WO2022205104A1 - 一种车辆预警方法、系统、装置、设备及存储介质 - Google Patents

一种车辆预警方法、系统、装置、设备及存储介质 Download PDF

Info

Publication number
WO2022205104A1
WO2022205104A1 PCT/CN2021/084491 CN2021084491W WO2022205104A1 WO 2022205104 A1 WO2022205104 A1 WO 2022205104A1 CN 2021084491 W CN2021084491 W CN 2021084491W WO 2022205104 A1 WO2022205104 A1 WO 2022205104A1
Authority
WO
WIPO (PCT)
Prior art keywords
real
target
time
video
early warning
Prior art date
Application number
PCT/CN2021/084491
Other languages
English (en)
French (fr)
Inventor
金晨
卢红喜
李嫩
李国庆
李向阳
Original Assignee
浙江吉利控股集团有限公司
宁波吉利汽车研究开发有限公司
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 浙江吉利控股集团有限公司, 宁波吉利汽车研究开发有限公司 filed Critical 浙江吉利控股集团有限公司
Priority to CN202180071144.4A priority Critical patent/CN116368545A/zh
Priority to PCT/CN2021/084491 priority patent/WO2022205104A1/zh
Publication of WO2022205104A1 publication Critical patent/WO2022205104A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • 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
    • G08G1/16Anti-collision systems

Definitions

  • the present application relates to the technical field of remote control driving, and in particular to a vehicle early warning method, system, device, device and storage medium.
  • the remote control driving system has been developed and promoted by traditional car companies and Internet companies in recent years. It is mainly used in harsh environments and weather, port operations, logistics and transportation, and remote takeover of L3/L4 autonomous vehicles.
  • the remote control driving system on the market now monitors the surrounding environment of the vehicle through the camera on the vehicle end, and the vehicle terminal transmits the monitoring video to the remote driving remote control end and displays it on the monitor of the remote control end.
  • the driver of the remote control terminal controls the vehicle terminal by observing the surrounding environment of the vehicle terminal displayed on the display screen, and using the cockpit simulator of the remote control terminal to control the vehicle terminal. control command.
  • the present application provides a vehicle early warning method, system, device, device and storage medium.
  • the technical solution is as follows:
  • a vehicle warning method comprising:
  • the on-board terminal of the target vehicle obtains real-time monitoring videos of multiple on-board cameras on the target vehicle;
  • the on-board terminal sends the real-time monitoring video to the remote driving remote terminal of the target vehicle;
  • the remote driving remote control terminal performs early warning scene recognition on the target vehicle based on the real-time monitoring video, and obtains a recognition result, where the recognition result includes an early warning target;
  • the remote driving remote control terminal processes the real-time monitoring video to obtain the perception information of the early warning target;
  • the remote driving remote control terminal marks the early warning target in the real-time monitoring video based on the perception information to obtain a real-time early warning video;
  • the remote driving remote control terminal outputs the real-time warning video to a display device.
  • a vehicle warning method comprising:
  • the real-time warning video is output to a display device.
  • a vehicle early warning system includes: a vehicle-mounted terminal of a target vehicle and a remote driving remote control terminal of the target vehicle; wherein,
  • the vehicle-mounted terminal is used for acquiring real-time monitoring videos of multiple vehicle-mounted cameras on the target vehicle; sending the real-time monitoring videos to the remote driving remote control terminal;
  • the remote driving remote control terminal is configured to receive the real-time monitoring video sent by the vehicle-mounted terminal; perform early warning scene recognition on the target vehicle based on the real-time monitoring video, and obtain a recognition result, where the recognition result includes an early warning target; processing the real-time monitoring video to obtain the perception information of the early warning target; based on the perception information, marking the early warning target in the real-time monitoring video to obtain a real-time early warning video; converting the real-time early warning video output to a display device.
  • a vehicle warning device comprising:
  • a real-time monitoring video receiving module configured to receive the real-time monitoring videos of multiple on-board cameras on the target vehicle sent by the on-board terminal of the target vehicle;
  • an early warning scene recognition module configured to perform early warning scene recognition on the target vehicle based on the real-time monitoring video, and obtain a recognition result, where the recognition result includes an early warning target;
  • a perception information module configured to process the real-time monitoring video to obtain the perception information of the early warning target
  • an early warning target marking module configured to mark the early warning target in the real-time monitoring video based on the perception information to obtain a real-time early warning video
  • the real-time warning video output module is used for outputting the real-time warning video to a display device.
  • a vehicle early warning device in another aspect, includes a processor and a memory, the memory stores at least one instruction or at least a piece of program, and the at least one instruction or the at least one piece of program is processed by the The device is loaded and executed to realize the vehicle early warning method as described above.
  • a computer-readable storage medium stores at least one instruction or at least one piece of program, and the at least one instruction or at least one piece of program is loaded and executed by a processor to achieve the above vehicle warning method.
  • the vehicle early warning method, system, device, device and storage medium provided by this application have the following technical effects:
  • the technical solution provided by this application is by placing the monitoring, marking and early warning algorithms in the remote driving remote control terminal, and the remote driving remote control terminal receives the real-time monitoring video sent by the vehicle terminal, so that the dangerous objects in the upcoming dangerous scene are displayed in the real-time monitoring video in the real-time monitoring video.
  • Mark generate the corresponding early warning video, and output the early warning video to the display device to visually and clearly display the dangerous targets around the vehicle, helping the remote driver to discover the dangerous targets in time to avoid danger.
  • it realizes timely early warning and accurate Early warning protects the property and personal safety of drivers and passengers.
  • FIG. 1 is a schematic diagram of an application environment provided by an embodiment of the present application.
  • FIG. 2 is a schematic flowchart of a vehicle early warning method provided by an embodiment of the present application.
  • Fig. 3 is a kind of schematic flow chart of sending real-time monitoring video provided by the embodiment of the present application.
  • FIG. 4 is a schematic flowchart of a method for processing real-time monitoring video provided by an embodiment of the present application
  • FIG. 5 is a schematic flowchart of a method for coordinate positioning and marking of an early warning target provided by an embodiment of the present application
  • FIG. 6 is a schematic flowchart of a judging warning range provided by an embodiment of the present application.
  • FIG. 7 is a schematic flowchart of determining a target display device provided by an embodiment of the present application.
  • FIG. 8 is a schematic flow chart of outputting a real-time early warning video provided by an embodiment of the present application.
  • FIG. 9 is a schematic flowchart of another vehicle early warning method provided by an embodiment of the present application.
  • FIG. 10 is a schematic diagram of a vehicle early warning device provided by an embodiment of the present application.
  • FIG. 11 is a schematic diagram of a perception information module provided by an embodiment of the present application.
  • FIG. 1 is a schematic diagram of an application environment provided by an embodiment of the present application. As shown in FIG. 1 , the above application environment includes a vehicle terminal 01 and a remote driving remote control terminal 02 .
  • the in-vehicle terminal 01 may include a display device, an in-vehicle camera, a network communication unit, a processor, a memory, and the like. Specifically, the in-vehicle terminal 01 may acquire real-time monitoring videos of multiple in-vehicle cameras on the target vehicle; and send the above-mentioned real-time monitoring videos to the above-mentioned remote driving remote control terminal.
  • the remote control terminal 02 for remote driving may include a cockpit simulator, a display device, a network communication unit, a processor, a memory, and the like.
  • the remote driving remote control terminal 02 can receive the above-mentioned real-time monitoring video sent by the above-mentioned vehicle-mounted terminal; based on the above-mentioned real-time monitoring video, the above-mentioned target vehicle can be recognized by an early warning scene, and a recognition result is obtained, and the above-mentioned recognition result includes the early-warning target; the above-mentioned real-time monitoring video Perform processing to obtain the perception information of the above-mentioned early warning target; based on the above-mentioned perception information, mark the above-mentioned early-warning target in the above-mentioned real-time monitoring video to obtain a real-time early warning video; and output the above-mentioned real-time early warning video to a display device.
  • the remote driving remote control terminal can exchange information with the vehicle terminal through wireless communication.
  • the remote driving remote control terminal perceives the surrounding environment of the vehicle terminal through the real-time monitoring video transmitted by the vehicle terminal, and then sends the corresponding information to the vehicle terminal from the remote driving remote control terminal. Control commands to ensure the normal driving of the vehicle.
  • FIG. 2 is a schematic flowchart of a vehicle early warning method provided by an embodiment of the present application.
  • this specification provides method operation steps as described in the embodiments or flow charts, but more or less operation steps may be included based on routine or non-creative work.
  • the sequence of steps enumerated in the embodiments is only one of the execution sequences of many steps, and does not represent the only execution sequence.
  • the methods shown in the embodiments or the accompanying drawings may be executed sequentially or in parallel (for example, a parallel processor or a multi-threaded processing environment).
  • the above method may include:
  • the on-board terminal of the target vehicle acquires real-time monitoring videos of multiple on-board cameras on the target vehicle.
  • the above-mentioned target vehicle may be a vehicle based on remote control driving, and the above-mentioned target vehicle may include a vehicle-mounted terminal.
  • the remote-driving remote control terminal may interact with the vehicle-mounted terminal through wireless communication. Specifically, the remote driving remote control terminal sends vehicle control instructions to the vehicle-mounted terminal, and the vehicle-mounted terminal controls the vehicle-end actuator to execute the corresponding instructions in response to the vehicle control instructions.
  • the vehicle based on remote control driving is equipped with multiple on-board cameras in various directions at the vehicle end to monitor the surrounding environment situation of the vehicle end in real time.
  • the on-board terminal sends the real-time monitoring video to the remote driving remote control terminal of the target vehicle.
  • the driver of the remote driving remote control terminal observes the surrounding environment of the vehicle terminal based on the real-time monitoring video sent by the vehicle terminal, and controls the vehicle terminal through the cockpit simulator of the remote driving remote terminal.
  • the sending of the real-time monitoring video by the vehicle-mounted terminal to the remote driving remote terminal of the target vehicle may include:
  • the vehicle-mounted terminal encodes the real-time monitoring video to obtain an encoded real-time monitoring video.
  • the data volume of the video is reduced, the speed of video transmission is improved, and the real-time nature of the monitoring video received by the remote driving remote control terminal is ensured.
  • the vehicle-mounted terminal sends the encoded real-time monitoring video to the remote driving remote control terminal.
  • the receiving the encoded real-time monitoring video sent by the vehicle-mounted terminal may include: receiving the encoded real-time monitoring video sent by the vehicle-mounted terminal based on 5G communication.
  • video transmission based on 5G communication can further reduce the delay of video transmission, so that the remote control terminal of remote driving can receive the monitoring video of the vehicle terminal in real time.
  • the above-mentioned remote driving remote control terminal decodes the above-mentioned encoded real-time monitoring video to obtain the above-mentioned real-time monitoring video.
  • the encoded real-time monitoring video is decoded to obtain the original real-time monitoring video, so as to ensure accurate transmission of information in the video.
  • the remote driving remote control terminal performs an early warning scene recognition on the target vehicle based on the real-time monitoring video, and obtains a recognition result, where the recognition result includes an early warning target.
  • an early warning scene is identified for the situation of the surrounding environment of the vehicle. If there is an early warning target that affects the safety of the target vehicle in the environment where the current target vehicle is located, the environment where the current target vehicle is located is identified as an early warning scene.
  • the early warning scene may include a scene that threatens the safety of the target vehicle.
  • the early warning scene may include but is not limited to: a lane change warning scene, a reversing warning scene, and a door opening warning scene.
  • the real-time surveillance video is identified based on the preset early warning scene identification algorithm, and the identification result is obtained.
  • the preset early warning scene recognition algorithm may include multiple recognition algorithms for identifying different types of early warning scenes, and each recognition algorithm may be obtained by analyzing and summarizing a large number of sample video images corresponding to each early warning scene.
  • the above-mentioned early warning target may be a dynamic early warning target or a static early warning target
  • the dynamic early warning target may include a person or thing that is in motion in the environment where the target vehicle is located and affects the safe driving of the target vehicle.
  • Early warning targets may include, but are not limited to, moving vehicles and pedestrians; static early warning targets may include persons or objects that are stationary in the environment where the target vehicle is located and affect the safe driving of the target vehicle.
  • static early warning targets may include But not limited to stationary vehicles, roads, trees.
  • the remote driving remote control terminal processes the real-time monitoring video to obtain the perception information of the early warning target.
  • FIG. 4 is a schematic flowchart of a method for processing real-time surveillance video provided by an embodiment of this specification. Specifically, it may include:
  • the remote driving remote control terminal performs target detection on the real-time monitoring video, and determines the type information of the warning target.
  • determining the type information of the above-mentioned early warning target may include:
  • the above-mentioned characteristic information may include but not limited to: height information, width information, and pixel information;
  • type identification of the above-mentioned early warning target is performed, and type information of the early-warning target is determined.
  • the type information of the above-mentioned early-warning target can represent the classification category of the early-warning target.
  • the type information can include but not limited to: pedestrians, vehicles , roads, trees.
  • the remote driving remote control terminal performs target tracking on the real-time monitoring video, and determines multiple frames of images including the early warning target.
  • multiple frames of images including the above-mentioned early warning target are collected, and the above-mentioned multiple frames of images may include consecutive frame images.
  • the remote driving remote control terminal analyzes the motion trajectory of the warning target based on the multi-frame images, and determines real-time relative position information, real-time speed information, and real-time heading angle information of the warning target relative to the target vehicle.
  • the motion trajectory of the early warning target is extracted from the above-mentioned multi-frame images, and the above-mentioned motion trajectory is analyzed to obtain the real-time relative position information, real-time speed information and real-time heading angle information of the above-mentioned target vehicle.
  • the above-mentioned relative position information may include But not limited to distance information.
  • the remote driving remote control terminal uses the type information of the early warning target, the real-time relative position information, the real-time speed information, and the real-time heading angle information as the sensing information.
  • the perception information of the early warning target including type information, real-time relative position information, real-time speed information and real-time heading angle information can be obtained.
  • the remote driving remote control terminal based on the sensing information, marks the warning target in the real-time monitoring video to obtain a real-time warning video.
  • the above-mentioned method may further include:
  • the remote driving remote control terminal constructs a target coordinate system corresponding to the multi-frame images on the multi-frame images with the target vehicle as the coordinate origin respectively.
  • a corresponding target coordinate system is constructed on each frame of the above-mentioned multi-frame image, and the coordinate origin of the target coordinate system is the target vehicle. Coordinate System.
  • the above-mentioned remote driving remote control terminal marks the above-mentioned early warning target in the above-mentioned real-time monitoring video based on the above-mentioned perception information, and obtaining the real-time early warning video may include:
  • the remote driving remote control terminal determines, according to the real-time relative position information, the coordinate information of the warning target in the target coordinate system corresponding to the multi-frame images.
  • the coordinate information of the early warning target in the target coordinate system with the target vehicle as the coordinate origin is determined, so as to facilitate the positioning of the early warning target in the multi-frame images.
  • the remote driving remote control terminal marks the early warning target in the multi-frame images according to the type information of the early warning target and the coordinate information, and obtains the real-time early warning video.
  • marking the early warning target may include:
  • corresponding label style information is preset for each type of information.
  • the above label style information is used to distinguish early warning targets of different types of information.
  • the above label style information may include but not limited to: color, outline format , fill format.
  • the early warning target is marked in the multi-frame images, and the technical effect of distinguishing the early warning target from other environments in the multi-frame images is realized, so that the early warning can be displayed to the driver and passengers intuitively.
  • FIG. 6 is a schematic flowchart of judging an early warning range provided by an embodiment of this specification. Specifically, it may include:
  • the above-mentioned remote driving remote control terminal generates the predicted motion trajectory of the above-mentioned early warning target based on the above-mentioned real-time relative position information, the above-mentioned real-time speed information and the above-mentioned real-time heading angle information.
  • the predicted motion trajectory is generated through real-time relative position information, real-time velocity information and real-time heading angle information.
  • the above-mentioned motion trajectory generation algorithm may be based on multiple sample real-time position information, sample real-time velocity information, The real-time heading angle information of the sample and the corresponding movement trajectory of the sample are summarized.
  • the remote driving remote control terminal determines whether the warning target is within the warning range of the target vehicle based on the predicted motion trajectory.
  • the real-time driving information of the target vehicle can be obtained and the predicted motion trajectory of the target vehicle can be generated based on the real-time driving information.
  • the predicted motion trajectory of the early warning target and the predicted motion trajectory of the target vehicle have an intersection, it is determined that the early warning target is in the target vehicle. within the warning range.
  • the remote driving remote control terminal performs the above-mentioned step of marking the above-mentioned early warning target in the above-mentioned real-time monitoring video based on the above-mentioned perception information to obtain a real-time warning video.
  • the above warning target is marked in the real-time monitoring video to obtain a real-time early warning video.
  • the remote driving remote control terminal outputs the real-time warning video to a display device.
  • the above-mentioned identification result further includes the type information of the above-mentioned warning scene
  • the above-mentioned remote driving remote control terminal outputs the above-mentioned real-time warning video to the display device
  • the above-mentioned method may also include:
  • the remote driving remote control terminal determines the target display device of the warning video based on the type information of the warning scene.
  • remote control driving has a wide range of application scenarios.
  • the alighting terminal and the remote control terminal require joint early warning or separate early warning for dangerous scenarios.
  • the target warning personnel among the drivers and passengers corresponding to the warning video is determined, and the corresponding target display device is determined based on the target warning personnel.
  • the target warning person is a driver
  • the corresponding target display device may include a remote driving remote control terminal display device
  • the target warning person is a passenger
  • the corresponding target display device may include a vehicle-side display device.
  • Display device for remote driving remote control terminal when the type information of the warning scene is door opening warning, the target warning person is the passenger, and the target display device is the vehicle-side display device; when the type information of the warning scene is lane change warning, the target warning person is the driver, and the target display device is the driver.
  • the remote driving remote control terminal outputting the real-time warning video to the display device includes the remote driving remote terminal outputting the real-time warning video to the vehicle display device.
  • the above-mentioned remote driving remote control terminal outputting the above-mentioned real-time warning video to the above-mentioned vehicle-side display device may include:
  • the remote driving remote control terminal encodes the real-time warning video to obtain an encoded real-time warning video.
  • the real-time warning video is encoded to compress the video pixel data, reduce the data volume of the video, improve the speed of video transmission, and ensure the real-time nature of the warning video received by the vehicle terminal.
  • the remote driving remote control terminal sends the encoded real-time warning video to the vehicle terminal.
  • sending the encoded real-time warning video by the remote driving remote control terminal to the vehicle terminal may include: the remote driving remote terminal sending the encoded real-time warning video to the vehicle terminal based on 5G communication.
  • video transmission based on 5G communication can further reduce the delay in video transmission, so that the vehicle terminal can receive surveillance video in real time.
  • the vehicle-mounted terminal decodes the encoded real-time warning video to obtain the real-time warning video.
  • the encoded real-time early warning video is decoded to obtain the original real-time early warning video, so as to ensure the accurate transmission of information in the video.
  • the on-board terminal outputs the real-time warning video to the on-board display device.
  • the early warning video can be quickly transmitted from the remote driving remote control terminal to the vehicle terminal, and the vehicle terminal passengers can be warned in time.
  • the above-mentioned target display device includes a remote-driving remote-control terminal display device
  • the above-mentioned remote-driving remote-control terminal outputting the above-mentioned real-time warning video to the display device includes the above-mentioned remote-driving remote-control terminal outputting the above-mentioned real-time warning video to the above-mentioned remote driving remote control terminal display device .
  • the outputting the real-time warning video to the display device may include outputting the real-time warning video to the remote driving remote control display device and The above-mentioned real-time warning video is output to the above-mentioned vehicle-end display device.
  • the warning audio corresponding to the type information of the warning scene can also be preset.
  • the target display device outputs the real-time warning video
  • the corresponding warning audio is output to further remind drivers and passengers to avoid danger.
  • the remote driving remote control terminal receives the real-time monitoring video sent by the vehicle terminal, and marks the early warning target in the real-time monitoring video to obtain the real-time early warning video and displays it to the remote driver through the remote driving remote control terminal display device, It helps remote drivers to avoid risks.
  • the warning target will affect the passengers in the car
  • the real-time warning video is sent to the on-board terminal, and displayed to the passengers in the car through the on-board display device, which effectively improves the accuracy and pertinence of the warning.
  • FIG. 9 is a schematic flowchart of another vehicle early warning method provided by the embodiment of the present application.
  • the operation steps of the method described in the flow chart may include more or less operation steps based on routine or non-creative work.
  • the sequence of steps enumerated in the embodiments is only one of the execution sequences of many steps, and does not represent the only execution sequence.
  • an actual system or server product is executed, it can be executed sequentially or in parallel (for example, in a parallel processor or multi-threaded processing environment) according to the embodiments or the methods shown in the accompanying drawings.
  • the above method may include:
  • S901 Receive real-time monitoring videos of multiple on-board cameras on the target vehicle and sent by the on-board terminal of the target vehicle.
  • the real-time monitoring videos of multiple on-board cameras on the target vehicle sent by the on-board terminal receiving the target vehicle may include:
  • the receiving the encoded real-time monitoring video sent by the vehicle-mounted terminal may include: receiving the encoded real-time monitoring video sent by the vehicle-mounted terminal based on 5G communication.
  • S905 Process the real-time monitoring video to obtain the perception information of the early warning target.
  • the above-mentioned processing of the above-mentioned real-time monitoring video to obtain the perception information of the above-mentioned early warning target may include:
  • Target tracking is performed on the above-mentioned real-time monitoring video, and multiple frames of images including the above-mentioned early warning target are determined.
  • the movement trajectory analysis of the above-mentioned early warning target is carried out, and the real-time relative position information, real-time speed information and real-time heading angle information of the above-mentioned early warning target relative to the above-mentioned target vehicle are determined.
  • the type information of the early warning target, the real-time relative position information, the real-time speed information, and the real-time heading angle information are used as the sensing information.
  • the perception information of the early warning target including type information, real-time relative position information, real-time speed information and real-time heading angle information can be obtained.
  • the above-mentioned method may further include:
  • the target coordinate system corresponding to the multi-frame images is constructed with the target vehicle as the coordinate origin respectively on the multi-frame images.
  • marking the above-mentioned early warning target in the above-mentioned real-time monitoring video based on the above-mentioned perception information, and obtaining the real-time early warning video may include:
  • the warning target is marked in the above-mentioned multi-frame images, and the above-mentioned real-time warning video is obtained.
  • marking the early-warning target in the above-mentioned multi-frame images, and obtaining the above-mentioned real-time early-warning video may include:
  • the above-mentioned early warning target is marked in the above-mentioned real-time monitoring video, and before the real-time early warning video is obtained, the above-mentioned method may further include:
  • the above-mentioned method may further include:
  • the above-mentioned outputting the above-mentioned real-time warning video to the display equipment includes outputting the above-mentioned real-time warning video to the above-mentioned vehicle-side display equipment.
  • the above-mentioned outputting the above-mentioned real-time warning video to the above-mentioned on-board display device may include:
  • the sending the encoded real-time warning video to the vehicle-mounted terminal may include: sending the encoded real-time warning video to the vehicle-mounted terminal based on 5G communication.
  • the above-mentioned outputting the above-mentioned real-time warning video to the display device includes outputting the above-mentioned real-time warning video to the above-mentioned remote-driving remote terminal display device.
  • the above-mentioned outputting the above-mentioned real-time warning video to the display device may include outputting the above-mentioned real-time warning video to the above-mentioned remote driving remote control. terminal display device and output the above-mentioned real-time warning video to the above-mentioned vehicle-side display device.
  • the present application receives the real-time monitoring video sent by the vehicle-mounted terminal through the remote driving remote control terminal, marks the dangerous target in the upcoming dangerous scene in the real-time monitoring video, generates the corresponding early warning video, and sends the corresponding warning video.
  • the early warning video is output to the display device, which intuitively and clearly displays the dangerous targets around the vehicle, realizing timely and accurate early warning.
  • the embodiment of the present application also provides a vehicle early warning system to implement the method shown in FIG. 2 .
  • This system embodiment corresponds to the foregoing method embodiment, and can implement all the contents in the foregoing method embodiment.
  • this system embodiment only briefly describes the content in the foregoing method embodiment, and does not repeat the detailed content in the method embodiment one by one.
  • the above-mentioned system includes the vehicle-mounted terminal of the target vehicle and the remote driving remote control terminal of the above-mentioned target vehicle; wherein,
  • the above-mentioned vehicle-mounted terminal is used to obtain the real-time monitoring video of a plurality of vehicle-mounted cameras on the above-mentioned target vehicle; and the above-mentioned real-time monitoring video is sent to the above-mentioned remote driving remote control terminal;
  • the above-mentioned remote driving remote control terminal is used for receiving the above-mentioned real-time monitoring video sent by the above-mentioned vehicle-mounted terminal; based on the above-mentioned real-time monitoring video, the above-mentioned target vehicle is recognized by an early warning scene, and a recognition result is obtained, and the above-mentioned recognition result includes an early-warning target; processing to obtain the perception information of the above-mentioned early warning target; based on the above-mentioned perception information, mark the above-mentioned early-warning target in the above-mentioned real-time monitoring video to obtain a real-time early warning video; and output the above-mentioned real-time early warning video to a display device.
  • the above-mentioned vehicle-mounted terminal is further configured to encode the above-mentioned real-time monitoring video to obtain an encoded real-time monitoring video; and send the above-mentioned encoded real-time monitoring video to the above-mentioned remote driving remote control terminal;
  • the above-mentioned remote driving remote control terminal is further configured to receive the above-mentioned encoded real-time monitoring video sent by the above-mentioned vehicle-mounted terminal; and decode the above-mentioned encoded real-time monitoring video to obtain the above-mentioned real-time monitoring video.
  • the above-mentioned vehicle-mounted terminal is further configured to send the above-mentioned encoded real-time monitoring video to the above-mentioned remote driving remote control terminal based on 5G communication;
  • the above-mentioned remote driving remote control terminal is also used for receiving the above-mentioned encoded real-time monitoring video sent by the above-mentioned vehicle-mounted terminal based on 5G communication.
  • the above-mentioned remote driving remote control terminal is further configured to perform target detection on the above-mentioned real-time monitoring video, and determine the type information of the above-mentioned early warning target; carry out target tracking on the above-mentioned real-time monitoring video, and determine the multi-frame image including the above-mentioned early warning target.
  • the motion trajectory analysis of the above-mentioned early-warning target is carried out, and the real-time relative position information, real-time speed information and real-time heading angle information of the above-mentioned early-warning target relative to the above-mentioned target vehicle are determined;
  • the relative position information, the above-mentioned real-time speed information, and the above-mentioned real-time heading angle information are used as the above-mentioned perception information.
  • the above-mentioned remote driving remote control terminal is further configured to use the above-mentioned target vehicle as the coordinate origin on the above-mentioned multi-frame images, respectively, to construct a target coordinate system corresponding to the above-mentioned multi-frame images; Coordinate information in the target coordinate system corresponding to the above-mentioned multi-frame images; According to the type information of the above-mentioned early-warning target and the above-mentioned coordinate information, carry out early-warning target marking in the above-mentioned multi-frame images, and obtain the above-mentioned real-time warning video.
  • the above-mentioned remote driving remote control terminal is further configured to generate the predicted movement trajectory of the above-mentioned early warning target based on the above-mentioned real-time relative position information, the above-mentioned real-time speed information and the above-mentioned real-time heading angle information; based on the above-mentioned predicted movement trajectory, determine the above-mentioned early warning Whether the target is within the early warning range of the target vehicle; when it is determined to be yes, the above-mentioned step of marking the above-mentioned warning target in the above-mentioned real-time monitoring video based on the above-mentioned perception information to obtain a real-time warning video is performed.
  • the above-mentioned remote driving remote control terminal is further configured to determine the target display device of the above-mentioned warning video based on the above-mentioned type information of the above-mentioned warning scene;
  • the target display device is a vehicle-side display device
  • the above-mentioned real-time warning video is output to the above-mentioned vehicle-side display device;
  • the above-mentioned target display device is a remote driving remote control terminal display device
  • the above-mentioned real-time warning video is output to the above-mentioned remote driving remote control terminal for display. equipment.
  • the above-mentioned remote driving remote control terminal is further configured to encode the above-mentioned real-time warning video to obtain an encoded real-time warning video; and send the above-mentioned encoded real-time warning video to the above-mentioned vehicle-mounted terminal;
  • the above-mentioned vehicle-mounted terminal is further configured to receive the above-mentioned encoded real-time warning video sent by the above-mentioned remote driving remote control terminal.
  • the above-mentioned remote driving remote control terminal is further configured to send the above-mentioned encoded real-time warning video to the above-mentioned vehicle-mounted terminal based on 5G communication;
  • the above-mentioned vehicle terminal is further configured to receive the above-mentioned encoded real-time early warning video sent by the above-mentioned remote driving remote control terminal based on 5G communication.
  • the on-board terminal of the target vehicle acquires real-time monitoring videos of multiple on-board cameras on the target vehicle and sends the real-time monitoring videos to the remote driving remote control of the target vehicle.
  • the above-mentioned remote driving remote control terminal performs early-warning scene recognition on the above-mentioned target vehicle based on the above-mentioned real-time monitoring video, and obtains a recognition result, and the above-mentioned recognition result includes an early-warning target, and processes the above-mentioned real-time monitoring video to obtain the above-mentioned early warning target perception information, Then, based on the above-mentioned perception information, the above-mentioned early warning target is marked in the above-mentioned real-time monitoring video, and a real-time warning video is obtained.
  • the above-mentioned real-time warning video is output to a display device to visually and clearly display the dangerous targets around the vehicle and help the remote driver. Dangerous targets can be detected in time to avoid danger, which improves the timeliness and accuracy of early warning.
  • An embodiment of the present application provides a video early warning device, as shown in FIG. 10 , the above device includes:
  • a real-time monitoring video receiving module 1010 configured to receive the real-time monitoring videos of multiple on-board cameras on the target vehicle sent by the on-board terminal of the target vehicle;
  • an early warning scene recognition module 1020 configured to perform early warning scene recognition on the target vehicle based on the real-time monitoring video, and obtain a recognition result, where the recognition result includes an early warning target;
  • a perception information module 1030 configured to process the real-time monitoring video to obtain the perception information of the early warning target
  • an early warning target marking module 1040 configured to mark the early warning target in the real-time monitoring video based on the perception information to obtain a real-time early warning video
  • the real-time warning video output module 1050 is configured to output the real-time warning video to a display device.
  • the above-mentioned real-time monitoring video receiving module 1010 may include:
  • a first video receiving unit configured to receive the encoded real-time monitoring video sent by the vehicle-end central control, where the encoded real-time monitoring video is obtained after the vehicle-end central control encodes the real-time monitoring video;
  • the video decoding unit is used for decoding the above-mentioned encoded real-time monitoring video to obtain the above-mentioned real-time monitoring video.
  • the above-mentioned first video receiving unit may include:
  • the second video receiving unit is configured to receive the encoded real-time monitoring video sent by the vehicle-end central control based on 5G communication.
  • the above-mentioned perception information module 1030 may include:
  • a target detection unit 1031 configured to perform target detection on the above-mentioned real-time monitoring video, and determine the type information of the above-mentioned early warning target;
  • a target tracking unit 1032 configured to perform target tracking on the above-mentioned real-time monitoring video, and determine multiple frames of images including the above-mentioned early warning target;
  • the motion trajectory analysis unit 1033 performs motion trajectory analysis on the above-mentioned early warning target based on the above-mentioned multi-frame images, and determines the real-time relative position information, real-time speed information and real-time heading angle information of the above-mentioned early warning target relative to the above-mentioned target vehicle;
  • the sensing information unit 1034 is configured to use the type information of the early warning target, the real-time relative position information, the real-time speed information, and the real-time heading angle information as the sensing information.
  • the above-mentioned apparatus may further include:
  • a target coordinate system construction unit configured to construct a target coordinate system corresponding to the above-mentioned multi-frame images on the above-mentioned multi-frame images with the above-mentioned target vehicle as the coordinate origin respectively;
  • the above-mentioned early warning target marking module 1040 may include:
  • a coordinate information determining unit configured to determine, according to the real-time relative position information, the coordinate information of the warning target in the target coordinate system corresponding to the multi-frame images
  • the early warning target marking unit is configured to mark the early warning target in the above-mentioned multi-frame images according to the type information of the above-mentioned early-warning target and the above-mentioned coordinate information, and obtain the above-mentioned real-time warning video.
  • the above-mentioned device may further include:
  • a predicted motion trajectory generation unit configured to generate the predicted motion trajectory of the above-mentioned early warning target based on the above-mentioned real-time relative position information, the above-mentioned real-time speed information and the above-mentioned real-time heading angle information;
  • an early warning range judgment unit configured to judge whether the above-mentioned early-warning target is within the early-warning range of the above-mentioned target vehicle based on the above-mentioned predicted motion trajectory;
  • the step executing unit is configured to execute the above-mentioned step of marking the above-mentioned early warning target in the above-mentioned real-time monitoring video based on the above-mentioned perception information when the judgment is yes, to obtain a real-time early-warning video.
  • the above-mentioned apparatus may further include:
  • a target display device determination unit configured to determine the target display device of the above-mentioned early warning video based on the type information of the above-mentioned early warning scene
  • the above-mentioned target display device includes a vehicle-side display device
  • the above-mentioned real-time warning video output module 850 may include:
  • a first video output unit configured to output the above-mentioned real-time warning video to the above-mentioned on-board display device
  • the above-mentioned target display device includes a remote control terminal display device
  • the above-mentioned real-time warning video output module 850 may include:
  • the second video output unit is used for outputting the above-mentioned real-time warning video to the above-mentioned remote control terminal display device.
  • the above-mentioned first video output unit may further include:
  • a video encoding unit used for encoding the above-mentioned real-time warning video to obtain an encoded real-time warning video
  • the first video sending unit is configured to send the above-mentioned encoded real-time warning video to the above-mentioned vehicle-end central control, so that the above-mentioned encoded real-time warning video is decoded by the above-mentioned vehicle-end central control to obtain the above-mentioned real-time warning video, and the above-mentioned real-time warning video is obtained.
  • the real-time warning video is output to the above-mentioned on-board display device.
  • the above-mentioned first video sending unit may further include:
  • the second video sending unit is configured to send the encoded real-time warning video to the vehicle-end central control based on 5G communication.
  • An embodiment of the present application provides a vehicle early warning device, the vehicle early warning device includes a processor and a memory, and the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded by the processor and executed. It is executed to realize the vehicle early warning method provided by the above method embodiments.
  • the memory can be used to store software programs and modules, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory.
  • the memory may mainly include a storage program area and a storage data area, wherein the storage program area may store the operating system, application programs required for functions, etc.; the storage data area may store data created according to the use of the above-mentioned devices, etc.
  • the memory may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device. Accordingly, the memory may also include a memory controller to provide processor access to the memory.
  • the method embodiments provided in the embodiments of the present application may be executed in a mobile terminal, a computer terminal, or a similar computing device, that is, the above-mentioned computer equipment may include a mobile terminal, a computer terminal, or a similar computing device.
  • An embodiment of the present application further provides a storage medium, where the storage medium can be set in a server to store at least one instruction or at least a piece of program related to implementing one of the vehicle early warning methods in the method embodiments, the at least one instruction Or the at least one section of the program is loaded and executed by the processor to implement the vehicle warning method provided by the above method embodiments.
  • the above-mentioned storage medium may be located in at least one network server among multiple network servers of a computer network.
  • the above-mentioned storage medium may include but is not limited to: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic Various media that can store program codes, such as discs or optical discs.
  • the technical solution provided by the present application can intuitively and clearly display the dangerous objects around the vehicle, helping the remote driver to discover the dangerous objects in time In order to avoid danger, improve the timeliness and accuracy of early warning, and protect the property and personal safety of drivers and passengers; it can also further determine the target display device corresponding to the target early warning personnel among the drivers and passengers, and output real-time early warning video To the target display device, improve the pertinence and effectiveness of early warning.

Abstract

一种车辆预警方法、系统、装置、设备及存储介质,所述方法包括:接收目标车辆的车载终端发送的所述目标车辆上多个车载摄像头的实时监控视频;基于所述实时监控视频对所述目标车辆进行预警场景识别,得到识别结果,所述识别结果包括预警目标(S205);对所述实时监控视频进行处理,得到所述预警目标的感知信息(S207);基于所述感知信息,在所述实时监控视频中对所述预警目标进行标记,得到实时预警视频(S209);将所述实时预警视频输出至显示设备(S211)。利用上述方法,可以直观、清晰地显示车辆周围的危险目标,帮助远程驾驶员及时发现危险目标以规避危险,一方面,提升了预警的及时性和精准性,另一方面,保护了驾乘人员的财产及人身安全。

Description

一种车辆预警方法、系统、装置、设备及存储介质 技术领域
本申请涉及远程遥控驾驶技术领域,具体涉及一种车辆预警方法、系统、装置、设备及存储介质。
背景技术
远程遥控驾驶系统近年来被传统车企及互联网公司所研发及推广,主要适用在恶劣环境及天气、港口运营、物流运输、L3/L4级自动驾驶汽车远程接管等方向。现在市场上的远程遥控驾驶系统是通过车端的摄像头监控车辆周围环境,由车载终端将监控视频传输到远程驾驶遥控端并显示在遥控端的显示器上。远程驾驶遥控端驾驶员通过观察显示屏中展现的车端周围环境,使用遥控端的座舱模拟器对车端进行控制,具体的,遥控端发送控制指令到车载终端,再由车端执行器执行相应的控制指令。
然而,目前的远程遥控驾驶系统中没有相应的危险预警系统,当车端周围有危险场景发生时,例如车辆追尾/碰撞、行人/电动车突然出现、视频死角等,遥控端驾驶员只能通过遥控端的视频显示器画面判断场景的危险性再通过操控座舱模拟器对车端进行危险的规避,由于受制于显示器的帧率以及摄像头的角度局限,遥控端驾驶员对很多危险场景不能做出及时、准确地判断,导致远程驾驶的安全性无法得到保证,因此,需要提供更加有效的技术方案。
发明内容
为了解决现有技术的问题,本申请提供了一种车辆预警方法、系统、装置、设备及存储介质。所述技术方案如下:
一方面,提供了一种车辆预警方法,所述方法包括:
目标车辆的车载终端获取所述目标车辆上多个车载摄像头的实时监控视频;
所述车载终端向所述目标车辆的远程驾驶遥控端发送所述实时监控视频;
所述远程驾驶遥控端基于所述实时监控视频对所述目标车辆进行预警场景识别,得到识别结果,所述识别结果包括预警目标;
所述远程驾驶遥控端对所述实时监控视频进行处理,得到所述预警目标的感知信息;
所述远程驾驶遥控端基于所述感知信息,在所述实时监控视频中对所述预警目标进行标记,得到实时预警视频;
所述远程驾驶遥控端将所述实时预警视频输出至显示设备。
另一方面,提供了一种车辆预警方法,所述方法包括:
接收目标车辆的车载终端发送的所述目标车辆上多个车载摄像头的实时监控视频;
基于所述实时监控视频对所述目标车辆进行预警场景识别,得到识别结果,所述识别结果包括预警目标;
对所述实时监控视频进行处理,得到所述预警目标的感知信息;
基于所述感知信息,在所述实时监控视频中对所述预警目标进行标记,得到实时预警视频;
将所述实时预警视频输出至显示设备。
另一方面,提供了一种车辆预警系统,所述系统包括:目标车辆的车载终端和所述目标车辆的远程驾驶遥控端;其中,
所述车载终端,用于获取所述目标车辆上多个车载摄像头的实时监控视频;将所述实时监控视频发送到所述远程驾驶遥控端;
所述远程驾驶遥控端,用于接收所述车载终端发送的所述实时监控视频;基于所述实时监控视频对所述目标车辆进行预警场景识别,得到识别结果,所述识别结果包括预警目标;对所述实时监控视频进行处理,得到所述预警目标的感知信息;基于所述感知信息,在所述实时监控视频中对所述预警目标进行标记,得到实时预警视频;将所述实时预警视频输出至显示设备。
另一方面,提供了一种车辆预警装置,所述装置包括:
实时监控视频接收模块,用于接收目标车辆的车载终端发送的所述目标车辆上多个车载摄像头的实时监控视频;
预警场景识别模块,用于基于所述实时监控视频对所述目标车辆进行预警场景识别,得到识别结果,所述识别结果包括预警目标;
感知信息模块,用于对所述实时监控视频进行处理,得到所述预警目标的感知信息;
预警目标标记模块,用于基于所述感知信息,在所述实时监控视频中对所述预警目标进行标记,得到实时预警视频;
实时预警视频输出模块,用于将所述实时预警视频输出至显示设备。
另一方面,提供了一种车辆预警设备,所述设备包括处理器和存储器,所述存储器中存储有至少一条指令或至少一段程序,所述至少一条指令或所述至少一段程序由所述处理器加载并执行以实现如上述的车辆预警方法。
另一方面,提供了一种计算机可读存储介质,所述存储介质中存储有至少一条指令或至少一段程序,所述至少一条指令或所述至少一段程序由处理器加载并执行以实现如上述的车辆预警方法。
本申请提供的车辆预警方法、系统、装置、设备及存储介质,具有如下技术效果:
本申请提供的技术方案通过将监控、标记、预警算法置入远程驾驶遥控端中,远程驾驶遥控端接收车载终端发送的实时监控视频,对即将发生的危险场景中的危险目标在实时监控视频中进行标记,生成相应的预警视频,并将预警视频输出到显示设备,直观、清晰地显示车辆周围的危险目标,帮助远程驾驶员及时发现危险目标以规避危险,一方面,实现了及时预警和精确预警,另一方面,保护了驾乘人员的财产及人身安全。
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和优点,下面将对实施例或现有技术描述中所需要使用的附图作简单的介绍,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它附图。
图1是本申请实施例提供的一种应用环境的示意图;
图2是本申请实施例提供的一种车辆预警方法的流程示意图;
图3是本申请实施例提供的一种发送实时监控视频的流程示意图;
图4是本申请实施例提供的一种对实时监控视频的处理方法的流程示意图;
图5是本申请实施例提供的一种对预警目标的坐标定位及标记方法的流程示意图;
图6是本申请实施例提供的一种判断预警范围的流程示意图;
图7是本申请实施例提供的一种确定目标显示设备的流程示意图;
图8是本申请实施例提供的一种输出实时预警视频的流程示意图;
图9是本申请实施例提供的另一种车辆预警方法的流程示意图;
图10是本申请实施例提供的一种车辆预警装置示意图;
图11是本申请实施例提供的一种感知信息模块示意图;
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在没有做出创造性劳动的前提下所获得的所有其他实施例,都属于本申请保护的范围。
需要说明的是,本申请的说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别类似的对象,而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换,以便这里描述的本申请的实施例能够以除了在这里图示或描述的那些以外的顺序实施。此外,术语“包括” 和“具有”以及他们的任何变形,意图在于覆盖不排他的包含,例如,包含了一系列步骤或单元的过程、方法、系统、产品或服务器不必限于清楚地列出的那些步骤或单元,而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。
请参阅图1,图1是本申请实施例提供的一种应用环境的示意图,如图1所示,上述应用环境包括车载终端01、远程驾驶遥控端02。
具体的,车载终端01可以包括有显示设备、车载摄像头、网络通信单元、处理器和存储器等等。具体的,车载终端01可以获取目标车辆上多个车载摄像头的实时监控视频;将上述实时监控视频发送到上述远程驾驶遥控端。
具体的,远程驾驶遥控端02可以包括有座舱模拟器、显示设备、网络通信单元、处理器和存储器等等。具体的,远程驾驶遥控端02可以接收上述车载终端发送的上述实时监控视频;基于上述实时监控视频对上述目标车辆进行预警场景识别,得到识别结果,上述识别结果包括预警目标;对上述实时监控视频进行处理,得到上述预警目标的感知信息;基于上述感知信息,在上述实时监控视频中对上述预警目标进行标记,得到实时预警视频;将上述实时预警视频输出至显示设备。
在实际应用中,远程驾驶遥控端可以通过无线通信与车载终端进行信息交互,远程驾驶遥控端通过车载终端传输的实时监控视频感知车端周围环境,再由远程驾驶遥控端向车载终端发出相应的控制指令以保证车辆的正常行驶。
以下介绍本申请实施例提供的一种车辆预警方法,图2为本申请实施例提供的一种车辆预警方法的流程示意图。需要说明的是,本说明书提供了如实施例或流程图所述的方法操作步骤,但基于常规或者无创造性的劳动可以包括更多或者更少的操作步骤。实施例中列举的步骤顺序仅仅为众多步骤执行顺序中的一种方式,不代表唯一的执行顺序。在实际中的系统或产品执行时,可以按照实施例或者附图所示的方法顺序执行或者并行执行(例如并行处理器或者多线程处理的环境)。具体的如图2所示,上述方法可以包括:
S201,目标车辆的车载终端获取上述目标车辆上多个车载摄像头的实时监控视频。
在本说明书实施例中,上述目标车辆可以为基于远程遥控驾驶的车辆,上述目标车辆可以包括有车载终端,可选的,远程驾驶遥控端可以通过无线通信与车载终端进行交互。具体的,远程驾驶遥控端向车载终端发送车辆控制指令,车载终端响应于车辆控制指令,控制车端执行器执行相应指令。
在实际应用中,基于远程遥控驾驶的车辆在车端配备有各个方位的多个车载摄像头以对车端周围环境势态进行实时监控。
S203,上述车载终端向上述目标车辆的远程驾驶遥控端发送所述实时监控视频。
在实际应用中,远程驾驶遥控端的驾驶员基于车载终端发送的实时监控视频观察车端周围环境,并通过远程驾驶遥控端的座舱模拟器对车端进行控制。
在一个可选的实施例中,如图3所示,上述车载终端向上述目标车辆的远程驾驶遥控端发送所述实时监控视频可以包括:
S301,上述车载终端对上述实时监控视频进行编码,得到编码后的实时监控视频。
具体的,通过对实时监控视频进行编码以对视频像素数据进行压缩,减少视频的数据量,提升视频传输的速度,保证远程驾驶遥控端接收监控视频的实时性。
S303,上述车载终端向上述远程驾驶遥控端发送上述编码后的实时监控视频。
可选的,上述接收上述车载终端发送的编码后的实时监控视频可以包括:接收所述车载终端基于5G通信发送的编码后的实时监控视频。
在实际应用中,基于5G通信进行视频传输能够进一步减少视频传输时的时延,使得远程驾驶遥控端能够实时接收到车端的监控视频。
S305,上述远程驾驶遥控端对上述编码后的实时监控视频进行解码,得到 上述实时监控视频。
具体的,对编码后的实时监控视频进行解码,得到原始的实时监控视频,以保证视频中信息的准确传递。
S205,上述远程驾驶遥控端基于上述实时监控视频对上述目标车辆进行预警场景识别,得到识别结果,上述识别结果包括预警目标。
具体的,根据实时监控视频,对车端周围环境势态进行预警场景识别,若当前目标车辆所处环境存在影响目标车辆安全的预警目标,则将当前目标车辆所处环境识别为预警场景,具体的,预警场景可以包括对目标车辆的安全造成威胁的场景,具体的实施例中,预警场景可以包括但不限于:变道预警场景、倒车预警场景、开门预警场景。
在本说明书实施例中,基于预设的预警场景识别算法对实时监控视频进行识别,得到识别结果。具体的,上述预设的预警场景识别算法可以包括多种识别不同类型的预警场景的识别算法,每种识别算法可以基于每种预警场景对应的大量样本视频图像进行分析归纳后得到。
在实际应用中,上述预警目标可以为动态预警目标或静态预警目标,动态预警目标可以包括目标车辆所处环境中处于运动状态且影响目标车辆安全行驶的人或物,具体的实施例中,动态预警目标可以包括但不限于行驶中的车辆、行人;静态预警目标可以包括目标车辆所处环境中处于静止状态且影响目标车辆安全行驶的人或物,具体的实施例中,静态预警目标可以包括但不限于静止的车辆、道路、树木。
S207,上述远程驾驶遥控端对上述实时监控视频进行处理,得到上述预警目标的感知信息。
在一些实施例中,如图4所示,图4是本说明书实施例提供的一种对实时监控视频的处理方法的流程示意图,具体的,可以包括:
S401,上述远程驾驶遥控端对上述实时监控视频进行目标检测,确定上述预警目标的类型信息。
在一个具体的实施例中,确定上述预警目标的类型信息可以包括:
1)对上述实时监控视频进行目标检测,得到预警目标的特征信息,上述特征信息可以包括但不限于:高度信息、宽度信息、像素信息;
2)基于上述特征信息,对上述预警目标进行类型识别,确定预警目标的类型信息,上述预警目标的类型信息可以表征预警目标的分类类别,具体的,类型信息可以包括但不限于:行人、车辆、道路、树木。
S403,上述远程驾驶遥控端对上述实时监控视频进行目标追踪,确定包括上述预警目标的多帧图像。
具体的,从上述实时监控视频的图像中,采集包含上述预警目标的多帧图像,上述多帧图像可以包括连续帧图像。
S405,上述远程驾驶遥控端基于上述多帧图像对上述预警目标进行运动轨迹分析,确定上述预警目标的相对于上述目标车辆的实时相对位置信息、实时速度信息和实时航向角信息。
具体的,从上述多帧图像中提取出预警目标的运动轨迹,并对上述运动轨迹进行分析,得到上述目标车辆的实时相对位置信息、实时速度信息和实时航向角信息,上述相对位置信息可以包括但不限于距离信息。
S407,上述远程驾驶遥控端将上述预警目标的类型信息、上述实时相对位置信息、上述实时速度信息和上述实时航向角信息作为上述感知信息。
利用上述说明书实施例的技术方案,可以得到包括类型信息、实时相对位置信息、实时速度信息和实时航向角信息的预警目标的感知信息。
S209,上述远程驾驶遥控端基于上述感知信息,在上述实时监控视频中对上述预警目标进行标记,得到实时预警视频。
在一些实施例中,如图5所示,在上述远程驾驶遥控端对上述实时监控视频进行处理,得到上述预警目标的感知信息之后,上述方法还可以包括:
S409,上述远程驾驶遥控端在所述多帧图像上分别以所述目标车辆为坐标原点,构建所述多帧图像对应的目标坐标系。
具体的,在上述多帧图像的每帧图像上分别构建对应的目标坐标系,该目标坐标系的坐标原点为目标车辆,上述目标坐标系可以包括但不限于包括X轴和Y轴的二维坐标系。
相应的,上述远程驾驶遥控端基于上述感知信息,在上述实时监控视频中对上述预警目标进行标记,得到实时预警视频可以包括:
S411,上述远程驾驶遥控端根据上述实时相对位置信息,确定上述预警目标在上述多帧图像对应的目标坐标系中的坐标信息。
具体的,根据预警目标相对于目标车辆的实时相对位置信息,确定预警目标在以目标车辆为坐标原点的目标坐标系中的坐标信息,以便于对预警目标在多帧图像中进行定位。
S413,上述远程驾驶遥控端根据上述预警目标的类型信息和上述坐标信息,在上述多帧图像中进行预警目标标记,得到上述实时预警视频。
在一个具体的实施例中,对预警目标进行标记可以包括:
1)根据上述预警目标的类型信息,确定上述预警目标对应的标注样式信息。
在实际应用中,针对每种类型信息,预先设置对应的标注样式信息,上述标注样式信息用于区分不同类型信息的预警目标,具体的,上述标注样式信息可以包括但不限于:颜色、轮廓格式、填充格式。
2)根据上述标注样式信息和上述坐标信息,在上述多帧图像中进行预警目标标记,得到标记后的多帧图像。
具体的,根据标注样式信息和坐标信息,在多帧图像中对预警目标进行标记,实现将预警目标与多帧图像中的其他环境进行区分的技术效果,以便于直观地向驾乘人员展示预警目标的类型信息和相对于目标车辆的位置信息。
3)根据上述标记后的多帧图像,生成上述实时预警视频。
在一些实施例中,如图6所示,图6是本说明书实施例提供的一种判断预警范围的流程示意图,具体的,可以包括:
S601,上述远程驾驶遥控端基于上述实时相对位置信息、上述实时速度信 息和上述实时航向角信息生成上述预警目标的预测运动轨迹。
具体的,根据预设的运动轨迹生成算法,通过实时相对位置信息、实时速度信息和实时航向角信息生成预测运动轨迹,上述运动轨迹生成算法可以基于多个样本实时位置信息、样本实时速度信息、样本实时航向角信息和对应的样本运动轨迹进行归纳总结得到。
S603,上述远程驾驶遥控端基于上述预测运动轨迹,判断上述预警目标是否在上述目标车辆的预警范围内。
具体的,可以通过获取目标车辆的实时驾驶信息并基于实时驾驶信息生成目标车辆的预测运动轨迹,当预警目标的预测运动轨迹与目标车辆的预测运动轨迹出现交点时,则判断预警目标在目标车辆的预警范围内。
S605,当判断为是时,上述远程驾驶遥控端执行上述基于上述感知信息,在上述实时监控视频中对上述预警目标进行标记,得到实时预警视频的步骤。
在实际应用中,当预警目标在目标车辆的预警范围内时,基于感知信息,在实时监控视频中对上述预警目标进行标记,得到实时预警视频。
S211,上述远程驾驶遥控端将上述实时预警视频输出至显示设备。
在本说明书实施例中,如图7所示,当上述识别结果还包括上述预警场景的类型信息时,在上述远程驾驶遥控端将上述实时预警视频输出至显示设备之前,上述方法还可以包括:
S701,上述远程驾驶遥控端基于上述预警场景的类型信息,确定上述预警视频的目标显示设备。
在实际应用中,远程遥控驾驶的应用场景广泛,部分应用场景下车端和遥控端对危险场景需要共同预警或者分别预警。
在一个具体的实施例中,根据预警场景的类型信息和预警目标的感知信息,确定预警视频对应的驾乘人员中的目标预警人员,基于目标预警人员确定对应的目标显示设备,具体的,若目标预警人员为驾驶员,对应的目标显示设备可以包括远程驾驶遥控端显示设备;若目标预警人员为乘客,对应的目标显示设 备可以包括车端显示设备。例如,当预警场景的类型信息为开门预警时,目标预警人员为乘客,目标显示设备为车端显示设备;当预警场景的类型信息为变道预警时,目标预警人员为驾驶员,目标显示设备为远程驾驶遥控端显示设备。
S703,当上述目标显示设备包括车端显示设备时,上述远程驾驶遥控端将上述实时预警视频输出至显示设备包括上述远程驾驶遥控端将上述实时预警视频输出至上述车端显示设备。
在一个可选的实施例中,如图8所示,上述远程驾驶遥控端将上述实时预警视频输出至上述车端显示设备可以包括:
S801,上述远程驾驶遥控端对上述实时预警视频进行编码,得到编码后的实时预警视频。
在实际应用中,通过对实时预警视频进行编码,以对视频像素数据进行压缩,减少视频的数据量,提升视频传输的速度,保证车载终端接收预警视频的实时性。
S803,上述远程驾驶遥控端向上述车载终端发送上述编码后的实时预警视频。
可选的,上述远程驾驶遥控端向上述车载终端发送上述编码后的实时预警视频可以包括:上述远程驾驶遥控端基于5G通信向上述车载终端发送上述编码后的实时预警视频。
在实际应用中,基于5G通信进行视频传输能够进一步减少视频传输时的时延,使得车载终端能够实时接收到监控视频。
S805,上述车载终端将上述编码后的实时预警视频进行解码,得到上述实时预警视频。
在实际应用中,对编码后的实时预警视频进行解码,得到原始的实时预警视频,以保证视频中信息的准确传递。
S807,上述车载终端将上述实时预警视频输出至上述车端显示设备。
利用上述说明书实施例的技术方案,可以快速地将预警视频从远程驾驶遥 控端传输到车载终端,对车端乘客及时预警。
S705,当上述目标显示设备包括远程驾驶遥控端显示设备时,上述远程驾驶遥控端将上述实时预警视频输出至显示设备包括上述远程驾驶遥控端将上述实时预警视频输出至上述远程驾驶遥控端显示设备。
具体的,当上述目标显示设备包括上述远程驾驶遥控端显示设备和车端显示设备时,上述将上述实时预警视频输出至显示设备可以包括将上述实时预警视频输出至上述远程驾驶遥控端显示设备和将上述实时预警视频输出至上述车端显示设备。
在实际应用中,还可以预先设置与预警场景的类型信息对应的预警音频,在目标显示设备输出实时预警视频的同时,输出对应的预警音频,进一步提醒驾乘人员注意规避危险。
由以上本说明书实施例可见,远程驾驶遥控端接收车载终端发送的实时监控视频,以及将实时监控视频中的预警目标进行标记得到实时预警视频并通过远程驾驶遥控端显示设备向远程驾驶员显示,帮助远程驾驶员规避风险,当预警目标会影响到车内乘客时,将实时预警视频发送到车载终端,通过车端显示设备向车内乘客显示,有效地提升了预警的精准性和针对性。
以下以远程驾驶遥控端为执行主体介绍本说明书另一种车辆预警方法的具体实施例,图9是本申请实施例提供的另一种车辆预警方法的流程示意图,本说明书提供了如实施例或流程图所述的方法操作步骤,但基于常规或者无创造性的劳动可以包括更多或者更少的操作步骤。实施例中列举的步骤顺序仅仅为众多步骤执行顺序中的一种方式,不代表唯一的执行顺序。在实际中的系统或服务器产品执行时,可以按照实施例或者附图所示的方法顺序执行或者并行执行(例如并行处理器或者多线程处理的环境)。具体的如图9所示,上述方法可以包括:
S901,接收目标车辆的车载终端发送的上述目标车辆上多个车载摄像头的实时监控视频。
在一个可选的实施例中,上述接收目标车辆的车载终端发送的上述目标车辆上多个车载摄像头的实时监控视频可以包括:
1)接收上述车载终端发送的编码后的实时监控视频,上述编码后的实时监控视频由上述车载终端对上述实时监控视频进行编码后得到。
可选的,上述接收上述车载终端发送的编码后的实时监控视频可以包括:接收所述车载终端基于5G通信发送的编码后的实时监控视频。
2)对上述编码后的实时监控视频进行解码,得到上述实时监控视频。
S903,基于上述实时监控视频对上述目标车辆进行预警场景识别,得到识别结果,上述识别结果包括预警目标。
S905,对上述实时监控视频进行处理,得到上述预警目标的感知信息。
在一些实施例中,上述对上述实时监控视频进行处理,得到上述预警目标的感知信息可以包括:
1)对上述实时监控视频进行目标检测,确定上述预警目标的类型信息。
2)对上述实时监控视频进行目标追踪,确定包括上述预警目标的多帧图像。
3)基于上述多帧图像对上述预警目标进行运动轨迹分析,确定上述预警目标的相对于上述目标车辆的实时相对位置信息、实时速度信息和实时航向角信息。
4)将上述预警目标的类型信息、上述实时相对位置信息、上述实时速度信息和上述实时航向角信息作为上述感知信息。
利用上述说明书实施例的技术方案,可以得到包括类型信息、实时相对位置信息、实时速度信息和实时航向角信息的预警目标的感知信息。
S907,基于上述感知信息,在上述实时监控视频中对上述预警目标进行标记,得到实时预警视频。
在一些实施例中,在上述对上述实时监控视频进行处理,得到上述预警目标的感知信息之后,上述方法还可以包括:
在所述多帧图像上分别以所述目标车辆为坐标原点,构建所述多帧图像对 应的目标坐标系。
相应的,上述基于上述感知信息,在上述实时监控视频中对上述预警目标进行标记,得到实时预警视频可以包括:
1)根据上述实时相对位置信息,确定上述预警目标在上述多帧图像对应的目标坐标系中的坐标信息。
2)根据上述预警目标的类型信息和上述坐标信息,在上述多帧图像中进行预警目标标记,得到上述实时预警视频。
在一个具体的实施例中,上述根据上述预警目标的类型信息和上述坐标信息,在上述多帧图像中进行预警目标标记,得到上述实时预警视频可以包括:
1)根据上述预警目标的类型信息,确定上述预警目标对应的标注样式信息。
2)根据上述标注样式信息和上述坐标信息,在上述多帧图像中进行预警目标标记,得到标记后的多帧图像。
3)根据上述标记后的多帧图像,生成上述实时预警视频。
在一些实施例中,上述基于上述感知信息,在上述实时监控视频中对上述预警目标进行标记,得到实时预警视频之前,上述方法还可以包括:
1)基于上述实时相对位置信息、上述实时速度信息和上述实时航向角信息生成上述预警目标的预测运动轨迹。
2)基于上述预测运动轨迹,判断上述预警目标是否在上述目标车辆的预警范围内。
3)当判断为是时,执行上述基于上述感知信息,在上述实时监控视频中对上述预警目标进行标记,得到实时预警视频的步骤。
S909,将上述实时预警视频输出至显示设备。
在一个具体的实施例中,当上述识别结果还包括上述预警场景的类型信息时,在上述将上述实时预警视频输出至显示设备之前,上述方法还可以包括:
1)基于上述预警场景的类型信息,确定上述预警视频的目标显示设备。
2)当上述目标显示设备包括上述车端显示设备时,上述将上述实时预警视 频输出至显示设备包括将上述实时预警视频输出至上述车端显示设备。
在一个可选的实施例中,上述将上述实时预警视频输出至上述车端显示设备可以包括:
(1)对上述实时预警视频进行编码,得到编码后的实时预警视频。
(2)向上述车载终端发送上述编码后的实时预警视频,以使上述车载终端将上述编码后的实时预警视频进行解码,得到上述实时预警视频,以及将上述实时预警视频输出至上述车端显示设备。
可选的,上述向上述车载终端发送上述编码后的实时预警视频可以包括:基于5G通信向上述车载终端发送上述编码后的实时预警视频。
3)当上述目标显示设备包括上述远程驾驶遥控端显示设备时,上述将上述实时预警视频输出至显示设备包括将上述实时预警视频输出至上述远程驾驶遥控端显示设备。
在本说明书实施例中,当上述目标显示设备包括上述远程驾驶遥控端显示设备和车端显示设备时,上述将上述实时预警视频输出至显示设备可以包括将上述实时预警视频输出至上述远程驾驶遥控端显示设备和将上述实时预警视频输出至上述车端显示设备。
由以上本说明书实施例可见,本申请通过远程驾驶遥控端接收车载终端发送的实时监控视频,对即将发生的危险场景中的危险目标在实时监控视频中进行标记,生成相应的预警视频,并将预警视频输出到显示设备,直观、清晰地显示车辆周围的危险目标,实现了及时预警和精确预警。
上述从远程驾驶遥控端单侧撰写的车辆预警方法的实施例的具体细化步骤可以参见上述交互侧车辆预警方法的实施例,在此不再赘述。
本申请实施例还提供了一种车辆预警系统,用以实现图2所示的方法。本系统实施例与前述方法实施例对应,能够实现前述方法实施例中的全部内容。为便于阅读,本系统实施例仅对前述方法实施例中的内容进行概要性描述,不对方法实施例中的细节内容进行逐一赘述。上述系统包括目标车辆的车载终端 和上述目标车辆的远程驾驶遥控端;其中,
上述车载终端,用于获取上述目标车辆上多个车载摄像头的实时监控视频;将上述实时监控视频发送到上述远程驾驶遥控端;
上述远程驾驶遥控端,用于接收上述车载终端发送的上述实时监控视频;基于上述实时监控视频对上述目标车辆进行预警场景识别,得到识别结果,上述识别结果包括预警目标;对上述实时监控视频进行处理,得到上述预警目标的感知信息;基于上述感知信息,在上述实时监控视频中对上述预警目标进行标记,得到实时预警视频;将上述实时预警视频输出至显示设备。
在一个可选的实施例中,上述车载终端,还用于将上述实时监控视频进行编码,得到编码后的实时监控视频;向上述远程驾驶遥控端发送上述编码后的实时监控视频;
上述远程驾驶遥控端,还用于接收上述车载终端发送的上述编码后的实时监控视频;对上述编码后的实时监控视频进行解码,得到上述实时监控视频。
在一个可选的实施例中,上述车载终端,还用于基于5G通信向上述远程驾驶遥控端发送上述编码后的实时监控视频;
上述远程驾驶遥控端,还用于基于5G通信接收上述车载终端发送的上述编码后的实时监控视频。
在一些实施例中,上述远程驾驶遥控端,还用于对上述实时监控视频进行目标检测,确定上述预警目标的类型信息;对上述实时监控视频进行目标追踪,确定包括上述预警目标的多帧图像;基于上述多帧图像对上述预警目标进行运动轨迹分析,确定上述预警目标的相对于上述目标车辆的实时相对位置信息、实时速度信息和实时航向角信息;将上述预警目标的类型信息、上述实时相对位置信息、上述实时速度信息和上述实时航向角信息作为上述感知信息。
具体的,上述远程驾驶遥控端,还用于在上述多帧图像上分别以上述目标车辆为坐标原点,构建上述多帧图像对应的目标坐标系;根据上述实时相对位置信息,确定上述预警目标在上述多帧图像对应的目标坐标系中的坐标信息; 根据上述预警目标的类型信息和上述坐标信息,在上述多帧图像中进行预警目标标记,得到上述实时预警视频。
在一些实施例中,上述远程驾驶遥控端,还用于基于上述实时相对位置信息、上述实时速度信息和上述实时航向角信息生成上述预警目标的预测运动轨迹;基于上述预测运动轨迹,判断上述预警目标是否在上述目标车辆的预警范围内;当判断为是时,执行上述基于上述感知信息,在上述实时监控视频中对上述预警目标进行标记,得到实时预警视频的步骤。
在一个具体的实施例中,当上述识别结果还包括上述预警场景的类型信息时,上述远程驾驶遥控端,还用于基于上述预警场景的类型信息,确定上述预警视频的目标显示设备;当上述目标显示设备为车端显示设备时,将上述实时预警视频输出至上述车端显示设备;当上述目标显示设备为远程驾驶遥控端显示设备时,将上述实时预警视频输出至上述远程驾驶遥控端显示设备。
在一个可选的实施例中,上述远程驾驶遥控端,还用于对上述实时预警视频进行编码,得到编码后的实时预警视频;向上述车载终端发送上述编码后的实时预警视频;
上述车载终端,还用于接收上述远程驾驶遥控端发送的上述编码后的实时预警视频。
在一个可选的实施例中,上述远程驾驶遥控端,还用于基于5G通信向上述车载终端发送上述编码后的实时预警视频;
上述车载终端,还用于基于5G通信接收上述远程驾驶遥控端发送的上述编码后的实时预警视频。
从以上的描述中,可以看出,在本申请实施例中,目标车辆的车载终端获取上述目标车辆上多个车载摄像头的实时监控视频并将上述实时监控视频发送到上述目标车辆的远程驾驶遥控端;然后上述远程驾驶遥控端基于上述实时监控视频对上述目标车辆进行预警场景识别,得到识别结果,上述识别结果包括预警目标,并对上述实时监控视频进行处理,得到上述预警目标的感知信息, 接着基于上述感知信息,在上述实时监控视频中对上述预警目标进行标记,得到实时预警视频,最后将上述实时预警视频输出至显示设备,直观、清晰地显示车辆周围的危险目标,帮助远程驾驶员及时发现危险目标以规避危险,提升了预警的及时性和精准性。
本申请实施例提供了一种视频预警装置,如图10所示,上述装置包括:
实时监控视频接收模块1010,用于接收目标车辆的车载终端发送的所述目标车辆上多个车载摄像头的实时监控视频;
预警场景识别模块1020,用于基于所述实时监控视频对所述目标车辆进行预警场景识别,得到识别结果,所述识别结果包括预警目标;
感知信息模块1030,用于对所述实时监控视频进行处理,得到所述预警目标的感知信息;
预警目标标记模块1040,用于基于所述感知信息,在所述实时监控视频中对所述预警目标进行标记,得到实时预警视频;
实时预警视频输出模块1050,用于将所述实时预警视频输出至显示设备。
在一个可选的实施例中,上述实时监控视频接收模块1010可以包括:
第一视频接收单元,用于接收上述车端中控发送的编码后的实时监控视频,上述编码后的实时监控视频由上述车端中控对上述实时监控视频进行编码后得到;
视频解码单元,用于对上述编码后的实时监控视频进行解码,得到上述实时监控视频。
在一个可选的实施例中,上述第一视频接收单元可以包括:
第二视频接收单元,用于接收上述车端中控基于5G通信发送的编码后的实时监控视频。
在一些实施例中,如图11所示,上述感知信息模块1030可以包括:
目标检测单元1031,用于对上述实时监控视频进行目标检测,确定上述预警目标的类型信息;
目标追踪单元1032,用于对上述实时监控视频进行目标追踪,确定包括上述预警目标的多帧图像;
运动轨迹分析单元1033,基于上述多帧图像对上述预警目标进行运动轨迹分析,确定上述预警目标的相对于上述目标车辆的实时相对位置信息、实时速度信息和实时航向角信息;
感知信息单元1034,用于将上述预警目标的类型信息、上述实时相对位置信息、上述实时速度信息和上述实时航向角信息作为上述感知信息。
在一些实施例中,在上述感知信息模块1030之后,上述装置还可以包括:
目标坐标系构建单元,用于在上述多帧图像上分别以上述目标车辆为坐标原点,构建上述多帧图像对应的目标坐标系;
对应的,上述预警目标标记模块1040可以包括:
坐标信息确定单元,用于根据上述实时相对位置信息,确定上述预警目标在上述多帧图像对应的目标坐标系中的坐标信息;
预警目标标记单元,用于根据上述预警目标的类型信息和上述坐标信息,在上述多帧图像中进行预警目标标记,得到上述实时预警视频。
在本说明书实施例中,上述装置还可以包括:
预测运动轨迹生成单元,用于基于上述实时相对位置信息、上述实时速度信息和上述实时航向角信息生成上述预警目标的预测运动轨迹;
预警范围判断单元,用于基于上述预测运动轨迹,判断上述预警目标是否在上述目标车辆的预警范围内;
步骤执行单元,用于当判断为是时,执行上述基于上述感知信息,在上述实时监控视频中对上述预警目标进行标记,得到实时预警视频的步骤。
在一个具体的实施例中,当上述识别结果还包括上述预警场景的类型信息时,上述装置还可以包括:
目标显示设备确定单元,用于基于上述预警场景的类型信息,确定上述预警视频的目标显示设备;
当上述目标显示设备包括车端显示设备时,上述实时预警视频输出模块850可以包括:
第一视频输出单元,用于将上述实时预警视频输出至上述车端显示设备;
当上述目标显示设备包括远程遥控端显示设备时,上述实时预警视频输出模块850可以包括:
第二视频输出单元,用于将上述实时预警视频输出至上述远程遥控端显示设备。
在一个具体的实施例中,上述第一视频输出单元还可以包括:
视频编码单元,用于对上述实时预警视频进行编码,得到编码后的实时预警视频;
第一视频发送单元,用于向上述车端中控发送上述编码后的实时预警视频,以使上述车端中控将上述编码后的实时预警视频进行解码,得到上述实时预警视频,以及将上述实时预警视频输出至上述车端显示设备。
在一个可选的实施例中,上述第一视频发送单元还可以包括:
第二视频发送单元,用于基于5G通信向上述车端中控发送上述编码后的实时预警视频。
上述的装置实施例中的装置与方法实施例基于同样地发明构思。
本申请实施例提供了一种车辆预警设备,该车辆预警设备包括处理器和存储器,该存储器中存储有至少一条指令或至少一段程序,该至少一条指令或该至少一段程序由该处理器加载并执行以实现如上述方法实施例所提供的车辆预警方法。
存储器可用于存储软件程序以及模块,处理器通过运行存储在存储器的软件程序以及模块,从而执行各种功能应用以及数据处理。存储器可主要包括存储程序区和存储数据区,其中,存储程序区可存储操作系统、功能所需的应用程序等;存储数据区可存储根据上述设备的使用所创建的数据等。此外,存储器可以包括高速随机存取存储器,还可以包括非易失性存储器,例如至少一个 磁盘存储器件、闪存器件、或其他易失性固态存储器件。相应地,存储器还可以包括存储器控制器,以提供处理器对存储器的访问。
本申请实施例所提供的方法实施例可以在移动终端、计算机终端或者类似的运算装置中执行,即上述计算机设备可以包括移动终端、计算机终端或者类似的运算装置。
本申请实施例还提供了一种存储介质,上述存储介质可设置于服务器之中以保存用于实现方法实施例中一种的车辆预警方法相关的至少一条指令或至少一段程序,该至少一条指令或该至少一段程序由该处理器加载并执行以实现上述方法实施例提供的车辆预警方法。
可选地,在本实施例中,上述存储介质可以位于计算机网络的多个网络服务器中的至少一个网络服务器。可选地,在本实施例中,上述存储介质可以包括但不限于:U盘、只读存储器(ROM,Read-Only Memory)、随机存取存储器(RAM,Random Access Memory)、移动硬盘、磁碟或者光盘等各种可以存储程序代码的介质。
由上述本申请提供的车辆预警方法、系统、装置、设备或存储介质的实施例可见,利用本申请提供的技术方案可以直观、清晰地显示车辆周围的危险目标,帮助远程驾驶员及时发现危险目标以规避危险,提升了预警的及时性和精准性,同时保护了驾乘人员的财产及人身安全;还可以进一步确定与驾乘人员中的目标预警人员对应的目标显示设备,将实时预警视频输出至目标显示设备,提高预警的针对性和有效性。
需要说明的是:上述本申请实施例先后顺序仅仅为了描述,不代表实施例的优劣。且上述对本说明书特定实施例进行了描述。其它实施例在所附权利要求书的范围内。在一些情况下,在权利要求书中记载的动作或步骤可以按照不同于实施例中的顺序来执行并且仍然可以实现期望的结果。另外,在附图中描绘的过程不一定要求示出的特定顺序或者连续顺序才能实现期望的结果。在某些实施方式中,多任务处理和并行处理也是可以的或者可能是有利的。
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,对于装置、设备和存储介质实施例而言,由于其基本相似于方法实施例,所以描述的比较简单,相关之处参见方法实施例的部分说明即可。
本领域普通技术人员可以理解实现上述实施例的全部或部分步骤可以通过硬件来完成,也可以通过程序来指示相关的硬件完成,上述的程序可以存储于一种计算机可读存储介质中,上述提到的存储介质可以是只读存储器,磁盘或光盘等。
以上上述仅为本申请的较佳实施例,并不用以限制本申请,凡在本申请的精神和原则之内,所作的任何修改、等同替换、改进等,均应包含在本申请的保护范围之内。

Claims (30)

  1. 一种车辆预警方法,其特征在于,所述方法包括:
    目标车辆的车载终端获取所述目标车辆上多个车载摄像头的实时监控视频;
    所述车载终端向所述目标车辆的远程驾驶遥控端发送所述实时监控视频;
    所述远程驾驶遥控端基于所述实时监控视频对所述目标车辆进行预警场景识别,得到识别结果,所述识别结果包括预警目标;
    所述远程驾驶遥控端对所述实时监控视频进行处理,得到所述预警目标的感知信息;
    所述远程驾驶遥控端基于所述感知信息,在所述实时监控视频中对所述预警目标进行标记,得到实时预警视频;
    所述远程驾驶遥控端将所述实时预警视频输出至显示设备。
  2. 一种车辆预警方法,其特征在于,所述方法包括:
    接收目标车辆的车载终端发送的所述目标车辆上多个车载摄像头的实时监控视频;
    基于所述实时监控视频对所述目标车辆进行预警场景识别,得到识别结果,所述识别结果包括预警目标;
    对所述实时监控视频进行处理,得到所述预警目标的感知信息;
    基于所述感知信息,在所述实时监控视频中对所述预警目标进行标记,得到实时预警视频;
    将所述实时预警视频输出至显示设备。
  3. 根据权利要求2所述的方法,其特征在于,所述接收目标车辆的车载终端发送的所述目标车辆上多个车载摄像头的实时监控视频包括:
    接收所述车载终端发送的编码后的实时监控视频,所述编码后的实时监控视频由所述车载终端对所述实时监控视频进行编码后得到;
    对所述编码后的实时监控视频进行解码,得到所述实时监控视频。
  4. 根据权利要求3所述的方法,其特征在于,所述接收所述车载终端发送的编码后的实时监控视频包括:
    接收所述车载终端基于5G通信发送的编码后的实时监控视频。
  5. 根据权利要求2所述的方法,其特征在于,所述对所述实时监控视频进行处理,得到所述预警目标的感知信息包括:
    对所述实时监控视频进行目标检测,确定所述预警目标的类型信息;
    对所述实时监控视频进行目标追踪,确定包括所述预警目标的多帧图像;
    基于所述多帧图像对所述预警目标进行运动轨迹分析,确定所述预警目标的相对于所述目标车辆的实时相对位置信息、实时速度信息和实时航向角信息;
    将所述预警目标的类型信息、所述实时相对位置信息、所述实时速度信息和所述实时航向角信息作为所述感知信息。
  6. 根据权利要求5所述的方法,其特征在于,在所述对所述实时监控视频进行处理,得到所述预警目标的感知信息之后,所述方法还包括:
    在所述多帧图像上分别以所述目标车辆为坐标原点,构建所述多帧图像对应的目标坐标系;
    所述基于所述感知信息,在所述实时监控视频中对所述预警目标进行标记,得到实时预警视频包括:
    根据所述实时相对位置信息,确定所述预警目标在所述多帧图像对应的目标坐标系中的坐标信息;
    根据所述预警目标的类型信息和所述坐标信息,在所述多帧图像中进行预警目标标记,得到所述实时预警视频。
  7. 根据权利要求5所述的方法,其特征在于,在所述基于所述感知信息,在所述实时监控视频中对所述预警目标进行标记,得到实时预警视频之前,所述方法还包括:
    基于所述实时相对位置信息、所述实时速度信息和所述实时航向角信息生成所述预警目标的预测运动轨迹;
    基于所述预测运动轨迹,判断所述预警目标是否在所述目标车辆的预警范围内;
    当判断为是时,执行所述基于所述感知信息,在所述实时监控视频中对所述预警目标进行标记,得到实时预警视频的步骤。
  8. 根据权利要求2所述的方法,其特征在于,当所述识别结果还包括所述预警场景的类型信息时,在所述将所述实时预警视频输出至显示设备之前,所述方法还包括:
    基于所述预警场景的类型信息,确定所述预警视频的目标显示设备;
    当所述目标显示设备包括车端显示设备时,所述将所述实时预警视频输出至显示设备包括将所述实时预警视频输出至所述车端显示设备;
    当所述目标显示设备包括远程驾驶遥控端显示设备时,所述将所述实时预警视频输出至显示设备包括将所述实时预警视频输出至所述远程驾驶遥控端显示设备。
  9. 根据权利要求8所述的方法,其特征在于,所述将所述实时预警视频输出至所述车端显示设备包括:
    对所述实时预警视频进行编码,得到编码后的实时预警视频;
    向所述车载终端发送所述编码后的实时预警视频,以使所述车载终端将所述编码后的实时预警视频进行解码,得到所述实时预警视频,以及将所述实时预警视频输出至所述车端显示设备。
  10. 根据权利要求9所述的方法,其特征在于,所述向所述车载终端发送所述编码后的实时预警视频包括:
    基于5G通信向所述车载终端发送所述编码后的实时预警视频。
  11. 一种车辆预警系统,其特征在于,所述系统包括目标车辆的车载终端和所述目标车辆的远程驾驶遥控端;其中,
    所述车载终端,用于获取所述目标车辆上多个车载摄像头的实时监控视频;将所述实时监控视频发送到所述远程驾驶遥控端;
    所述远程驾驶遥控端,用于接收所述车载终端发送的所述实时监控视频;基于所述实时监控视频对所述目标车辆进行预警场景识别,得到识别结果,所述识别结果包括预警目标;对所述实时监控视频进行处理,得到所述预警目标的感知信息;基于所述感知信息,在所述实时监控视频中对所述预警目标进行标记,得到实时预警视频;将所述实时预警视频输出至显示设备。
  12. 根据权利要求11所述的系统,其特征在于,所述车载终端,还用于将所述实时监控视频进行编码,得到编码后的实时监控视频;向所述远程驾驶遥控端发送所述编码后的实时监控视频;
    所述远程驾驶遥控端,还用于接收所述车载终端发送的所述编码后的实时监控视频;对所述编码后的实时监控视频进行解码,得到所述实时监控视频。
  13. 根据权利要求12所述的系统,其特征在于,所述车载终端,还用于基于5G通信向所述远程驾驶遥控端发送所述编码后的实时监控视频;
    所述远程驾驶遥控端,还用于基于5G通信接收所述车载终端发送的所述编码后的实时监控视频。
  14. 根据权利要求11所述的系统,其特征在于,所述远程驾驶遥控端,还用 于对所述实时监控视频进行目标检测,确定所述预警目标的类型信息;对所述实时监控视频进行目标追踪,确定包括所述预警目标的多帧图像;基于所述多帧图像对所述预警目标进行运动轨迹分析,确定所述预警目标的相对于所述目标车辆的实时相对位置信息、实时速度信息和实时航向角信息;将所述预警目标的类型信息、所述实时相对位置信息、所述实时速度信息和所述实时航向角信息作为所述感知信息。
  15. 根据权利要求14所述的系统,其特征在于,所述远程驾驶遥控端,还用于在所述多帧图像上分别以所述目标车辆为坐标原点,构建所述多帧图像对应的目标坐标系;根据所述实时相对位置信息,确定所述预警目标在所述多帧图像对应的目标坐标系中的坐标信息;根据所述预警目标的类型信息和所述坐标信息,在所述多帧图像中进行预警目标标记,得到所述实时预警视频。
  16. 根据权利要求14所述的系统,其特征在于,所述远程驾驶遥控端,还用于基于所述实时相对位置信息、所述实时速度信息和所述实时航向角信息生成所述预警目标的预测运动轨迹;基于所述预测运动轨迹,判断所述预警目标是否在所述目标车辆的预警范围内;当判断为是时,执行所述基于所述感知信息,在所述实时监控视频中对所述预警目标进行标记,得到实时预警视频的步骤。
  17. 根据权利要求11所述的系统,其特征在于,当所述识别结果还包括所述预警场景的类型信息时,所述远程驾驶遥控端,还用于基于所述预警场景的类型信息,确定所述预警视频的目标显示设备;当所述目标显示设备包括车端显示设备时,将所述实时预警视频输出至所述车端显示设备;当所述目标显示设备包括远程驾驶遥控端显示设备时,将所述实时预警视频输出至所述远程驾驶遥控端显示设备。
  18. 根据权利要求17所述的系统,其特征在于,所述远程驾驶遥控端,还用于对所述实时预警视频进行编码,得到编码后的实时预警视频;向所述车载终端发送所述编码后的实时预警视频;
    所述车载终端,还用于接收所述远程驾驶遥控端发送的所述编码后的实时预警视频。
  19. 根据权利要求18所述的系统,其特征在于,所述远程驾驶遥控端,还用于基于5G通信向所述车载终端发送所述编码后的实时预警视频;
    所述车载终端,还用于基于5G通信接收所述远程驾驶遥控端发送的所述编码后的实时预警视频。
  20. 一种车辆预警装置,特征在于,所述装置包括:
    实时监控视频接收模块,用于接收目标车辆的车载终端发送的所述目标车辆上多个车载摄像头的实时监控视频;
    预警场景识别模块,用于基于所述实时监控视频对所述目标车辆进行预警场景识别,得到识别结果,所述识别结果包括预警目标;
    感知信息模块,用于对所述实时监控视频进行处理,得到所述预警目标的感知信息;
    预警目标标记模块,用于基于所述感知信息,在所述实时监控视频中对所述预警目标进行标记,得到实时预警视频;
    实时预警视频输出模块,用于将所述实时预警视频输出至显示设备。
  21. 根据权利要求20所述的装置,其特征在于,所述实时监控视频接收模块包括:
    第一视频接收单元,用于接收所述车载终端发送的编码后的实时监控视频,所述编码后的实时监控视频由所述车载终端对所述实时监控视频进行编码后得 到;
    视频解码单元,用于对所述编码后的实时监控视频进行解码,得到所述实时监控视频。
  22. 根据权利要求21所述的装置,其特征在于,所述第一视频接收单元包括:
    第二视频接收单元,用于接收所述车载终端基于5G通信发送的编码后的实时监控视频。
  23. 根据权利要求20所述的装置,其特征在于,所述感知信息模块包括:
    目标检测单元,用于对所述实时监控视频进行目标检测,确定所述预警目标的类型信息;
    目标追踪单元,用于对所述实时监控视频进行目标追踪,确定包括所述预警目标的多帧图像;
    运动轨迹分析单元,用于基于所述多帧图像对所述预警目标进行运动轨迹分析,确定所述预警目标的相对于所述目标车辆的实时相对位置信息、实时速度信息和实时航向角信息;
    感知信息单元,用于将所述预警目标的类型信息、所述实时相对位置信息、所述实时速度信息和所述实时航向角信息作为所述感知信息。
  24. 根据权利要求23所述的装置,其特征在于,所述装置还包括:
    目标坐标系构建单元,用于在所述多帧图像上分别以所述目标车辆为坐标原点,构建所述多帧图像对应的目标坐标系;
    所述预警目标标记模块包括:
    坐标信息确定单元,用于根据所述实时相对位置信息,确定所述预警目标在所述多帧图像对应的目标坐标系中的坐标信息;
    预警目标标记单元,用于根据所述预警目标的类型信息和所述坐标信息, 在所述多帧图像中进行预警目标标记,得到所述实时预警视频。
  25. 根据权利要求23所述的装置,其特征在于,所述装置还包括:
    预测运动轨迹生成单元,用于基于所述实时相对位置信息、所述实时速度信息和所述实时航向角信息生成所述预警目标的预测运动轨迹;
    预警范围判断单元,用于基于所述预测运动轨迹,判断所述预警目标是否在所述目标车辆的预警范围内;
    步骤执行单元,用于当判断为是时,执行所述基于所述感知信息,在所述实时监控视频中对所述预警目标进行标记,得到实时预警视频的步骤。
  26. 根据权利要求20所述的装置,其特征在于,当所述识别结果还包括所述预警场景的类型信息时,所述装置还包括:
    目标显示设备确定单元,用于基于所述预警场景的类型信息,确定所述预警视频的目标显示设备;
    当所述目标显示设备包括车端显示设备时,所述实时预警视频输出模块包括:
    第一视频输出单元,用于将所述实时预警视频输出至所述车端显示设备;
    当所述目标显示设备包括远程驾驶遥控端显示设备时,所述实时预警视频输出模块包括:
    第二视频输出单元,用于将所述实时预警视频输出至所述远程驾驶遥控端显示设备。
  27. 根据权利要求26所述的装置,其特征在于,所述第一视频输出单元包括:
    视频编码单元,用于对所述实时预警视频进行编码,得到编码后的实时预警视频;
    第一视频发送单元,用于向所述车载终端发送所述编码后的实时预警视频, 以使所述车载终端将所述编码后的实时预警视频进行解码,得到所述实时预警视频,以及将所述实时预警视频输出至所述车端显示设备。
  28. 根据权利要求27所述的装置,其特征在于,所述第一视频发送单元包括:
    第二视频发送单元,用于基于5G通信向所述车载终端发送所述编码后的实时预警视频。
  29. 一种车辆预警设备,其特征在于,所述设备包括处理器和存储器,所述存储器中存储有至少一条指令或至少一段程序,所述至少一条指令或所述至少一段程序由所述处理器加载并执行以实现如权利要求2至10任一所述的车辆预警方法。
  30. 一种计算机可读存储介质,其特征在于,所述存储介质中存储有至少一条指令或至少一段程序,所述至少一条指令或所述至少一段程序由处理器加载并执行以实现如权利要求2至10任一所述的车辆预警方法。
PCT/CN2021/084491 2021-03-31 2021-03-31 一种车辆预警方法、系统、装置、设备及存储介质 WO2022205104A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN202180071144.4A CN116368545A (zh) 2021-03-31 2021-03-31 一种车辆预警方法、系统、装置、设备及存储介质
PCT/CN2021/084491 WO2022205104A1 (zh) 2021-03-31 2021-03-31 一种车辆预警方法、系统、装置、设备及存储介质

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2021/084491 WO2022205104A1 (zh) 2021-03-31 2021-03-31 一种车辆预警方法、系统、装置、设备及存储介质

Publications (1)

Publication Number Publication Date
WO2022205104A1 true WO2022205104A1 (zh) 2022-10-06

Family

ID=83455416

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/084491 WO2022205104A1 (zh) 2021-03-31 2021-03-31 一种车辆预警方法、系统、装置、设备及存储介质

Country Status (2)

Country Link
CN (1) CN116368545A (zh)
WO (1) WO2022205104A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117235478A (zh) * 2023-11-14 2023-12-15 深圳万物安全科技有限公司 哑终端数据处理方法、终端设备及可读存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103544806A (zh) * 2013-10-31 2014-01-29 江苏物联网研究发展中心 基于视频绊线规则的重要物资运输车辆监控预警系统
CN107380163A (zh) * 2017-08-15 2017-11-24 上海电气自动化设计研究所有限公司 基于磁导航的汽车智能报警预测系统及其方法
US20180108252A1 (en) * 2015-07-03 2018-04-19 Marcelo Pividori Device known as real time total control digital tachograph (tcdt) for vehicle and other nearby vehicles by means of cameras and mobile connections
CN110009900A (zh) * 2019-03-12 2019-07-12 浙江吉利汽车研究院有限公司 一种车辆监控方法及系统
CN111583716A (zh) * 2020-04-29 2020-08-25 浙江吉利汽车研究院有限公司 一种车辆避障方法、装置、电子设备及存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103544806A (zh) * 2013-10-31 2014-01-29 江苏物联网研究发展中心 基于视频绊线规则的重要物资运输车辆监控预警系统
US20180108252A1 (en) * 2015-07-03 2018-04-19 Marcelo Pividori Device known as real time total control digital tachograph (tcdt) for vehicle and other nearby vehicles by means of cameras and mobile connections
CN107380163A (zh) * 2017-08-15 2017-11-24 上海电气自动化设计研究所有限公司 基于磁导航的汽车智能报警预测系统及其方法
CN110009900A (zh) * 2019-03-12 2019-07-12 浙江吉利汽车研究院有限公司 一种车辆监控方法及系统
CN111583716A (zh) * 2020-04-29 2020-08-25 浙江吉利汽车研究院有限公司 一种车辆避障方法、装置、电子设备及存储介质

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117235478A (zh) * 2023-11-14 2023-12-15 深圳万物安全科技有限公司 哑终端数据处理方法、终端设备及可读存储介质

Also Published As

Publication number Publication date
CN116368545A (zh) 2023-06-30

Similar Documents

Publication Publication Date Title
JP7101255B2 (ja) 目標対象の運動の向きを予測するための方法、車両制御方法及び装置
US11392131B2 (en) Method for determining driving policy
US10810871B2 (en) Vehicle classification system
Nidamanuri et al. A progressive review: Emerging technologies for ADAS driven solutions
US9881221B2 (en) Method and system for estimating gaze direction of vehicle drivers
CN109598066B (zh) 预测模块的效果评估方法、装置、设备和存储介质
CN106980813B (zh) 机器学习的注视生成
JP7239703B2 (ja) 領域外コンテキストを用いたオブジェクト分類
CN110796692A (zh) 用于同时定位与建图的端到端深度生成模型
US20190325597A1 (en) Simultaneous Localization And Mapping Constraints In Generative Adversarial Networks For Monocular Depth Estimation
JP2022507995A (ja) 無人運転車用の障害物回避方法及び装置
CN113240939B (zh) 一种车辆预警方法、装置、设备及存储介质
US10849543B2 (en) Focus-based tagging of sensor data
WO2018103023A1 (zh) 人机混合决策方法和装置
CN111216127A (zh) 一种机器人控制方法、装置、服务器及介质
KR20160135482A (ko) 동적 장애물 움직임 예측 장치 및 그 방법
WO2022205106A1 (zh) 一种车辆预警方法、装置、设备及存储介质
CN113096151B (zh) 对目标的运动信息进行检测的方法和装置、设备和介质
Li et al. DBUS: Human driving behavior understanding system
CN109070882A (zh) 利用摄像头影像的行驶信息提供方法及装置
WO2022205104A1 (zh) 一种车辆预警方法、系统、装置、设备及存储介质
US20220189210A1 (en) Occlusion-Aware Prediction of Human Behavior
CN114298908A (zh) 一种障碍物展示方法、装置、电子设备及存储介质
KR20230086457A (ko) 화재 감지 시스템의 구축을 위한 전차 장치 및 그 방법
JP7269694B2 (ja) 事象発生推定のための学習データ生成方法・プログラム、学習モデル及び事象発生推定装置

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: 21933774

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: 21933774

Country of ref document: EP

Kind code of ref document: A1