相机用于采集指定路段内目标的背向图像数据,所述第二监控相机用于采集指定路段内目标的正向图像数据;
所述方法包括:
收集所述RSU中的监控相机的信息采集结果;
根据所述信息采集结果对指定路段中的目标进行感知。
一种路况信息处理装置,应用于路况信息处理系统的路侧设备RSU中,所述路况信息处理系统中包括沿着道路行进方向部署的多个所述RSU,所述RSU配备有图像传感器,所述图像传感器包括多个监控相机,用于对道路上的目标进行信息采集;其中,所述多个监控相机中包括第一监控方向的第一监控相机,以及第二监控方向的第二监控相机,所述第一监控相机用于采集指定路段内目标的背向图像数据,所述第二监控相机用于采集指定路段内目标的正向图像数据;
所述装置包括:
采集结果收集单元,用于收集所述RSU中的监控相机的信息采集结果;
目标感知单元,用于根据所述信息采集结果对指定路段中的目标进行感知。
根据本申请提供的具体实施例,本申请公开了以下技术效果:
在本申请实施例中,通过为同一个RSU安装多个不同监控方向的监控相机,并将多个这种RSU部署到道路行进方向上,可以使得同一路段有多个不同的监控相机从不同的方向进行监控,从而可以通过组合同一路段两端的监控相机的有效监控区域,得到清晰度以及分辨率更高的图像,以实现更精确的目标感知,进而可以提升系统的安全性。
当然,实施本申请的任一产品并不一定需要同时达到以上所述的所有优点。
附图说明
为了更清楚地说明本申请实施例或现有技术中的技术方案,下面将对实施例中所需 要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本申请的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1是球式监控相机的工作区域示意图;
图2是枪式监控相机的工作区域示意图;
图3是枪式监控相机的工作方式示意图;
图4是本申请实施例提供的监控相机安装方式示意图;
图5是本申请实施例提供的系统的示意图;
图6是本申请实施例提供的另一系统的示意图;
图7是本申请实施例提供的方法的流程图;
图8是本申请实施例提供的装置的示意图。
具体实施方式
下面将结合本申请实施例中的附图,对本申请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申请一部分实施例,而不是全部的实施例。基于本申请中的实施例,本领域普通技术人员所获得的所有其他实施例,都属于本申请保护的范围。
在本申请实施例中,为了降低路侧设备RSU对车辆等交通参与者上报信息的依赖,为RSU实现了自主感知的能力。具体的,可以为RSU配备传感器,包括激光雷达、毫米波雷达、摄像头等,当车辆等目标进入到该RSU的感知范围时,RSU可以通过传感器自主感知到该目标,并且可以通过数据处理模块生成数据包,其中可以保存感知到的多个目标的信息(包括位置、速度、长宽高等等),并将数据包进行对外发送(包括通过广播的方式进行发送,或者,还可以通过其他方式发送给相关的接收方等等)。这样,即使道路上的某车辆不能上报其相关信息,RSU也能够感知到其存在,并将其广播给道路上的其他车辆,从而使得自动驾驶的安全性得以提升。
为了使得RSU具有自主感知能力,就需要在RSU中安装一些传感器,其中就包括监控相机。监控相机(也被称为CCTV相机)是把光学图象信号转变为电信号,以便于存储或者传输。当拍摄一个物体时,此物体上反射的光被摄像机镜头收集,使其聚焦在摄像器件的受光面上,再通过摄像器件把光转变为电能,即得到了“视频信号”。光电信号很微弱,需通过预放电路进行放大,再经过各种电路进行处理和调整,最后得到的 标准信号可以送到录像机等记录媒介上记录下来,或通过传播系统传播或送到监视器上显示出来。
现有的监控相机主要分为两大类,分别是球式监控相机以及枪式监控相机。其中,球式监控相机采用短焦距的超广角定镜头,可全方位无死角地覆盖整个相机下方的锥形区域(如图1所示的区域)。但相机装备的短焦距镜头会造成相机拍摄距离较短,对远处的物体解析度差且分辨率低,无法得到足够多的有效信息,难以满足监控系统的要求。因此,球式监控相机通常被用于小范围的监控场景,难以满足道路场景中长距离监控的需求。
枪式监控相机可灵活选择不同焦距的镜头,使其能够应用于多种应用场景。但由于光学镜头在监控范围和分辨率上难以两者兼顾。例如,为了加大监控距离需要采用长焦镜头,却会造成相机所拍摄的视野较小、有效工作范围有限(当焦距超过50mm时,水平视角小于6度)。另外,枪式相机在交通场景中只能监控单一方向的场景(如图2所示的区域),无法覆盖到相机的背后区域,造成了监控中的盲区。
可见,球式监控相机以及枪式监控相机各有各的优缺点,而在交通道路场景中的监控系统中使用时,也各自存在着不足。具体的,由于交通道路场景下,需要监控的场景是沿道路方向的狭长矩形区域,而球式监控相机的监控区域是小范围的圆形区域,覆盖的有效区域较小且难以满足道路场景中长距离大方位的监控需求。枪式监控相机虽然更适合在狭长矩形区域内使用,但是,枪式监控相机在道路平面上的作用区域近似为梯形区域。如图3所示,相机安装在监控目标矩形区域一端,且全面覆盖近端区域。而在需要覆盖近端所有区域(梯型区域的短底边一端)的情况下,会相应地造成远端场景的范围非常大(梯型区域的长底边一端),所需监控的道路区域在整个图像中的占比低。因此场景中目标的清晰度差、分辨率低,进而导致对较小的物体的检测和跟踪比较困难,难以达到实施检测的应用要求。另外,枪式监控相机的天然缺陷就是只能覆盖一个方向,因此,对于一条道路(假设是直线),一个明显的问题就是距离监控点越近的区域,反而检测的结果越差(如果是在当前监控点的背面)。
综上所述,在本申请实施例中,提供了更适合在交通道路场景中使用,并且获得更高清晰度、更高分辨率图像的实现方案。在该方案中,首先可以使用枪式监控相机,并且,可以将监控相机分别安装在场景的两端且负责覆盖其中场景中的一部分区域,通过组合两个相机覆盖的区域场景组合成完整的监控区域(如图4所示)。这样,由于每个相机覆盖场景中的一部分区域并确保其区域中具有足够的清晰度和分辨率。然后,通过 组合多个相机的有效区域,可以以达到对道路区域的全面且高清晰度地覆盖。
具体实现时,由于具体的监控相机是作为RSU的其中一种传感器存在的,因此,通常是安装在RSU上,用于对具体监控路段上的目标(车辆等交通参与者)进行监控。而为了达到上述目的,实现在同一个场景的两端分别安装监控相机,本申请实施例首先提供了RSU上的监控相机安装方式,具体的,可以在每个RSU上安装多个监控相机,所述多个监控相机中包括第一监控方向的第一监控相机,以及第二监控方向的第二监控相机,所述第一监控相机用于采集指定路段内目标的背向图像数据,所述第二监控相机用于采集指定路段内目标的正向图像数据。例如,一种典型的方式下,每个RSU上可以安装两个监控相机,其中一个为前向监控(也即,面向道路行进方向进行监控),另一个为后向监控(也即,面向道路行进相反的方向)。然后,沿着道路行进方向部署多个这种RSU,即可实现同一路段的两端分别部署监控相机,且监控方向相反,第一端的监控相机将该第一端作为其监控的近场区域,第二端的监控相机则将第二端作为监控的近场区域。路段的中央部分同为两个监控相机的远场区域,并且,两者可以存在交叠,这样,在对该路段的目标进行感知时,可以组合这两个监控相机的信息采集结果,获得更目标的正向和背向的全面数据,也能实现目标在跨基站中的交接,有助于保持目标在路面上的连续跟踪。例如,对于某路段的中央区域,该路段第一端的监控相机采集到的该区域的图像可能不够清晰,但是,可以利用该路段第二端的监控相机采集到的该区域的图像作为补充。
下面对具体的实现方式进行详细介绍。
实施例一
在该实施例一中,首先提供了一种路况信息处理系统,参见图5,所述系统中包括沿着道路行进方向部署的多个路侧设备RSU;
所述RSU配备有图像传感器501以及数据处理模块502;
所述图像传感器501包括多个监控相机,用于对道路上的目标进行信息采集;其中,所述多个监控相机中包括第一监控方向的第一监控相机5011,以及第二监控方向的第二监控相机5012,所述第一监控相机5011用于采集指定路段内目标的背向图像数据,所述第二监控相机5012用于采集指定路段内目标的正向图像数据;
所述数据处理模块502,用于通过收集所述RSU中的监控相机的信息采集结果,对指定路段内的目标进行感知。
也就是说,具体实现时,每个RSU上都可以安装监控方向相反(第一监控方向与第 二监控方向之间可以具有一夹角,并以RSU的立柱为轴,分别在立柱的两侧)的至少两个监控相机,实现在同一位置处,对行进方向及其反向的监控。例如,在一个RSU上安装两个监控相机的情况下,两个监控相机可以背靠背安装,这样,通过多个这种RSU的组合,可以实现对全路段的覆盖,并且对于每两个RSU之间的路段,在路段的两端的都分别具有一个监控相机,分别从不同的方向对该路段进行监控。其中,具体实现时,多个监控相机可以放到同一个硬件模块中,然后再安装到RSU上。并且,同一个RSU上可以仅具有同一个数据处理模块,用于收集监控相机采集到的数据(当然,在实际应用中,还可能会收集到其他类型的传感器采集到的数据,综合进行计算),然后,对对应路段中的目标进行感知。
其中,具体实现时,数据处理模块收集信息采集结果的方式可以有多种,例如,在一种方式下,数据处理模块具体可以根据当前所在RSU上配备的第一监控相机以及第二监控相机的信息采集结果,对从该RSU第一监控相机的有效采集区域到第二监控相机的有效采集区域之间的路段进行目标感知。也就是说,对于同一个RSU而言,其监控的路段,是以该RSU为中点,两个监控相机均以该RSU所在的位置为近端,向相反的方向进行监控,因此,同一个RSU监控的路段中,路段的两端均为远场区域。此时,由于RSU可以通过两个监控相机,使得路段中实际的近场区域得到延长,相应的,数据处理模块可以通过组合当前RSU的第二监控相机采集到的该指定路段内目标的正向图像数据,以及当前RSU的第一监控相机采集到的该指定路段内目标的背向图像数据,对该指定路段内的目标进行信息感知,因此,可以获得比安装一台监控相机时更清晰、更高分辨率的数据。
或者,在另一种实现方式下,所述RSU还可以配备有通讯模块;通过该通讯模块,还可以实现不同RSU之间的信号传输,以此实现其他的采集结果收集方式。
具体的,RSU的通讯模块可以具体无线通讯的方式,在RSU具有自主感知能力的情况下,由于不同RSU之间的距离通常会比较接近,而信号传输的范围又会大于该距离,因此,不同的RSU之间实际上是可以互相接收对方的信号的。利用该特点,一种可行的做法是,每个RSU利用自己的传感器实现对目标的感知,感知的结果中包括多个被感知到的目标及其详细信息(位置、速度、长宽高等等),然后将这些信息生成数据包,通过广播等形式对外发送。相应的,路上行驶的车辆等可以接收这种广播,还可以据此进行行驶决策,等等。同时,该RSU附近的其他RSU也可以收听到这种数据包,这样,收听到该数据包的RSU具体在生成自己的数据包时,除了可以参考自己的传感器采集到 的数据,还可以与这种来自其他RSU的数据包中的数据进行融合,获得更多目标的数据,或者同一在更多维度上的更详细的信息,等等,然后,再统一打包生成数据包,并对外广播。也就是说,在前述方式下,RSU对外广播的总是包含具体目标感知结果的数据包。
而在本申请实施例中,通讯模块除了可以向外广播带有具有目标感知结果的数据包,还可以向其相邻的下一个RSU发送其第一监控相机的信息采集结果。相应的,对于当前RSU而言,还可以接收相邻的上一个RSU的第一监控相机的采集结果;此时,指定路段具体可以是指,从上一RSU到当前RSU之间的路段,相应的,所述数据处理模块具体用于,通过组合上一RSU中第一监控相机采集到的该路段内的背向图像数据,以及当前RSU中第二监控相机采集到的该指定路段内目标的正向图像数据,对该指定路段上的目标的进行感知。
也就是说,如图6所示,假设RSU B是RSU A的下一个RSU,对于RSU A与RSU B之间的路段中的目标,RSU A上的第一监控相机A1首先采集到目标的图像数据(例如,主要是车辆背向的图像信息等);此时,RSU A的通讯模块可以将该第一监控相机A1采集到的图像数据发送给RSU B;这样,当目标进入到RSU B的第二监控相机B2的监控范围时,该第二监控相机B2也会采集到该目标的图像数据(主要是车辆正向的图像信息等)。然后,由RSU B的数据处理模块组合RSU A的第一监控相机A1拍摄到的数据,以及RSU B的第二监控相机B2拍摄到的数据,对该路段内的目标进行更精确的感知。
另外,在具体实现时,所述通讯模块还可以用于,接收上一RSU的通知消息,所述通知消息中包括有目标正在向该RSU的方向移动,以及该目标的外观、速度及编号信息;这样,所述数据处理模块还可以根据所述上一RSU的通知消息获得关于所述目标的先验信息,并利用所述先验信息,对所述从上一RSU到当前RSU之间的路段上的目标进行感知。由此可提高系统目标检测跟踪的准确率并降低计算的复杂度。同时,场景中的目标信息可在整个系统中传递,实现目标的全程跟踪和定位。
在实际应用中,同一RSU中可以分别部署一个第一监控相机以及一个第二监控相机,也即,前向及后向的监控相机各一个即可。或者,如果两个RSU之间的距离比较远,以至于两个RSU之间需要监控的路段比较长,还可以在同一个RSU中部署更多的监控相机。例如,同一RSU的所述多个监控相机中,监控方向相同的监控相机为多个,分别具有不同的监控视角,用于覆盖对应路段在相同监控方向上的不同区域。具体如,同一个RSU上可以安装两个第一监控相机,两个第二监控相机,其中,两个第一监控相机同 为沿着道路行进方向进行监控,但是两者的视角等参数可以不同,这样,可以将同一个RSU在同一方向上的监控区域划分为多段,分别由不同的监控相机进行监控,等等。
总之,在本申请实施例中,通过为同一个RSU安装多个不同监控方向的监控相机,并将多个这种RSU部署到道路行进方向上,可以使得同一路段有多个不同的监控相机从不同的方向进行监控,从而可以通过组合同一路段两端的监控相机的有效监控区域,得到清晰度以及分辨率更高的图像,进而可以实现更精确的目标感知。
实施例二
该实施例二将前述实施例一中提到的安装有多个不同监控相机的RSU单独进行保护,具体的,该实施例二具体提供了一种路况信息处理系统中的路侧设备,
所述RSU配备有图像传感器;
所述图像传感器包括多个监控相机,用于对道路上的目标进行信息采集;其中,所述多个监控相机中包括第一监控方向的第一监控相机,以及第二监控方向的第二监控相机,所述第一监控相机用于采集指定路段内目标的背向图像数据,所述第二监控相机用于采集指定路段内目标的正向图像数据。
在可选的实现方式下,所述RSU的所述多个监控相机中,监控方向相同的监控相机为多个,分别具有不同的监控视角,用于覆盖对应路段在相同监控方向上的不同区域。
实施例三
该实施例三是与实施例一相对应的,从RSU的角度,提供了一种路况信息处理方法,该方法应用于路况信息处理系统的路侧设备RSU中,所述路况信息处理系统中包括沿着道路行进方向部署的多个所述RSU,所述RSU配备有图像传感器,所述图像传感器包括多个监控相机,用于对道路上的目标进行信息采集;其中,所述多个监控相机中包括第一监控方向的第一监控相机,以及第二监控方向的第二监控相机,所述第一监控相机用于采集指定路段内目标的背向图像数据,所述第二监控相机用于采集指定路段内目标的正向图像数据;
具体的,参见图7,所述方法具体可以包括:
S701:收集所述RSU中的监控相机的信息采集结果;
S702:根据所述信息采集结果对指定路段中的目标进行感知。
其中,具体在收集所述RSU中的监控相机的信息采集结果时,可以收集当前所在RSU上配备的第一监控相机以及第二监控相机的信息采集结果;此时,所述指定路段包括:从当前RSU第二监控相机的有效采集区域到第一监控相机的有效采集区域之间的路 段;具体的,通过组合当前RSU的第二监控相机采集到的该指定路段内目标的正向图像数据,以及当前RSU的第一监控相机采集到的该指定路段内目标的背向图像数据,对该指定路段内的目标进行信息感知。
或者,在另一种实现方式下,还可以收集相邻的上一个RSU的第一监控相机的采集结果,以及当前RSU中第二监控相机的采集结果;此时,指定路段包括:从上一RSU到当前RSU之间的路段,可以通过组合上一RSU中第一监控相机采集到的该路段内的背向图像数据,以及当前RSU中第二监控相机采集到的该指定路段内目标的正向图像数据,实现对该指定路段上的目标的感知。
另外,还可以获得上一RSU的通知消息,所述通知消息中包括有目标正在向该RSU的方向移动,以及该目标的感知结果,所述感知结果包括至少包括目标的外观、速度及编号信息;然后,根据所述上一RSU的通知消息获得关于所述目标的先验信息,以便利用所述先验信息,对所述从上一RSU到当前RSU之间的路段上的目标进行感知。
关于上述实施例二以及实施例三中的未详述部分,可以参见前述实施例一中的记载,这里不再赘述。
与前述实施例三相对应,本申请实施例还提供了一种路况信息处理装置,应用于路况信息处理系统的路侧设备RSU中,所述路况信息处理系统中包括沿着道路行进方向部署的多个所述RSU,所述RSU配备有图像传感器,所述图像传感器包括多个监控相机,用于对道路上的目标进行信息采集;其中,所述多个监控相机中包括第一监控方向的第一监控相机,以及第二监控方向的第二监控相机,所述第一监控相机用于采集指定路段内目标的背向图像数据,所述第二监控相机用于采集指定路段内目标的正向图像数据;
参见图8,所述装置具体可以包括:
采集结果收集单元801,用于收集所述RSU中的监控相机的信息采集结果;
目标感知单元802,用于根据所述信息采集结果对指定路段中的目标进行感知。
具体的,所述采集结果收集单元具体可以用于:
收集当前所在RSU上配备的第一监控相机以及第二监控相机的信息采集结果;
相应的,目标感知单元具体可以用于:
根据收集到的信息采集结果,对从该RSU第一监控相机的有效采集区域到第二监控相机的有效采集区域之间的路段进行目标感知。
另一种方式下,所述采集结果收集单元具体可以用于:
收集相邻的上一个RSU的第一监控相机的采集结果,以及当前RSU中第二监控相 机的采集结果;
相应的,目标感知单元具体可以用于:
通过组合上一RSU中第一监控相机的有效采集区域,以及当前RSU中第二监控相机的有效采集区域,实现对从上一RSU到当前RSU之间的路段上的目标的感知。
另外,采集结果收集单元还可以用于,获得上一RSU的通知消息,所述通知消息中包括有目标正在向该RSU的方向移动,以及该目标的外观、速度及编号信息;
此时,该装置还可以包括:
先验信息获得单元,用于根据所述上一RSU的通知消息获得关于所述目标的先验信息,以便利用所述先验信息,对所述从上一RSU到当前RSU之间的路段上的目标进行感知。
通过以上的实施方式的描述可知,本领域的技术人员可以清楚地了解到本申请可借助软件加必需的通用硬件平台的方式来实现。基于这样的理解,本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来,该计算机软件产品可以存储在存储介质中,如ROM/RAM、磁碟、光盘等,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)执行本申请各个实施例或者实施例的某些部分所述的方法。
本说明书中的各个实施例均采用递进的方式描述,各个实施例之间相同相似的部分互相参见即可,每个实施例重点说明的都是与其他实施例的不同之处。尤其,对于系统或系统实施例而言,由于其基本相似于方法实施例,所以描述得比较简单,相关之处参见方法实施例的部分说明即可。以上所描述的系统及系统实施例仅仅是示意性的,其中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模块来实现本实施例方案的目的。本领域普通技术人员在不付出创造性劳动的情况下,即可以理解并实施。
以上对本申请所提供的路况信息处理系统及其信息处理方法及装置,进行了详细介绍,本文中应用了具体个例对本申请的原理及实施方式进行了阐述,以上实施例的说明只是用于帮助理解本申请的方法及其核心思想;同时,对于本领域的一般技术人员,依据本申请的思想,在具体实施方式及应用范围上均会有改变之处。综上所述,本说明书内容不应理解为对本申请的限制。

Claims (12)

  1. 一种路况信息处理系统,其特征在于,
    所述系统中包括沿着道路部署的多个路侧设备RSU;
    所述RSU配备有图像传感器以及数据处理模块;
    所述图像传感器包括多个监控相机,用于对道路上的目标进行信息采集;其中,所述多个监控相机中包括第一监控方向的第一监控相机,以及第二监控方向的第二监控相机,所述第一监控相机用于采集指定路段内目标的背向图像数据,所述第二监控相机用于采集指定路段内目标的正向图像数据;
    所述数据处理模块,用于通过收集所述RSU中的监控相机的信息采集结果,对指定路段内的目标进行感知。
  2. 根据权利要求1所述的系统,其特征在于,
    所述指定路段包括:从该RSU第二监控相机的有效采集区域到第一监控相机的有效采集区域之间的路段;
    所述数据处理模块具体用于,通过组合当前RSU的第二监控相机采集到的该指定路段内目标的正向图像数据,以及当前RSU的第一监控相机采集到的该指定路段内目标的背向图像数据,对该指定路段内的目标进行信息感知。
  3. 根据权利要求1所述的系统,其特征在于,
    所述RSU还配备有通讯模块;
    所述通讯模块用于,接收相邻的上一个RSU的第一监控相机的采集结果;
    所述指定路段包括:从上一RSU到当前RSU之间的路段;
    所述数据处理模块具体用于,通过组合上一RSU中第一监控相机采集到的该路段内的背向图像数据,以及当前RSU中第二监控相机采集到的该指定路段内目标的正向图像数据,对该指定路段上的目标的进行感知。
  4. 根据权利要求3所述的系统,其特征在于,
    所述通讯模块还用于,接收上一RSU的通知消息,所述通知消息中包括有目标正在向该RSU的方向移动,以及该目标的感知结果,所述感知结果至少包括:外观、速度及编号信息;
    所述数据处理模块还用于,根据所述上一RSU的通知消息获得关于所述目标的先验信息,并利用所述先验信息,对所述从上一RSU到当前RSU之间的路段上的目标进行感知。
  5. 根据权利要求1所述的系统,其特征在于,
    同一RSU的所述多个监控相机中,监控方向相同的监控相机为多个,分别具有不同的监控视角,用于覆盖对应路段在相同监控方向上的不同区域。
  6. 一种路况信息处理系统中的路侧设备,其特征在于,
    RSU配备有图像传感器;
    所述图像传感器包括多个监控相机,用于对道路上的目标进行信息采集;其中,所述多个监控相机中包括第一监控方向的第一监控相机,以及第二监控方向的第二监控相机,所述第一监控相机用于采集指定路段内目标的背向图像数据,所述第二监控相机用于采集指定路段内目标的正向图像数据。
  7. 根据权利要求6所述的路侧设备,其特征在于,
    所述RSU的所述多个监控相机中,监控方向相同的监控相机为多个,分别具有不同的监控视角,用于覆盖对应路段在相同监控方向上的不同区域。
  8. 一种路况信息处理方法,其特征在于,应用于路况信息处理系统的路侧设备RSU中,所述路况信息处理系统中包括沿着道路行进方向部署的多个所述RSU,所述RSU配备有图像传感器,所述图像传感器包括多个监控相机,用于对道路上的目标进行信息采集;其中,所述多个监控相机中包括第一监控方向的第一监控相机,以及第二监控方向的第二监控相机,所述第一监控相机用于采集指定路段内目标的背向图像数据,所述第二监控相机用于采集指定路段内目标的正向图像数据;
    所述方法包括:
    收集所述RSU中的监控相机的信息采集结果;
    根据所述信息采集结果对指定路段中的目标进行感知。
  9. 根据权利要求8所述的方法,其特征在于,
    所述收集所述RSU中的监控相机的信息采集结果,包括:
    收集当前所在RSU上配备的第一监控相机以及第二监控相机的信息采集结果;
    所述指定路段包括:从当前RSU第二监控相机的有效采集区域到第一监控相机的有效采集区域之间的路段;
    所述根据所述信息采集结果对指定路段中的目标进行感知,包括:
    通过组合当前RSU的第二监控相机采集到的该指定路段内目标的正向图像数据,以及当前RSU的第一监控相机采集到的该指定路段内目标的背向图像数据,对该指定路段内的目标进行信息感知。
  10. 根据权利要求8所述的方法,其特征在于,
    所述收集所述RSU中的监控相机的信息采集结果,包括:
    收集相邻的上一个RSU的第一监控相机的采集结果,以及当前RSU中第二监控相机的采集结果;
    所述指定路段包括:从上一RSU到当前RSU之间的路段;
    所述根据所述信息采集结果对指定路段中的目标进行感知,包括:
    通过组合上一RSU中第一监控相机采集到的该路段内的背向图像数据,以及当前RSU中第二监控相机采集到的该指定路段内目标的正向图像数据,实现对该指定路段上的目标的感知。
  11. 根据权利要求10所述的方法,其特征在于,还包括:
    获得上一RSU的通知消息,所述通知消息中包括有目标正在向该RSU的方向移动,以及该目标的感知结果,所述感知结果包括至少包括目标的外观、速度及编号信息;
    所述方法还包括:
    根据所述上一RSU的通知消息获得关于所述目标的先验信息,以便利用所述先验信息,对所述从上一RSU到当前RSU之间的路段上的目标进行感知。
  12. 一种路况信息处理装置,其特征在于,应用于路况信息处理系统的路侧设备RSU中,所述路况信息处理系统中包括沿着道路行进方向部署的多个所述RSU,所述RSU配备有图像传感器,所述图像传感器包括多个监控相机,用于对道路上的目标进行信息采集;其中,所述多个监控相机中包括第一监控方向的第一监控相机,以及第二监 控方向的第二监控相机,所述第一监控相机用于采集指定路段内目标的背向图像数据,所述第二监控相机用于采集指定路段内目标的正向图像数据;
    所述装置包括:
    采集结果收集单元,用于收集所述RSU中的监控相机的信息采集结果;
    目标感知单元,用于根据所述信息采集结果对指定路段中的目标进行感知。
PCT/CN2019/104124 2018-09-14 2019-09-03 路况信息处理系统及其信息处理方法及装置 WO2020052465A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811076738.0 2018-09-14
CN201811076738.0A CN110910632B (zh) 2018-09-14 2018-09-14 路况信息处理系统及其信息处理方法及装置

Publications (1)

Publication Number Publication Date
WO2020052465A1 true WO2020052465A1 (zh) 2020-03-19

Family

ID=69777396

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/104124 WO2020052465A1 (zh) 2018-09-14 2019-09-03 路况信息处理系统及其信息处理方法及装置

Country Status (2)

Country Link
CN (1) CN110910632B (zh)
WO (1) WO2020052465A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115457773A (zh) * 2022-09-19 2022-12-09 智道网联科技(北京)有限公司 路侧设备数据处理方法、装置及电子设备、存储介质
WO2024060575A1 (zh) * 2022-09-19 2024-03-28 智道网联科技(北京)有限公司 路侧设备数据处理方法、装置及电子设备、存储介质

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113077632A (zh) * 2021-06-07 2021-07-06 四川紫荆花开智能网联汽车科技有限公司 一种v2x智能网联路侧系统及实现方法
CN115880797A (zh) * 2021-09-28 2023-03-31 北京万集科技股份有限公司 一种基于v2x的车辆取证的方法、设备及系统
CN114973647A (zh) * 2022-04-11 2022-08-30 中智行(苏州)科技有限公司 一种路侧感知系统

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050027435A1 (en) * 2003-06-04 2005-02-03 Aps Technology Group, Inc. Systems and methods for monitoring and tracking movement and location of shipping containers and vehicles using a vision based system
CN203134149U (zh) * 2012-12-11 2013-08-14 武汉高德红外股份有限公司 基于不同波段成像融合图像处理的车辆辅助驾驶系统
CN105957401A (zh) * 2016-06-08 2016-09-21 上海汽车集团股份有限公司 基于车路协同的交叉路口行人防碰撞方法及其装置
CN205722415U (zh) * 2016-06-16 2016-11-23 北京万集科技股份有限公司 一种基于cpc复合通行卡的源头治超系统
CN106952476A (zh) * 2017-02-22 2017-07-14 成都华安视讯科技有限公司 可用于车辆违停动态监测的摄像机系统
CN107123303A (zh) * 2017-04-01 2017-09-01 智慧互通科技有限公司 雷达和智能相机联动管理路边停车位的系统及方法
WO2017176550A1 (en) * 2016-04-05 2017-10-12 Pcms Holdings, Inc. Method and system for autonomous vehicle sensor assisted selection of route with respect to dynamic route conditions
CN108417087A (zh) * 2018-02-27 2018-08-17 浙江吉利汽车研究院有限公司 一种车辆安全通行系统及方法

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA2513928A1 (en) * 2003-01-22 2004-08-05 Matsushita Electric Industrial Co., Ltd. A traffic information providing system, a traffic information representation method and apparatus therefor
CN108205892B (zh) * 2018-01-02 2020-07-21 江苏金晓电子信息股份有限公司 一种交通事件主动检测的方法及路侧智能信息终端

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050027435A1 (en) * 2003-06-04 2005-02-03 Aps Technology Group, Inc. Systems and methods for monitoring and tracking movement and location of shipping containers and vehicles using a vision based system
CN203134149U (zh) * 2012-12-11 2013-08-14 武汉高德红外股份有限公司 基于不同波段成像融合图像处理的车辆辅助驾驶系统
WO2017176550A1 (en) * 2016-04-05 2017-10-12 Pcms Holdings, Inc. Method and system for autonomous vehicle sensor assisted selection of route with respect to dynamic route conditions
CN105957401A (zh) * 2016-06-08 2016-09-21 上海汽车集团股份有限公司 基于车路协同的交叉路口行人防碰撞方法及其装置
CN205722415U (zh) * 2016-06-16 2016-11-23 北京万集科技股份有限公司 一种基于cpc复合通行卡的源头治超系统
CN106952476A (zh) * 2017-02-22 2017-07-14 成都华安视讯科技有限公司 可用于车辆违停动态监测的摄像机系统
CN107123303A (zh) * 2017-04-01 2017-09-01 智慧互通科技有限公司 雷达和智能相机联动管理路边停车位的系统及方法
CN108417087A (zh) * 2018-02-27 2018-08-17 浙江吉利汽车研究院有限公司 一种车辆安全通行系统及方法

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115457773A (zh) * 2022-09-19 2022-12-09 智道网联科技(北京)有限公司 路侧设备数据处理方法、装置及电子设备、存储介质
WO2024060575A1 (zh) * 2022-09-19 2024-03-28 智道网联科技(北京)有限公司 路侧设备数据处理方法、装置及电子设备、存储介质

Also Published As

Publication number Publication date
CN110910632B (zh) 2023-01-10
CN110910632A (zh) 2020-03-24

Similar Documents

Publication Publication Date Title
WO2020052465A1 (zh) 路况信息处理系统及其信息处理方法及装置
US20200074852A1 (en) Roadside sensing system based on vehicle infrastructure cooperation, and method for controlling vehicle thereof
CN103295396B (zh) 交通事故非现场快速取证方法及系统
KR101632168B1 (ko) 원샷 전방위탐지와 카메라영상으로 이루어진 하이브리드 분석형 스마트 감시카메라장치
US11027654B2 (en) Vehicle vision system with compressed video transfer via DSRC link
CN110719442A (zh) 一种安防监控系统
US8878927B2 (en) Method and apparatus for generating infrastructure-based basic safety message data
KR102045222B1 (ko) 데이터 융합을 통한 돌발상황 감시 제어 시스템
CN109102695B (zh) 智能交通服务站、智能交通服务方法及系统
CN204129798U (zh) 一种全景影像监控及行车记录系统
CN104077913A (zh) 融合多视角图像信息的交通事件监测方法及其装置
KR101477523B1 (ko) 차량간 통신을 이용한 협력 영상기록 장치 및 그 방법
CN111835998B (zh) 超视距全景图像获取方法、装置、介质、设备及系统
CN103886653A (zh) 一种全景影像监控及行车记录系统
CN110880235A (zh) 路况信息处理系统中的路侧设备、处理方法及装置
CN111311942A (zh) 基于v2x技术的路况显示方法、系统、v2x终端及v2x服务器
WO2019244425A1 (ja) 障害物検知システム及び障害物検知方法
US11782140B2 (en) SiPM based sensor for low level fusion
Chen et al. Low-latency high-level data sharing for connected and autonomous vehicular networks
WO2013111494A1 (ja) 監視システム
JPH10105863A (ja) 走行路情報収集装置
KR20200106810A (ko) 데이터 융합을 통한 돌발상황 감시 제어 시스템
CN103139539A (zh) 一种全景监视系统
CN113071498B (zh) 车辆控制方法、装置、系统、计算机设备及存储介质
KR101556581B1 (ko) 시야 확장용 실시간 무선 영상 공유 시스템

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: 19861136

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: 19861136

Country of ref document: EP

Kind code of ref document: A1