WO2021189970A1 - 交通信号灯的智能控制方法及系统、存储介质、计算设备 - Google Patents

交通信号灯的智能控制方法及系统、存储介质、计算设备 Download PDF

Info

Publication number
WO2021189970A1
WO2021189970A1 PCT/CN2020/136057 CN2020136057W WO2021189970A1 WO 2021189970 A1 WO2021189970 A1 WO 2021189970A1 CN 2020136057 W CN2020136057 W CN 2020136057W WO 2021189970 A1 WO2021189970 A1 WO 2021189970A1
Authority
WO
WIPO (PCT)
Prior art keywords
traffic signal
signal light
traffic
preset
pedestrians
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2020/136057
Other languages
English (en)
French (fr)
Inventor
曾昱为
王健宗
黄章成
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ping An Technology Shenzhen Co Ltd
Original Assignee
Ping An Technology Shenzhen Co Ltd
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 Ping An Technology Shenzhen Co Ltd filed Critical Ping An Technology Shenzhen Co Ltd
Publication of WO2021189970A1 publication Critical patent/WO2021189970A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G08SIGNALLING
    • G08GTRAFFIC CONTROL SYSTEMS
    • G08G1/00Traffic control systems for road vehicles
    • G08G1/09Arrangements for giving variable traffic instructions
    • G08G1/095Traffic lights
    • GPHYSICS
    • G08SIGNALLING
    • G08GTRAFFIC CONTROL SYSTEMS
    • G08G1/00Traffic control systems for road vehicles
    • G08G1/07Controlling traffic signals
    • G08G1/08Controlling traffic signals according to detected number or speed of vehicles
    • GPHYSICS
    • G08SIGNALLING
    • G08GTRAFFIC CONTROL SYSTEMS
    • G08G1/00Traffic control systems for road vehicles
    • G08G1/07Controlling traffic signals
    • G08G1/085Controlling traffic signals using a free-running cyclic timer

Definitions

  • This application relates to the field of intelligent control technology, in particular to an intelligent control method and system for traffic signal lights, storage media, and computing equipment.
  • Traffic signal lights are signal lights that direct traffic operation, and are generally composed of red, green, and yellow lights.
  • a red light means no traffic, a green light means permission to pass, and a yellow light means a warning.
  • the existing traffic signal light system is a work equipment that is planned based on a static time interval scheduling algorithm, which switches the color of the signal light according to a preset time period.
  • the inventor found that the biggest disadvantage of such a planning algorithm is that it cannot dynamically adapt to changes and changes in the flow of people and vehicles. At some traffic signal intersections, there are often situations where there are clearly no pedestrians, but vehicles still have to wait for the red light for the previous section, or there are clearly no vehicles but pedestrians have to wait.
  • this application is proposed to provide an intelligent control method and system, storage medium, and computing device for traffic lights that overcome the above-mentioned problems or at least partially solve the above-mentioned problems.
  • an intelligent control method for traffic signal lights including:
  • the traffic signal lamp has the preset starting condition of the dual-track operating mode, controlling the traffic signal lamp to operate in the preset dual-track operating mode;
  • the color control mechanism of the traffic signal light switches between a preset static time allocation mechanism and a visual dynamic control mechanism according to the image information of the predetermined range of the traffic signal light.
  • an intelligent control system for traffic lights including:
  • the acquisition module is suitable for acquiring the operating parameters of the traffic signal light
  • the judging module is suitable for judging whether the traffic signal has a preset starting condition of the dual-track working mode according to the operating parameters of the traffic signal;
  • the first control module is adapted to set the operating mode of the traffic signal light to the dual-track operating mode when the traffic signal light has a preset starting condition for the dual-track operating mode;
  • the color control mechanism of the traffic signal light switches between a preset static time allocation mechanism and a visual dynamic control mechanism according to the image information of the predetermined range of the traffic signal light.
  • a computer-readable storage medium is also provided, the computer-readable storage medium is used to store program code, and the program code is used to execute an intelligent control method of a traffic signal light;
  • the intelligent control method of the traffic signal light includes:
  • the traffic signal lamp has the preset starting condition of the dual-track operating mode, controlling the traffic signal lamp to operate in the preset dual-track operating mode;
  • the color control mechanism of the traffic signal light switches between a preset static time allocation mechanism and a visual dynamic control mechanism according to the image information of the traffic road controlled by the traffic signal light.
  • a computing device including a processor and a memory:
  • the memory is used to store program code and transmit the program code to the processor
  • the processor is configured to execute an intelligent control method for traffic signal lights according to instructions in the program code
  • the intelligent control method of the traffic signal light includes:
  • the traffic signal lamp has the preset starting condition of the dual-track operating mode, controlling the traffic signal lamp to operate in the preset dual-track operating mode;
  • the color control mechanism of the traffic signal light switches between a preset static time allocation mechanism and a visual dynamic control mechanism according to the image information of the traffic road controlled by the traffic signal light.
  • This application provides an intelligent control method, system, storage medium, and computing device for traffic signal lights.
  • the traffic signal lamp can be controlled to operate in a preset dual-track working mode.
  • the dual-track working mode provided in this application uses visual intelligent recognition technology to efficiently identify pedestrians and vehicles for dynamic control of traffic lights, while taking into account the traditional static time allocation mechanism, and using a dual-track work control strategy to control the color of traffic lights , Can improve the resource allocation efficiency of the entire highway system to a certain extent, and also improve the road use experience of pedestrians and vehicles.
  • Fig. 1 shows a schematic flow chart of an intelligent control method for traffic signal lights according to an embodiment of the present application
  • Fig. 2 shows a schematic diagram of a traffic signal light control scene according to an embodiment of the present application
  • Fig. 3 shows a schematic diagram of a traffic signal light control scene according to another embodiment of the present application.
  • Fig. 4 shows a schematic flow chart of a method for intelligent control of a traffic signal light according to another embodiment of the present application
  • Fig. 5 shows a schematic structural diagram of an intelligent control device for a traffic signal light according to an embodiment of the present application
  • Fig. 6 shows a schematic structural diagram of an intelligent control device for a traffic signal light according to another embodiment of the present application.
  • FIG. 1 shows a schematic flow chart of a method for intelligent control of traffic lights according to an embodiment of the present application.
  • the method for intelligent control of traffic lights provided by an embodiment of the present application may at least include the following steps S102 to S106.
  • the traffic signal lamp can communicate with the cloud central server for data transmission.
  • the network communication debugging can be carried out with the cloud central server to ensure that the traffic signal lamp and the cloud central server Maintain a good communication connection between them to provide a network communication basis for the subsequent intelligent control of traffic lights.
  • a wireless communication module may be provided in the traffic signal light, and the network connection parameters of the traffic signal light may be acquired based on the wireless communication module.
  • the traffic signal lamp can emit red, green, and yellow prompts, there will be a light-emitting device inside. Therefore, when obtaining the working parameters of the hardware device of the traffic signal light, the working parameters of the light-emitting device can be obtained, that is, whether the light-emitting device is normal Luminous working parameters.
  • the hardware device in this embodiment may also include an image collection device for collecting image information within a preset range of the traffic signal light, for example, collecting image information of a traffic road under the jurisdiction of the traffic signal light to which the image collection device belongs. Therefore, when acquiring the operating parameters of the hardware device of the traffic signal light, it is also necessary to acquire the operating parameters of the image acquisition device to determine whether it can normally acquire image information.
  • the image acquisition device may be a camera device, a camera device, and other related devices that can realize image acquisition.
  • the operating parameters of other hardware devices of the traffic signal light can also be acquired. Go into details.
  • the image recognition parameters of the processing device may include the recognition accuracy of the object in the image information.
  • the processing device may use a preset target detection algorithm to perform object recognition on the image information collected by the image acquisition device to identify the object.
  • test pictures can be pre-stored in the processing device, communicate with the cloud central server by simulating a normal work process, and then the image recognition parameters of the processing device can be obtained according to the recognition result and related reaction actions.
  • the processing device may be a processor, a processing chip, or other components that can execute an image detection algorithm.
  • the processing device may be integrated in a traffic signal light or an independent device connected to the traffic signal.
  • the operating parameters of the traffic signal light can be acquired periodically during the initial work phase or the work relay phase of the traffic signal light.
  • the above embodiments respectively introduced the network connection parameters of the traffic signal lamp, the operating parameters of the hardware device of the traffic signal lamp, and the image recognition parameters of the processing device in the traffic signal lamp.
  • the operating parameters of the traffic signal lamp can be obtained. One or more.
  • the network connection parameters, the operating parameters of the hardware devices, and the image recognition parameters of the processing devices mentioned in the above-mentioned embodiment are also acquired.
  • the network connection parameters, hardware device parameters, and the algorithm accuracy of the processing device described in the foregoing embodiments it is also possible to obtain other operating parameters of the traffic signal lamp, which this application does not do. limited.
  • the operating parameters of the traffic signal lamp have been obtained in the above step S102, and further, it can be determined whether the traffic signal lamp has the preset starting condition of the dual-track operating mode according to the operating parameters of the traffic signal lamp.
  • step S102 when acquiring the operating parameters of the traffic signal light, the network connection parameters of the traffic signal light, the operating parameters of the hardware device of the traffic signal light, and the image recognition parameters of the processing equipment in the traffic signal light can be acquired respectively.
  • each type of operating parameter can be used as a basis for judging whether the traffic signal light has the preset opening condition of the dual-track operating mode.
  • the traffic signal light when judging whether the traffic signal light has the preset dual-track operating mode on condition, it is necessary to judge whether the network is connected normally according to the network connection parameters of the traffic signal light; judging whether the hardware device is working normally according to the working parameters of the hardware device of the traffic signal light; According to the image recognition parameters of the processing device in the traffic signal light, it is judged whether the processing device has a certain degree of recognition accuracy when executing the target detection algorithm, and the above three judgment steps need to be executed.
  • the traffic signal has the preset dual-track operating mode opening conditions; if the result of the above three judgment steps is no, it is judged that any device in the traffic signal is not In a normal working state, it can be judged that the traffic signal light does not have the preset opening conditions for the dual-track working mode.
  • S106 If the traffic signal lamp has the preset starting condition of the dual-track operating mode, control the traffic signal lamp to operate in the preset dual-track operating mode. In the dual-track working mode, the color control mechanism of the traffic signal light switches between the preset static time allocation mechanism and the visual dynamic control mechanism according to the image information of the preset range of the traffic signal light.
  • the dual-track working mode is a dual-track mechanism based on the parallel of the traditional static time allocation mechanism and the visual dynamic control mechanism based on the target detection AI vision technology.
  • the original intention of the static time allocation mechanism is to maintain the maximum circulation of road conditions while ensuring the fair use of pedestrians and vehicles.
  • the static time allocation mechanism pre-stores the flashing time, flashing frequency, and switching sequence of the red, yellow, and green lights.
  • the color control mechanism of the traffic signal light is the preset static time allocation mechanism, the red light, yellow light, and green light can be prompted according to the preset switching sequence, flashing time, and flashing frequency.
  • the color control mechanism of traffic lights is a visual dynamic control mechanism
  • the color of traffic signals and other prompts can be intelligently adjusted according to the dynamic image of the traffic road, and the current environment can be intelligently perceived and processed, especially for vehicles and pedestrians.
  • controlling the traffic signal lamp to operate in a preset dual-track working mode in step S106 may include:
  • S106-1 Collect image information of multiple angles in a preset range of the traffic signal light.
  • image information can include multiple single-frame image frames acquired at intervals, or it can be Continuous image frames within a certain period of time are not limited in this application.
  • S106-2 Identify the number of pedestrians and the number of vehicles in the image information.
  • target detection algorithms can be used, such as the R-CNN series algorithm in the Two-stage method, the SSD (Single Shot MultiBox Detector) algorithm in the One-stage method, and the Yolo (You Only Look Once) algorithm and so on.
  • the specific identification process can be as follows:
  • Step S1 Recognize multiple recognition objects corresponding to different object categories in the image information through the target detection algorithm, and obtain the confidence level of each recognition object.
  • the object category includes at least one of animals, pedestrians, and vehicles.
  • the target detection algorithm can identify all object categories such as vehicles, animals, pedestrians and other objects in the image information, and will simultaneously output the confidence of each identified object, that is, the accuracy of the judgment.
  • Step S2 selecting a plurality of first objects with a confidence level greater than a preset threshold from the plurality of recognition objects according to the confidence level of each recognition object.
  • the recognition object with a confidence level greater than the preset threshold may be used as the first object.
  • the preset threshold may be 0.1 to 0.2, preferably 0.2.
  • the confidence is less than or equal to the preset threshold, the corresponding recognition object will be considered as a misjudgment, that is, ignored.
  • the preset threshold can also be set to other values according to different requirements, which is not limited in this application.
  • Step S3 Recognize and filter out the repeated objects among the multiple first objects to obtain the target object.
  • NMS Non-Maximum Suppression , Non-maximum suppression
  • the threshold can be set to 0.45. If you want to ensure the optimal result, you can also do a grid search based on the test set to find the most accurate optimal value.
  • Step S4 Count the number of pedestrians and the number of vehicles in the target object.
  • the number of pedestrians and the number of vehicles in the image information can be counted.
  • the Yolo-v4 algorithm is preferentially used to identify the number of pedestrians and the number of vehicles in the image information.
  • Yolo-v4 is the best model currently made in the field of target detection between high performance and high detection frame rate. It can guarantee an ultra-high detection speed of 65 frames per second FPS with such a fairly high AP average accuracy of 43.5%.
  • a sufficiently precise detection frame number FPS is also an important indicator.
  • the use of Yolo-v4 algorithm can ensure the real-time visual detection of the system to the greatest extent and adapt to the vehicle speed to the greatest extent. A time difference with the signal switch of the semaphore improves the experience and safety of the application.
  • the above step S106-2 identifies the number of pedestrians and the number of vehicles in the image information, which can be executed in the processing device on the side of the traffic signal, or the image information can be uploaded to the cloud central server via 5G through the traffic signal, and received
  • the cloud center server performs target detection and returns the detection results including the number of pedestrians and the number of vehicles. Specifically, it can be set independently according to the traffic volume and traffic volume of the traffic road where the traffic signal light is located.
  • the target detection algorithm can be executed through the cloud central server to efficiently determine the number of pedestrians and the number of vehicles.
  • the target detection algorithm can be executed locally by the traffic lights to quickly determine the number of pedestrians and vehicles.
  • the visual dynamic control mechanism is a control mechanism based on pedestrian monopoly or vehicle monopoly respectively.
  • the traffic signal light mentioned in this embodiment may include a pedestrian crossing signal light.
  • the warning color of the traffic signal light on the pedestrian side is controlled to be green; or, when the traffic road has only vehicles, the traffic signal light is controlled The color of the tip on the pedestrian side is red.
  • traffic signal lights may also include motor vehicle signal lights and non-motor vehicle signal lights.
  • the traffic signal light can also communicate with other signal lights on the same road direction or on the intersecting traffic road to realize the linkage control of the traffic signal light.
  • the pedestrian crossing signal light is B
  • the motor vehicle signal light is C and D
  • C and D perform traffic lights based on the traffic of the vehicles identified by each Switch
  • B is a red light
  • B is a red light
  • B is the green light.
  • the foregoing embodiment only schematically introduces the traffic signal light control method. In practical applications, it is also possible to predict pedestrian flow, vehicle flow, etc., to achieve intelligent control of the traffic signal light.
  • the method provided in the embodiments of this application can perform intelligent control of traffic lights according to the number of vehicles and pedestrians around the traffic lights, so that the traffic lights can effectively adapt to changes in the flow of people and vehicles, and perform intelligent control when there are only pedestrians or only vehicles. , It can save traffic time while meeting traffic demands, and effectively realize the intelligent control of traffic lights.
  • step S102 determines whether the traffic signal has the preset conditions of the dual-track operating mode according to the network and device status, the method further includes:
  • step S108 if the traffic signal light does not have the preset opening condition of the dual-track working mode, the prompt color of the traffic signal light is sequentially switched according to the preset static time allocation mechanism.
  • the operating parameters of the traffic signal light can also be directionally updated through an air download dynamic update mechanism.
  • Over the air (OTA) online upgrade mechanism is a technology for downloading upgrade packages on remote servers through wireless networks to upgrade systems or applications. Through the OTA download technology, you can download various business menus provided by the network to your mobile phone using OTA mechanism according to your personal preferences with simple operations, and you can also customize specific services according to your own wishes.
  • the traffic signal light side can deploy an over-the-air (OTA) online upgrade mechanism, so that the operator can directionally (designate a certain device or certain devices) in the cloud to easily complete various system configuration parameters Updates, or related software upgrades.
  • OTA over-the-air
  • the operating parameters such as the duration of each prompt color of the signal lamp, the flashing frequency, and the prompt color switching sequence can be mainly updated or modified online through the OTA online update mechanism.
  • the traffic signal lights are provided with processing equipment to execute the target detection algorithm to obtain the number of pedestrians and the number of vehicles. Therefore, the target detection algorithm can also be dynamically upgraded through the OTA online upgrade mechanism to continuously improve the accuracy and granularity of object recognition. In this way, the work effect is continuously improved, and the response strategy to the recognition results can also be dynamically adjusted through OTA upgrade; for example, how the lights change when there are vehicles and pedestrians at the same time; how far away pedestrians or vehicles need to trigger corresponding changes, and so on.
  • the relevant traffic and transportation departments can lead the development of the update and upgrade of the traffic signal operating parameters according to the actual road conditions and urban development plan; Suppliers related to detection algorithms can provide new models based on technical upgrades, and the relevant departments will decide whether to implement upgrades based on actual measurement results and local conditions.
  • the directional configuration of traffic signal lights can be quickly changed and upgraded in the cloud, which greatly improves the operability and flexible response capability of the entire signal light system.
  • Fig. 4 shows a method for intelligent control of a traffic signal light according to another embodiment of the present application. Referring to Fig. 3, it can be seen that the method provided by the embodiment of the present application may include:
  • Step S402 start the traffic lights; control the traffic lights to switch the prompt colors of the traffic lights in sequence according to the static time allocation mechanism; when the traffic lights are in start-up operation, first use the static time allocation mechanism as a control strategy to control the prompt colors of the traffic lights to ensure traffic Road traffic safety.
  • Step S404 acquiring the operating parameters of the traffic signal light
  • Step S406 judging whether the traffic signal has the preset dual-track operating mode turn-on conditions according to the operating parameters; if yes, execute step S408; if not, execute step S420;
  • Step S408 controlling the traffic signal lamp to operate in a preset dual-track working mode
  • Step S410 collecting image information within a preset range of the traffic signal light
  • Step S412 Send the above-mentioned image information to the cloud central server for target detection.
  • the Yolo-V4 target detection algorithm can be used to first identify each object in the image information, and then screen out objects with low confidence and duplicate objects.
  • the target object counts the number of pedestrians and vehicles in the image information;
  • Step S414 receiving the number of pedestrians and the number of vehicles returned by the cloud center server;
  • Step S416 it is judged whether there are specific pedestrians and vehicles at the same time; if yes, go to step 420; if not, go to step S418;
  • Step S418 Control the traffic signal to operate in the pedestrian exclusive mode or the vehicle exclusive mode; among them, the pedestrian exclusive mode means that the traffic signal on the pedestrian side is a green light, and the traffic signal on the vehicle side is a red light; the vehicle exclusive mode refers to The traffic light on the side is green, and the traffic light on the pedestrian side is red
  • step S420 the prompt colors of the traffic signal lights are sequentially switched according to the preset static time allocation mechanism.
  • the method provided in the embodiments of this application adopts advanced Yolo-V4 visual intelligent recognition technology to efficiently recognize pedestrians and vehicles for dynamic control of traffic lights, while taking into account the traditional static time allocation mechanism, and using a dual-track control strategy control
  • the prompt color of the traffic signal light will ensure safety, fairness, stability, efficiency and integration.
  • AI technology is used to empower traditional traffic lights, thereby improving the rationality of the allocation of highway resources, which can improve the allocation efficiency of the entire highway system to a certain extent, and also improve the road use experience of pedestrians and vehicles.
  • the embodiment of the present application also provides a traffic signal light intelligent control system.
  • the traffic signal light intelligent control system provided by the embodiment of the present application may include:
  • the obtaining module 510 is adapted to obtain the operating parameters of the traffic signal light
  • the judging module 520 is suitable for judging whether the traffic signal has the preset starting condition of the dual-track working mode according to the operating parameters of the traffic signal;
  • the first control module 530 is adapted to control the traffic signal lamp to operate in the preset dual-track operating mode when the traffic signal lamp has the preset dual-track operating mode on condition; wherein, in the dual-track operating mode, the color control mechanism of the traffic signal light is based on The image information of the preset range of traffic lights is switched between the preset static time allocation mechanism and the visual dynamic control mechanism.
  • the obtaining module 510 may also be adapted to:
  • the hardware device includes at least one of an image acquisition device and a light-emitting device.
  • the first control module 530 may be further adapted to:
  • the color of the traffic signal light is controlled according to the preset static time allocation mechanism or the visual dynamic control mechanism.
  • the first control module 530 may be further adapted to:
  • the first control module 530 may be further adapted to:
  • control the color of the traffic signal light according to the visual dynamic control mechanism; among them, when the traffic road has only pedestrians, control the traffic signal light to be green on the pedestrian side; or, when the traffic road has only vehicles, control the traffic signal light The color of the tip on the pedestrian side is red.
  • the foregoing system may further include a second control module 540:
  • the second control module 540 is adapted to sequentially switch the prompt colors of the traffic signal lights according to the preset static time allocation mechanism if the traffic signal lights do not have the preset opening conditions of the dual-track working mode.
  • the foregoing system may further include an update module 550;
  • the update module 550 is adapted to update the operating parameters of the traffic signal light directionally through an air download dynamic update mechanism.
  • a computer-readable storage medium is also provided.
  • the computer-readable storage medium may be non-volatile or volatile.
  • the computer-readable storage medium is used to store program codes.
  • the program code is used to execute the intelligent control method of the traffic signal light of any of the above embodiments.
  • a computing device including a processor and a memory:
  • the memory is used to store the program code and transmit the program code to the processor
  • the processor is configured to execute the intelligent control method for traffic signal lights of any one of the above embodiments according to the instructions in the program code.
  • the functional units in the various embodiments of the present application may be physically independent of each other, or two or more functional units may be integrated together, or all functional units may be integrated in one processing unit.
  • the above-mentioned integrated functional unit may be implemented in the form of hardware, or may be implemented in the form of software or firmware.
  • the integrated functional unit is implemented in the form of software and sold or used as an independent product, it can be stored in a computer readable storage medium.
  • the technical solution of the present application is essentially or all or part of the technical solution can be embodied in the form of a software product.
  • the computer software product is stored in a storage medium and includes several instructions to make a computer
  • a computing device for example, a personal computer, a server, or a network device, etc.
  • the aforementioned storage media include: U disk, mobile hard disk, read only memory (ROM), random access memory (RAM), magnetic disks or optical disks and other media that can store program codes.
  • all or part of the steps of the foregoing method embodiments may be implemented by program instructions related to hardware (computing devices such as personal computers, servers, or network devices), and the program instructions may be stored in a computer readable storage.
  • the program instructions when executed by the processor of the computing device, the computing device executes all or part of the steps of the methods described in the embodiments of the present application.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Traffic Control Systems (AREA)

Abstract

一种交通信号灯的智能控制方法、系统、存储介质和计算设备,涉及人工智能领域,该方法包括:获取交通信号灯的运行参数(S102);根据交通信号灯的运行参数判断交通信号灯是否具备预设的双轨工作模式的开启条件(S104);若交通信号灯具备预设的双轨工作模式的开启条件,则控制所述交通信号灯以预设的双轨工作模式运行(S106)。所述方法兼顾视觉智能识别技术和传统的静态时间分配机制,以双轨工作的控制策略控制交通信号灯的提示颜色,能够提高整个公路系统资源的配置效率,也提高行人车辆的公路使用体验。

Description

交通信号灯的智能控制方法及系统、存储介质、计算设备
本申请要求于2020年10月28日提交中国专利局、申请号为202011171429.9,发明名称为“交通信号灯的智能控制方法及系统、存储介质、计算设备”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
本申请涉及智能控制技术领域,特别是一种交通信号灯的智能控制方法及系统、存储介质、计算设备。
背景技术
交通信号灯是指挥交通运行的信号灯,一般由红灯、绿灯、黄灯组成。红灯表示禁止通行,绿灯表示准许通行,黄灯表示警示。
现有的交通信号灯系统是一种基于静态时间间隔调度算法去进行规划的工作设备,它根据预先设定好的时间周期切换信号灯颜色。发明人发现,这样的规划算法的最大缺点就是无法动态适应人流与车流的变迁与变动。在某些交通信号灯路口,经常会出现明明没有行人,但是车辆却还是不得不等上一段的红灯,或者明明没有车辆,行人却必须等待的情况。
同时,伴随着城市与区域发展,人流与车流量也会动态变动,原有的静态系统无法适应这一特点,其预设算法一定会慢慢过时。
技术问题
鉴于上述问题,提出了本申请以便提供一种克服上述问题或者至少部分地解决上述问题的交通信号灯的智能控制方法及系统、存储介质、计算设备。
技术解决方案
根据本申请的一个方面,提供了一种交通信号灯的智能控制方法,包括:
获取交通信号灯的运行参数;
根据所述交通信号灯的运行参数判断所述交通信号灯是否具备预设的双轨工作模式的开启条件;
若所述交通信号灯具备预设的双轨工作模式的开启条件,则控制所述交通信号灯以预设的双轨工作模式运行;
其中,在所述双轨工作模式下,所述交通信号灯的颜色控制机制依据所述交通信号灯预设范围的图像信息在预设的静态时间分配机制和视觉动态控制机制之间进行切换。
根据本申请的另一个方面,还提供了一种交通信号灯智能控制系统,包括:
获取模块,适于获取交通信号灯的运行参数;
判断模块,适于根据所述交通信号灯的运行参数判断所述交通信号灯是否具备预设的双轨工作模式的开启条件;
第一控制模块,适于当所述交通信号灯具备预设的双轨工作模式的开启条件时,将所述交通信号灯的工作模式设置为双轨工作模式;
其中,在所述双轨工作模式下,所述交通信号灯的颜色控制机制依据所述交通信号灯预设范围的图像信息在预设的静态时间分配机制和视觉动态控制机制之间进行切换。
根据本申请的另一个方面,还提供了一种计算机可读存储介质,所述计算机可读存储介质用于存储程序代码,所述程序代码用于执行一种交通信号灯的智能控制方法;
其中,所述交通信号灯的智能控制方法,包括:
获取交通信号灯的运行参数;
根据所述交通信号灯的运行参数判断所述交通信号灯是否具备预设的双轨工作模式的开启条件;
若所述交通信号灯具备预设的双轨工作模式的开启条件,则控制所述交通信号灯以预设的双轨工作模式运行;
其中,在所述双轨工作模式下,所述交通信号灯的颜色控制机制依据所述交通信号灯所控制交通道路的图像信息在预设的静态时间分配机制和视觉动态控制机制之间进行切换。
根据本申请的另一个方面,还提供了一种计算设备,包括处理器以及存储器:
所述存储器用于存储程序代码,并将所述程序代码传输给所述处理器;
所述处理器用于根据所述程序代码中的指令执行一种交通信号灯的智能控制方法;
其中,所述交通信号灯的智能控制方法,包括:
获取交通信号灯的运行参数;
根据所述交通信号灯的运行参数判断所述交通信号灯是否具备预设的双轨工作模式的开启条件;
若所述交通信号灯具备预设的双轨工作模式的开启条件,则控制所述交通信号灯以预设的双轨工作模式运行;
其中,在所述双轨工作模式下,所述交通信号灯的颜色控制机制依据所述交通信号灯所控制交通道路的图像信息在预设的静态时间分配机制和视觉动态控制机制之间进行切换。
有益效果
本申请提供了一种交通信号灯的智能控制方法及系统、存储介质、计算设备,在本申请提供的方法中,根据交通信号灯的运行参数判断交通信号灯具备预设的双轨工作模式的开启条件后,可控制交通信号灯以预设的双轨工作模式运行。本申请提供的双轨工作模式,采用视觉智能识别技术进行行人和车辆高效识别以进行交通信号灯动态控制的同时,兼顾传统的静态时间分配机制,以一种双轨工作的控制策略控制交通信号灯的提示颜色,能够在一定程度上提高整个公路系统资源的配置效率,也提高行人车辆的公路使用体验。
附图说明
图1示出了根据本申请一实施例的交通信号灯的智能控制方法流程示意图;
图2示出了根据本申请一实施例的交通信号灯控制场景示意图;
图3示出了根据本申请另一实施例的交通信号灯控制场景示意图;
图4示出了根据本申请另一实施例的交通信号灯的智能控制方法流程示意图;
图5示出了根据本申请一实施例的交通信号灯的智能控制装置结构示意图;
图6示出了根据本申请另一实施例的交通信号灯的智能控制装置结构示意图。
本发明的最佳实施方式
下面将参照附图更详细地描述本公开的示例性实施例。虽然附图中显示了本公开的示例性实施例,然而应当理解,可以以各种形式实现本公开而不应被这里阐述的实施例所限制。相反,提供这些实施例是为了能够更透彻地理解本公开,并且能够将本公开的范围完整的传达给本领域的技术人员。
图1示出了根据本申请一实施例的交通信号灯的智能控制方法流程示意图,参见图1可知,本申请实施例提供的交通信号灯的智能控制方法至少可以包括以下步骤S102~S106。
S102,获取交通信号灯的运行参数。
在本申请实施例中,对交通信号灯进行智能控制之前,需要先获取交通信号灯的运行参数,可选地,可以包括获取交通信号灯的网络连接参数、获取交通信号灯的硬件设备的工作参数以及获取交通信号灯内处理设备的图像识别参数等。下面分别对获取交通信号灯的运行参数的具体过程进行详细说明。
1、获取交通信号灯的网络连接参数
在本实施例中,交通信号灯可以和云端中心服务器之间进行通信连接,以进行数据传输,获取交通信号灯的网络连接参数时,可以与云端中心服务器进行网络通信调试,确保交通信号灯和云端中心服务器之间保持良好的通信连接,为后续在交通信号灯的智能控制过程中提供网络通信基础。
可选地,交通信号灯中可以设置有无线通信模块,获取交通信号灯的网络连接参数时,可以基于无线通信模块进行获取。
2、获取交通信号灯的硬件设备的工作参数
由于交通信号灯可以发出红绿黄三种提示颜色,其内部则会设置有发光设备,因此,在获取交通信号灯的硬件设备的工作参数时,可以获取发光设备的工作参数,即获取发光设备是否正常发光的工作参数。
另外,本实施例中的硬件设备还可以包括图像采集设备,用于采集交通信号灯预设范围内的图像信息,例如,采集图像采集设备所属交通信号灯所管辖交通道路的图像信息。因此,在获取交通信号灯的硬件设备的工作参数时,还需要获取图像采集设备的工作参数,以判断其是否可以正常采集图像信息。其中,图像采集设备可以是照相设备、摄像设备等可以实现图像采集的相关设备。
实际应用中,由于交通信号灯中还可能设置其他设备组件,因此,除了对上述发光设备、图像采集设备的工作参数进行获取之外,还可以获取交通信号灯的其他硬件设备的工作参数,此处不多赘述。
3、获取交通信号灯内处理设备的图像识别参数
处理设备的图像识别参数,可以包括对图像信息中对象的识别准确度,具体来讲,处理设备可以采用预设的目标检测算法对图像采集设备所采集的图像信息进行对象识别,以识别出其中包含的各个对象。在本实施例中,可以在处理设备中预先存储测试图片,通过模拟正常工作流程与云端中心服务器进行通信,进而根据识别结果与相关的反应动作可以获取到处理设备的图像识别参数。该处理设备可以为可执行图像检测算法的处理器、处理芯片或是其他组件,该处理设备可以集成于交通信号灯中,或是与交通信号连接的独立设置的设备。
基于本申请实施例提供的方法,通过对交通信号灯的运行参数进行获取,可以有效为判断交通信号灯是否正常运行提供准确的判断依据。实际应用中,可以周期性地在交通信号灯的开始工作阶段或是工作中继阶段过程中对交通信号灯的运行参数进行获取。上述实施例分别介绍了获取交通信号灯的网络连接参数、交通信号灯的硬件设备的工作参数以及交通信号灯内处理设备的图像识别参数,实际应用中,可以获取交通信号灯的运行参数时,可以获取其中的一种或是多种。优选地,获取交通信号灯的运行参数时,同时获取上述实施例提及的网络连接参数、硬件设备的工作参数以及处理设备的图像识别参数。并且,除了对上述实施例所介绍的对网络连接参数、硬件设备参数以及处理设备的算法准确度进行获取之外,还可以对交通信号灯的其他方面的运行参数进行获取,本申请对此不做限定。
S104,根据交通信号灯的运行参数判断交通信号灯是否具备预设的双轨工作模式的开启条件。
上述步骤S102中已经获取出交通信号灯的运行参数,进一步地,可以根据交通信号灯的运行参数判断交通信号灯是否具备预设的双轨工作模式的开启条件。
上述步骤S102提及,获取交通信号灯的运行参数时,可以分别获取交通信号灯的网络连接参数、获取交通信号灯的硬件设备的工作参数以及获取交通信号灯内处理设备的图像识别参数。
进一步地,在判断交通信号灯是否具备预设的双轨工作模式的开启条件时,可以根据所获取到的一种或多种运行参数进行判断。优选地,可将根据各类型运行参数同时作为交通信号灯是否具备预设的双轨工作模式的开启条件的判断依据。举例来讲,判断交通信号灯是否具备预设的双轨工作模式的开启条件时,需要根据交通信号灯的网络连接参数判断网络是否连接正常;根据交通信号灯的硬件设备的工作参数判断硬件设备是否正常工作;根据交通信号灯内处理设备的图像识别参数判断处理设备在执行目标检测算法时是否具有一定的识别准确度,上述三个判断步骤均需执行。若上述三个判断步骤的结果均为是,则可以判断交通信号灯具备预设的双轨工作模式的开启条件;若上述三个判断步骤的结果有一项为否,即判断交通信号灯中任一设备不处于正常工作状态,则可以判断交通信号灯不具备预设的双轨工作模式的开启条件。
S106,若交通信号灯具备预设的双轨工作模式的开启条件,则控制交通信号灯以预设的双轨工作模式运行。在双轨工作模式下,交通信号灯的颜色控制机制依据交通信号灯预设范围的图像信息在预设的静态时间分配机制和视觉动态控制机制之间进行切换。
在本申请实施例中,双轨工作模式,是一种基于传统的静态时间分配机制与基于目标检测AI视觉技术的视觉动态控制机制并行的双轨机制。静态时间分配机制的初衷是维持路况最大程度流通的同时保证行人与车辆尽可能公平的使用权。
当行人与车辆同时对公路具有使用需求的时候,就需要传统的静态时间分配机制以保证公平。一般情况下,静态时间分配机制中预先存储了红灯、黄灯、绿灯各自的闪烁时间、闪烁频率以及切换顺序。当交通信号灯的颜色控制机制为预设的静态时间分配机制时,则可以按照预设的切换顺序以及闪烁时间、闪烁频率进行红灯、黄灯、绿灯的提示。
当交通信号灯的颜色控制机制为视觉动态控制机制,可以根据交通道路的动态图像对交通信号等的提示颜色进行智能调控,对当前环境进行智能化感知和处理,特别是可以实现对车辆与行人的AI识别能力,实现交通信号灯的智能切换。
在本申请可选实施例中,上述步骤S106控制交通信号灯以预设的双轨工作模式运行可以包括:
S106-1,采集交通信号灯预设范围的多个角度的图像信息。采集图像信息时,可以采集交通信号预设范围的多个角度、且距离交通信号等设定距离范围内的图像信息,该图像信息可以包括间隔获取到的多个单帧图像帧,也可以是在一定时间段内的连续图像帧,本申请对此不做限定。
S106-2,识别图像信息中的行人数量和车辆数量。
识别图像信息汇总的行人数量和车辆数量时,可以采用目标检测算法,例如Two-stage方法中的R-CNN系列算法,以及One-stage方法中的SSD(Single Shot MultiBox Detector)算法、Yolo(You Only Look Once)算法等等。具体识别过程可以如下:
步骤S1,通过目标检测算法识别图像信息中对应不同对象类别的多个识别对象,并获取各识别对象的置信度。其中,对象类别包括动物、行人、车辆中至少之一。
实际应用中,目标检测算法可以把图像信息中所有的对象类别如车辆,动物,行人等对象分别识别出来,并且会同时输出各识别对象的置信度,即判断的准确程度。
步骤S2,依据各识别对象的置信度在多个识别对象中选取置信度大于预设阈值的多个第一对象。
由于各识别对象具有各自的置信度,此时,可以将置信度大于预设阈值的识别对象作为第一对象。可选地,预设阈值可以为0.1~0.2,优选为0.2,当置信度小于或等于预设阈值时,那么对应的识别对象则会认为是错判,即忽略不计。实际应用中,还可以根据不同的需求将预设阈值设置为其他数值,本申请对此不做限定。
步骤S3,识别并筛除多个第一对象中的重复对象,获得目标对象。
基于图像信息进行目标检测时,对于同一识别对象而言,可能会检测到多次,此时,可以采用重复检测的方式筛除第一对象中的重复对象,具体可采用NMS(Non-Maximum Suppression,非极大值抑制)算法检测第一对象中的重复对象。可选地,对重复对象进行筛除时,其阈值可以设置为0.45,如果要保证最优结果也可以根据测试集去做网格搜索找最精准的最优值。
步骤S4,统计目标对象中的行人数量和车辆数量。
当对识别对象进行去重筛选后,即可统计图像信息中的行人数量和车辆数量。
本申请实施例中优先采用Yolo-v4算法识别图像信息中的行人数量和车辆数量,Yolo-v4是目标检测领域目前在高性能与高检测帧率之间做的最好的一个模型。它可以做到在43.5%的这样一个相当高的AP平均精确度的情况下保证每秒65帧FPS的超高检测速度。对于信号灯实时检测而言,除了要求比较精确的检测精度,足够精密的检测帧数FPS也是很重要的一个指标,采用Yolo-v4算法能够最大程度保证系统的视觉检测实时性,最大程度适应车辆速度与信号灯信号切换的一个时差,提高应用上的体验感与安全性。
实际应用中,上述步骤S106-2识别图像信息中的行人数量和车辆数量,可以在交通信号灯侧的处理设备中执行,也可以通过交通信号灯将图像信息通过5G上传至云端中心服务器,并接收由云端中心服务器进行目标检测后返回包括行人数量和车辆数量的检测结果,具体可以根据交通信号灯所处交通道路的人流量和车流量的大小进行自主设置,当交通信号灯所处交通道路的人流量和车流量较大时,由于云端中心服务器具备较大的数据处理能力,因此,可以通过云端中心服务器执行目标检测算法,高效确定出行人数量和车辆数量。当交通信号灯所处交通道路的人流量和车流量较小时,可以通过交通信号灯本地执行目标检测算法,进而快速确定出行人数量和车辆数量。
S106-3,根据行人数量和车辆数量依据预设的静态时间分配机制或视觉动态控制机制控制交通信号灯提示颜色。
具体来讲,可以根据行人数量和车辆数量判断交通道路是否同时具有行人和车辆;若是,则根据预设的静态时间分配机制顺序切换交通信号灯的提示颜色;若否,则按照视觉动态控制机制控制交通信号灯提示颜色。
本申请实施例中,视觉动态控制机制是分别根据行人独占或车辆独占的控制机制。本实施例提及的交通信号灯可以包括人行横道信号灯,可选地,在交通道路只具有行人时,控制交通信号灯在行人侧的提示颜色为绿色;或,在交通道路只具有车辆时,控制交通信号灯在行人侧的提示颜色为红色。
实际应用中,交通信号灯还可以包括机动车信号灯以及非机动车信号灯,此时,该交通信号灯还可以与同一道路方向上或是相交交通道路的其他信号灯进行通信,实现交通信号灯的联动控制。
举例来讲,如图2所示,假设人行横道信号灯为A,在交通道路只具有行人时,A为绿灯;在交通道路只具有车辆时,A为红灯。
如图3所示,假设人行横道信号灯为B,机动车信号灯为C和D,假设C和D周围只有车辆,没有行人,此时,C和D进行依据各自所识别出的车辆的流量进行红绿灯的切换,B为红灯;假设C和D周围只有行人,没有车辆,那么,C和D进为红灯,B为绿灯。上述实施例仅示意性的介绍了交通信号灯控制方法,实际应用中,还可以根据对行人流量、车辆流量等进行预测,以实现交通信号灯的智能控制。
本申请实施例提供的方法,可根据交通信号灯周围的车辆和行人数量进行交通信号灯的智能控制,进而使得交通信号灯有效适应人流与车流的变动,在只有行人或只有车辆的情况下,进行智能调控,满足交通需求的同时节省交通时间,有效实现交通信号灯的智能控制。
在本申请一可选实施例中,继续参见图1可知,上述步骤S102根据网络与设备状态判断交通信号灯是否具备双轨工作模式的预设条件之后,还包括:
步骤S108,若交通信号灯不具备预设的双轨工作模式的开启条件,则根据预设的静态时间分配机制顺序切换交通信号灯的提示颜色。
在本申请一可选实施例中,还可以通过空中下载动态更新机制定向更新交通信号灯的运行参数。空中下载(Over the Air,OTA)在线升级机制,是通过无线网络下载远程服务器上的升级包,对系统或应用进行升级的技术。通过OTA空中下载技术,只需要进行简单操作,就可以按照个人喜好把网络所提供的各种业务菜单利用OTA机制下载到手机中,并且还可以根据自己的意愿定制具体业务。
在本申请实施例中,交通信号灯侧可以部署空中下载(Over the Air,OTA)在线升级机制,使得操作员可以定向地(指定某个或者某些设备)在云端轻松完成各种系统配置参数的更新,或者是相关的软件升级。
可选地,主要可以通过OTA在线升级机制在线升级或修改信号灯各提示颜色的持续时间、闪烁频率、提示颜色切换顺序等运行参数。另外,上述实施例提及,交通信号灯中设置有处理设备执行目标检测算法获取行人数量和车辆数量,因此,还可以通过OTA在线升级机制动态升级目标检测算法,持续提升对象识别的精度和粒度,从而持续提升工作效果,而对于识别结果的反应策略也可以通过OTA升级动态调整;例如同时有车辆和行人时灯该如何变化;距离多远的行人或者车辆需要触发什么样的对应变化等等。
基于本申请实施例提供的方法,通过在交通信号灯侧设置OTA在线升级机制,可以根据交通运输相关部门根据实际路况,城市发展规划来主导制定交通信号灯的运行参数的更新和升级应该由;对于目标检测算法相关的供应商可根据技术升级提供新的模型,并根据实测效果与地方因地制宜由有关部门决定是否落地升级使用。基于本申请实施例提供的方法可以在云端对交通信号灯快速进行定向配置的改变与升级,极大提高了整个信号灯系统的可操作性与灵活响应能力。
图4示出了根据本申请又一实施例的交通信号灯智能控制方法,参见图3可知,本申请实施例提供的方法可以包括:
步骤S402,启动交通信号灯;控制交通信号灯以静态时间分配机制顺序切换交通信号灯的提示颜色;在交通信号灯进行启动运行时,先以静态时间分配机制作为控制策略控制交通信号灯的提示颜色,以确保交通道路的交通安全。
步骤S404,获取交通信号灯的运行参数;
步骤S406,根据运行参数判断交通信号灯是否具备预设的双轨工作模式的开启条件;若是,则执行步骤S408;若否,则执行步骤S420;
步骤S408,控制交通信号灯以预设的双轨工作模式运行;
步骤S410,采集交通信号灯预设范围内的图像信息;
步骤S412,将上述图像信息发送到云端中心服务器进行目标检测,具体可以采用Yolo-V4目标检测算法先识别出图像信息中的各个对象,进而对经过筛除置信度较低的对象以及重复对象的目标对象统计图像信息中的行人数量和车辆数量;
步骤S414,接收云端中心服务器返回的行人数量和车辆数量;
步骤S416,判断是否同时具体行人和车辆;若是,则执行步骤420;若否,则执行步骤S418;
步骤S418,控制交通信号灯以行人独占模式或车辆独占模式运行;其中,行人独占模式是指,在行人侧的交通信号灯为绿灯,车辆侧的交通信号灯为红灯;车辆独占模式是指,在车辆侧的交通信号灯为绿灯,行人侧的交通信号灯为红灯
步骤S420,根据预设的静态时间分配机制顺序切换交通信号灯的提示颜色。
本申请实施例提供的方法,采用先进的Yolo-V4视觉智能识别技术进行行人和车辆高效识别以进行交通信号灯动态控制的同时,兼顾传统的静态时间分配机制,以一种双轨工作的控制策略控制交通信号灯的提示颜色,将保证安全、公平、稳定、效率于一体化。其中,通过AI技术赋能传统的交通信号灯,从而提高了公路资源在分配上的合理性,能够在一定程度上提高整个公路系统资源的配置效率,也提高行人车辆的公路使用体验。
基于同一发明构思,本申请实施例还提供了交通信号灯智能控制系统,参见图5可知,本申请实施例提供的交通信号灯智能控制系统可以包括:
获取模块510,适于获取交通信号灯的运行参数;
判断模块520,适于根据交通信号灯的运行参数判断交通信号灯是否具备预设的双轨工作模式的开启条件;
第一控制模块530,适于当交通信号灯具备预设的双轨工作模式的开启条件时,控制交通信号灯以预设的双轨工作模式运行;其中,在双轨工作模式下,交通信号灯的颜色控制机制依据交通信号灯预设范围的图像信息在预设的静态时间分配机制和视觉动态控制机制之间进行切换。
在本申请一可选实施例中,获取模块510还可以适于:
获取交通信号灯的网络连接参数;和/或
获取交通信号灯的硬件设备的工作参数;和/或
获取交通信号灯内处理设备的图像识别参数;
其中,硬件设备包括图像采集设备、发光设备中至少之一。
在本申请一可选实施例中,第一控制模块530还可以适于:
采集交通信号灯预设范围的多个角度的图像信息;
识别图像信息中的行人数量和车辆数量;
根据行人数量和车辆数量依据预设的静态时间分配机制或视觉动态控制机制控制交通信号灯提示颜色。
在本申请一可选实施例中,第一控制模块530还可以适于:
通过目标检测算法识别图像信息中对应不同对象类别的多个识别对象,并获取各识别对象的置信度;其中,对象类别包括动物、行人、车辆中至少之一;
依据各识别对象的置信度在多个识别对象中选取置信度大于预设阈值的多个第一对象;
识别并筛除多个第一对象中的重复对象,获得目标对象;
统计目标对象中的行人数量和车辆数量。
在本申请一可选实施例中,第一控制模块530还可以适于:
根据行人数量和车辆数量判断交通道路是否同时具有行人和车辆;
若是,则根据预设的静态时间分配机制顺序切换交通信号灯的提示颜色;
若否,则按照视觉动态控制机制控制交通信号灯提示颜色;其中,在交通道路只具有行人时,控制交通信号灯在行人侧的提示颜色为绿色;或,在交通道路只具有车辆时,控制交通信号灯在行人侧的提示颜色为红色。
在本申请一可选实施例中,如图6所示,上述系统还可以包括第二控制模块540:
第二控制模块540,适于若交通信号灯不具备预设的双轨工作模式的开启条件,则根据预设的静态时间分配机制顺序切换交通信号灯的提示颜色。
在本申请一可选实施例中,如图5所示,上述系统还可以包括更新模块550;
更新模块550,适于通过空中下载动态更新机制定向更新交通信号灯的运行参数。
在本申请一可实施例中,还提供了一种计算机可读存储介质,该计算机可读存储介质可以是非易失性,也可以是易失性,计算机可读存储介质用于存储程序代码,程序代码用于执行上述任一实施例的交通信号灯的智能控制方法。
在本申请一可实施例中,还提供了一种计算设备,包括处理器以及存储器:
存储器用于存储程序代码,并将程序代码传输给处理器;
处理器用于根据程序代码中的指令执行上述任一实施例的交通信号灯的智能控制方法。
所属领域的技术人员可以清楚地了解到,上述描述的系统、装置、模块和单元的具体工作过程,可以参考前述方法实施例中的对应过程,为简洁起见,在此不另赘述。
另外,在本申请各个实施例中的各功能单元可以物理上相互独立,也可以两个或两个以上功能单元集成在一起,还可以全部功能单元都集成在一个处理单元中。上述集成的功能单元既可以采用硬件的形式实现,也可以采用软件或者固件的形式实现。
本领域普通技术人员可以理解:所述集成的功能单元如果以软件的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储介质中。基于这样的理解,本申请的技术方案本质上或者该技术方案的全部或部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质中,其包括若干指令,用以使得一台计算设备(例如个人计算机,服务器,或者网络设备等)在运行所述指令时执行本申请各实施例所述方法的全部或部分步骤。而前述的存储介质包括:U盘、移动硬盘、只读存储器(ROM)、随机存取存储器(RAM),磁碟或者光盘等各种可以存储程序代码的介质。
或者,实现前述方法实施例的全部或部分步骤可以通过程序指令相关的硬件(诸如个人计算机,服务器,或者网络设备等的计算设备)来完成,所述程序指令可以存储于一计算机可读取存储介质中,当所述程序指令被计算设备的处理器执行时,所述计算设备执行本申请各实施例所述方法的全部或部分步骤。
最后应说明的是:以上各实施例仅用以说明本申请的技术方案,而非对其限制;尽管参照前述各实施例对本申请进行了详细的说明,本领域的普通技术人员应当理解:在本申请的精神和原则之内,其依然可以对前述各实施例所记载的技术方案进行修改,或者对其中部分或者全部技术特征进行等同替换;而这些修改或者替换,并不使相应技术方案脱离本申请的保护范围。

Claims (20)

  1. 一种交通信号灯的智能控制方法,其中,包括:
    获取交通信号灯的运行参数;
    根据所述交通信号灯的运行参数判断所述交通信号灯是否具备预设的双轨工作模式的开启条件;
    若所述交通信号灯具备预设的双轨工作模式的开启条件,则控制所述交通信号灯以预设的双轨工作模式运行;
    其中,在所述双轨工作模式下,所述交通信号灯的颜色控制机制依据所述交通信号灯所控制交通道路的图像信息在预设的静态时间分配机制和视觉动态控制机制之间进行切换。
  2. 根据权利要求1所述的方法,其中,所述获取交通信号灯的运行参数,包括:
    获取所述交通信号灯的网络连接参数;和/或
    获取所述交通信号灯的硬件设备的工作参数;和/或
    获取所述交通信号灯内处理设备的图像识别参数;
    其中,所述硬件设备包括图像采集设备、发光设备中至少之一。
  3. 根据权利要求1所述的方法,其中,所述控制所述交通信号灯以预设的双轨工作模式运行,包括:
    采集所述交通信号灯所控制交通道路的多个角度的图像信息;
    识别所述图像信息中的行人数量和车辆数量;
    根据所述行人数量和车辆数量依据预设的静态时间分配机制或视觉动态控制机制控制所述交通信号灯提示颜色。
  4. 根据权利要求3所述的方法,其中,所述识别所述图像信息中的行人数量和车辆数量,包括:
    通过目标检测算法识别所述图像信息中对应不同对象类别的多个识别对象,并获取各所述识别对象的置信度;其中,所述对象类别包括动物、行人、车辆中至少之一;
    依据各所述识别对象的置信度在所述多个识别对象中选取置信度大于预设阈值的多个第一对象;
    识别并筛除所述多个第一对象中的重复对象,获得目标对象;
    统计所述目标对象中的行人数量和车辆数量。
  5. 根据权利要求3所述的方法,其中,所述根据所述行人数量和车辆数量依据预设的静态时间分配机制或视觉动态控制机制控制所述交通信号灯提示颜色,包括:
    根据所述行人数量和车辆数量判断所述交通道路是否同时具有行人和车辆;
    若是,则根据预设的静态时间分配机制顺序切换所述交通信号灯的提示颜色;
    若否,则按照视觉动态控制机制控制所述交通信号灯提示颜色;其中,在所述交通道路只具有行人时,控制所述交通信号灯在行人侧的提示颜色为绿色;或,在所述交通道路只具有车辆时,控制所述交通信号灯在行人侧的提示颜色为红色。
  6. 根据权利要求1-5任一项所述的方法,其中,所述根据所述网络与设备状态判断所述交通信号灯是否具备双轨工作模式的预设条件之后,还包括:
    若所述交通信号灯不具备预设的双轨工作模式的开启条件,则根据预设的静态时间分配机制顺序切换所述交通信号灯的提示颜色。
  7. 根据权利要求1-5任一项所述的方法,其中,所述方法还包括:
    通过空中下载动态更新机制定向更新所述交通信号灯的运行参数。
  8. 一种交通信号灯智能控制系统,其中,包括:
    获取模块,适于获取交通信号灯的运行参数;
    判断模块,适于根据所述交通信号灯的运行参数判断所述交通信号灯是否具备预设的双轨工作模式的开启条件;
    第一控制模块,适于当所述交通信号灯具备预设的双轨工作模式的开启条件时,将所述交通信号灯的工作模式设置为双轨工作模式;
    其中,在所述双轨工作模式下,所述交通信号灯的颜色控制机制依据所述交通信号灯预设范围的图像信息在预设的静态时间分配机制和视觉动态控制机制之间进行切换。
  9. 一种计算机可读存储介质,其中,所述计算机可读存储介质用于存储程序代码,所述程序代码用于执行一种交通信号灯的智能控制方法;
    其中,所述交通信号灯的智能控制方法,包括:
    获取交通信号灯的运行参数;
    根据所述交通信号灯的运行参数判断所述交通信号灯是否具备预设的双轨工作模式的开启条件;
    若所述交通信号灯具备预设的双轨工作模式的开启条件,则控制所述交通信号灯以预设的双轨工作模式运行;
    其中,在所述双轨工作模式下,所述交通信号灯的颜色控制机制依据所述交通信号灯所控制交通道路的图像信息在预设的静态时间分配机制和视觉动态控制机制之间进行切换。
  10. 根据权利要求9所述的计算机可读存储介质,其中,所述获取交通信号灯的运行参数,包括:
    获取所述交通信号灯的网络连接参数;和/或
    获取所述交通信号灯的硬件设备的工作参数;和/或
    获取所述交通信号灯内处理设备的图像识别参数;
    其中,所述硬件设备包括图像采集设备、发光设备中至少之一。
  11. 根据权利要求9所述的计算机可读存储介质,其中,所述控制所述交通信号灯以预设的双轨工作模式运行,包括:
    采集所述交通信号灯所控制交通道路的多个角度的图像信息;
    识别所述图像信息中的行人数量和车辆数量;
    根据所述行人数量和车辆数量依据预设的静态时间分配机制或视觉动态控制机制控制所述交通信号灯提示颜色。
  12. 根据权利要求11所述的计算机可读存储介质,其中,所述识别所述图像信息中的行人数量和车辆数量,包括:
    通过目标检测算法识别所述图像信息中对应不同对象类别的多个识别对象,并获取各所述识别对象的置信度;其中,所述对象类别包括动物、行人、车辆中至少之一;
    依据各所述识别对象的置信度在所述多个识别对象中选取置信度大于预设阈值的多个第一对象;
    识别并筛除所述多个第一对象中的重复对象,获得目标对象;
    统计所述目标对象中的行人数量和车辆数量。
  13. 根据权利要求11所述的计算机可读存储介质,其中,所述根据所述行人数量和车辆数量依据预设的静态时间分配机制或视觉动态控制机制控制所述交通信号灯提示颜色,包括:
    根据所述行人数量和车辆数量判断所述交通道路是否同时具有行人和车辆;
    若是,则根据预设的静态时间分配机制顺序切换所述交通信号灯的提示颜色;
    若否,则按照视觉动态控制机制控制所述交通信号灯提示颜色;其中,在所述交通道路只具有行人时,控制所述交通信号灯在行人侧的提示颜色为绿色;或,在所述交通道路只具有车辆时,控制所述交通信号灯在行人侧的提示颜色为红色。
  14. 根据权利要求9-13任一项所述的计算机可读存储介质,其中,所述根据所述网络与设备状态判断所述交通信号灯是否具备双轨工作模式的预设条件之后,还包括:
    若所述交通信号灯不具备预设的双轨工作模式的开启条件,则根据预设的静态时间分配机制顺序切换所述交通信号灯的提示颜色。
  15. 根据权利要求9-13任一项所述的计算机可读存储介质,其中,所述方法还包括:
    通过空中下载动态更新机制定向更新所述交通信号灯的运行参数。
  16. 一种计算设备,其中,包括处理器以及存储器:
    所述存储器用于存储程序代码,并将所述程序代码传输给所述处理器;
    所述处理器用于根据所述程序代码中的指令执行一种交通信号灯的智能控制方法;
    其中,所述交通信号灯的智能控制方法,包括:
    获取交通信号灯的运行参数;
    根据所述交通信号灯的运行参数判断所述交通信号灯是否具备预设的双轨工作模式的开启条件;
    若所述交通信号灯具备预设的双轨工作模式的开启条件,则控制所述交通信号灯以预设的双轨工作模式运行;
    其中,在所述双轨工作模式下,所述交通信号灯的颜色控制机制依据所述交通信号灯所控制交通道路的图像信息在预设的静态时间分配机制和视觉动态控制机制之间进行切换。
  17. 根据权利要求16所述的计算设备,其中,所述获取交通信号灯的运行参数,包括:
    获取所述交通信号灯的网络连接参数;和/或
    获取所述交通信号灯的硬件设备的工作参数;和/或
    获取所述交通信号灯内处理设备的图像识别参数;
    其中,所述硬件设备包括图像采集设备、发光设备中至少之一。
  18. 根据权利要求16所述的计算设备,其中,所述控制所述交通信号灯以预设的双轨工作模式运行,包括:
    采集所述交通信号灯所控制交通道路的多个角度的图像信息;
    识别所述图像信息中的行人数量和车辆数量;
    根据所述行人数量和车辆数量依据预设的静态时间分配机制或视觉动态控制机制控制所述交通信号灯提示颜色。
  19. 根据权利要求16所述的计算设备,其中,所述识别所述图像信息中的行人数量和车辆数量,包括:
    通过目标检测算法识别所述图像信息中对应不同对象类别的多个识别对象,并获取各所述识别对象的置信度;其中,所述对象类别包括动物、行人、车辆中至少之一;
    依据各所述识别对象的置信度在所述多个识别对象中选取置信度大于预设阈值的多个第一对象;
    识别并筛除所述多个第一对象中的重复对象,获得目标对象;
    统计所述目标对象中的行人数量和车辆数量。
  20. 根据权利要求16所述的计算设备,其中,所述根据所述行人数量和车辆数量依据预设的静态时间分配机制或视觉动态控制机制控制所述交通信号灯提示颜色,包括:
    根据所述行人数量和车辆数量判断所述交通道路是否同时具有行人和车辆;
    若是,则根据预设的静态时间分配机制顺序切换所述交通信号灯的提示颜色;
    若否,则按照视觉动态控制机制控制所述交通信号灯提示颜色;其中,在所述交通道路只具有行人时,控制所述交通信号灯在行人侧的提示颜色为绿色;或,在所述交通道路只具有车辆时,控制所述交通信号灯在行人侧的提示颜色为红色。
PCT/CN2020/136057 2020-10-28 2020-12-14 交通信号灯的智能控制方法及系统、存储介质、计算设备 Ceased WO2021189970A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011171429.9 2020-10-28
CN202011171429.9A CN112309146A (zh) 2020-10-28 2020-10-28 交通信号灯的智能控制方法及系统、存储介质、计算设备

Publications (1)

Publication Number Publication Date
WO2021189970A1 true WO2021189970A1 (zh) 2021-09-30

Family

ID=74331233

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/136057 Ceased WO2021189970A1 (zh) 2020-10-28 2020-12-14 交通信号灯的智能控制方法及系统、存储介质、计算设备

Country Status (2)

Country Link
CN (1) CN112309146A (zh)
WO (1) WO2021189970A1 (zh)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114155723A (zh) * 2021-11-29 2022-03-08 中国联合网络通信集团有限公司 一种红绿灯的转换方法及装置
CN114842655A (zh) * 2022-04-29 2022-08-02 杭州海康威视数字技术股份有限公司 一种交通信号控制方法、装置及设备
CN115593310A (zh) * 2022-09-29 2023-01-13 中国第一汽车股份有限公司(Cn) 车辆的信息显示方法、装置和车辆
CN116504082A (zh) * 2023-02-22 2023-07-28 每日互动股份有限公司 一种信号灯的控制方法、装置、设备及介质
CN116580547A (zh) * 2022-07-25 2023-08-11 郑州迈拓信息技术有限公司 一种地埋式交通信号灯智能控制方法及系统
CN116631213A (zh) * 2023-06-16 2023-08-22 安徽科力信息产业有限责任公司 一种信号机非法模块检测方法、装置及可读储存介质

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113257008B (zh) * 2021-05-12 2022-06-21 兰州交通大学 基于深度学习的行人流量动态控制系统及方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106846837A (zh) * 2017-03-27 2017-06-13 广州大学 一种交通灯智能控制系统、交通灯智能控制方法及装置
CN110992680A (zh) * 2019-12-31 2020-04-10 宁波宁工交通工程设计咨询有限公司 一种交通信号灯智能控制方法、存储介质及交通信号灯
CN111599187A (zh) * 2020-05-23 2020-08-28 广州市翔通交通设施工程有限公司 一种交通信号灯控制方法、设备及其存储介质

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105160924B (zh) * 2015-08-25 2017-10-17 公安部第三研究所 基于视频处理的智能信号灯状态检测方法及检测系统
CN106228817B (zh) * 2016-08-09 2019-03-26 浙江大华技术股份有限公司 一种控制交通信号灯的方法和装置
CN206363539U (zh) * 2016-12-26 2017-07-28 上海澳星照明电器制造有限公司 交通信号灯故障预警系统
CN106846833B (zh) * 2017-04-12 2019-11-05 辛国臣 交通信号灯控制方法及系统
CN108447279A (zh) * 2018-05-18 2018-08-24 上海会为智能技术有限公司 一种信号灯管理方法、装置、设备和介质
CN108564799A (zh) * 2018-06-06 2018-09-21 苏州思在科技有限公司 一种网络信号灯
CN208689753U (zh) * 2018-10-11 2019-04-02 赵佳锦 优化型智能交通信号灯调控系统
CN110060495A (zh) * 2019-05-27 2019-07-26 无锡安邦电气股份有限公司 交通信号灯状态监测系统和方法

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106846837A (zh) * 2017-03-27 2017-06-13 广州大学 一种交通灯智能控制系统、交通灯智能控制方法及装置
CN110992680A (zh) * 2019-12-31 2020-04-10 宁波宁工交通工程设计咨询有限公司 一种交通信号灯智能控制方法、存储介质及交通信号灯
CN111599187A (zh) * 2020-05-23 2020-08-28 广州市翔通交通设施工程有限公司 一种交通信号灯控制方法、设备及其存储介质

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114155723A (zh) * 2021-11-29 2022-03-08 中国联合网络通信集团有限公司 一种红绿灯的转换方法及装置
CN114155723B (zh) * 2021-11-29 2023-04-14 中国联合网络通信集团有限公司 一种红绿灯的转换方法及装置
CN114842655A (zh) * 2022-04-29 2022-08-02 杭州海康威视数字技术股份有限公司 一种交通信号控制方法、装置及设备
CN114842655B (zh) * 2022-04-29 2023-05-26 杭州海康威视数字技术股份有限公司 一种交通信号控制方法、装置及设备
CN116580547A (zh) * 2022-07-25 2023-08-11 郑州迈拓信息技术有限公司 一种地埋式交通信号灯智能控制方法及系统
CN116580547B (zh) * 2022-07-25 2024-05-03 郑州迈拓信息技术有限公司 一种地埋式交通信号灯智能控制方法及系统
CN115593310A (zh) * 2022-09-29 2023-01-13 中国第一汽车股份有限公司(Cn) 车辆的信息显示方法、装置和车辆
CN116504082A (zh) * 2023-02-22 2023-07-28 每日互动股份有限公司 一种信号灯的控制方法、装置、设备及介质
CN116631213A (zh) * 2023-06-16 2023-08-22 安徽科力信息产业有限责任公司 一种信号机非法模块检测方法、装置及可读储存介质

Also Published As

Publication number Publication date
CN112309146A (zh) 2021-02-02

Similar Documents

Publication Publication Date Title
WO2021189970A1 (zh) 交通信号灯的智能控制方法及系统、存储介质、计算设备
US11915593B2 (en) Systems and methods for machine learning based collision avoidance
CN109714730B (zh) 用于车车及车路协同的云控平台系统及协同系统和方法
JP4976409B2 (ja) センサ基盤ホームネットワークシステムのセンシング情報管理装置および方法
CN105794324B (zh) 具有基于来自相邻照明器和/或所连接设备的状态信息的自适应检测设置的传感器网络
EP2989865A2 (en) Adaptive outdoor lighting control system based on user behavior
CN108564786A (zh) 一种基于云计算的城市交通信号控制的方法和系统
CN105208132A (zh) 智能终端云管理系统
US12118882B2 (en) Method and Internet of Things system for tidal lane opening management in smart city
CN118429952B (zh) 用于交通管理的智能车牌识别平台
WO2016000325A1 (zh) 路灯控制方法及装置
CN105306560A (zh) 分布式终端实施动态管理平台
CN105227365A (zh) 基于安卓平台的物联网终端管控系统
US20250311075A1 (en) Detection and illumination of dark zones via collaborative lighting
CN110800324A (zh) 一种改善道路安全和/或管理的系统和方法
CN115905431A (zh) 数据处理方法及装置
US9336661B2 (en) Safety communication system and method thereof
CN117118906A (zh) 一种流量数据控制方法、装置、设备及介质
CN115762159A (zh) 路云协同下行人预测导向的自动驾驶路径规划系统
CN120075973A (zh) 基于大小模型融合的多智能体5g基站智能节能方法及装置
CN103607757A (zh) 车载终端实现接入网络选择的方法、系统及车载终端
CN118075299A (zh) 一种基于物联网的智慧城市系统
WO2024108810A1 (zh) 智能切片系统的数据参考集生成方法、装置、设备及介质
CN108510763A (zh) 一种交通信号灯控制方法及系统
CN116434573A (zh) 交通信号灯系统的控制方法、装置、计算机设备及介质

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: 20927194

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: 20927194

Country of ref document: EP

Kind code of ref document: A1