WO2019079965A1 - 一种公路遗撒物快速检测系统及方法 - Google Patents

一种公路遗撒物快速检测系统及方法

Info

Publication number
WO2019079965A1
WO2019079965A1 PCT/CN2017/107460 CN2017107460W WO2019079965A1 WO 2019079965 A1 WO2019079965 A1 WO 2019079965A1 CN 2017107460 W CN2017107460 W CN 2017107460W WO 2019079965 A1 WO2019079965 A1 WO 2019079965A1
Authority
WO
WIPO (PCT)
Prior art keywords
suspected
data packet
information
monitoring
confirmation
Prior art date
Application number
PCT/CN2017/107460
Other languages
English (en)
French (fr)
Inventor
费伦林
张炳琪
张一衡
徐立红
李俊卫
孙彪彪
Original Assignee
江西省高速公路投资集团有限责任公司
北京国交信通科技发展有限公司
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 江西省高速公路投资集团有限责任公司, 北京国交信通科技发展有限公司 filed Critical 江西省高速公路投资集团有限责任公司
Priority to PCT/CN2017/107460 priority Critical patent/WO2019079965A1/zh
Publication of WO2019079965A1 publication Critical patent/WO2019079965A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled

Definitions

  • the present application belongs to the field of highway detection, and in particular relates to a rapid detection system and method for road debris.
  • the Ministry of Transport and other ceremonies jointly issued a document requesting that the key operating vehicles must be equipped with a vehicle-mounted positioning terminal with satellite navigation function. It is required to install a video image monitoring terminal in the passenger vehicle, and send the location information data and part of the image data to the monitoring platform to enable monitoring.
  • the platform brings together a large amount of real-time location and image data.
  • the vehicle data of the access monitoring platform has reached more than 5 million units.
  • the vehicle terminal will further improve the positioning accuracy and image quality, and the coverage of the vehicle will be further expanded. It is expected that 2020 will reach more than 8 million units.
  • Road traffic is caused by the bumping, collision or packaging damage, the failure of the fixture, etc., causing the transported goods or vehicle parts to be scattered from the vehicle to the road surface.
  • the means of discovery of roads are very limited.
  • Manual observation or image analysis methods can be used for road sections with video surveillance. In the sections without video surveillance, only road inspection personnel can manually detect or report to and from personnel vehicles. At the end of 2016, China's highway mileage has reached nearly 4.7 million kilometers, and the expressway has reached 130,000 kilometers. Most of the road sections have no video surveillance. How to discover the traces on these roads has no effective means.
  • the application for solving the road residual monitoring on the road segment without video monitoring in the prior art has the problems of wasted manpower, difficult real-time monitoring and low monitoring efficiency.
  • a technical solution of the present application is to provide a road debris rapid detection system, comprising: a data receiving module, configured to receive and store vehicle monitoring information of a plurality of groups of operating vehicles, wherein the operating vehicle monitoring information includes a vehicle speed , monitoring time, monitoring location and monitoring images;
  • the data preprocessing module is configured to process the monitoring image in the monitoring information of each operating vehicle, analyze and obtain the suspected survivor information, and generate the suspected residual data packet according to the suspected survivor information and the corresponding operating vehicle monitoring information;
  • the suspected residual analysis module is configured to store the suspected lost data packet; for each suspected legacy data packet, extract the suspected number of the first predetermined space and time according to the monitoring time and the monitoring position in the suspected legacy data packet. According to the package; comparing the suspected legacy data packet with the extracted suspected data packet, calculating the suspected probability of the suspected legacy data packet according to the comparison result; pushing the suspected legacy data packet whose suspected probability is greater than a predetermined threshold value to the legacy information confirmation module;
  • the residual confirmation module is configured to receive the suspected lost data packet pushed by the suspected residual analysis module, send the suspected lost data packet to the corresponding operating vehicle for confirmation or hand-in-delivery, and if confirmed, generate a residual according to the suspected legacy data packet in a predetermined format. Information report form.
  • a method for quickly detecting a road remnant including:
  • the operating vehicle monitoring information includes vehicle speed, monitoring time, monitoring position, and monitoring image
  • the monitoring images in the monitoring information of each operating vehicle are processed, the suspected survivor information is analyzed, and the suspected residual data packet is generated according to the suspected survivor information and the corresponding operating vehicle monitoring information;
  • each suspected legacy data packet For each suspected legacy data packet, according to the monitoring time and the monitoring location in the suspected legacy data packet, extract the suspected legacy data packet that satisfies the first predetermined space and time; compare the suspected legacy data packet with the extracted suspected legacy data packet. Calculating the suspected probability of the suspected lost data packet based on the comparison result;
  • the suspected legacy data packet whose suspect probability is greater than the predetermined threshold is sent to the corresponding operational vehicle for confirmation or manual confirmation. If the confirmation is passed, the residual information report is generated according to the suspected legacy data packet in a predetermined format.
  • a computer device including a memory, a processor, and a computer program stored on the memory and operable on the processor, and the processor executes the computer program to implement the road described in the foregoing embodiment. Rapid detection method for lost materials.
  • a computer readable storage medium storing a computer program for performing the method for quickly detecting a road remnant according to the above embodiments is further provided.
  • the present invention can obtain the suspected survivor information, generate the suspected residual data packet according to the suspected survivor information and the operating vehicle monitoring information, calculate the suspected probability of the suspected lost data packet, and filter out the suspected probability.
  • the legacy data packet larger than the predetermined threshold is sent to the operating vehicle for confirmation or manual confirmation. If the confirmation is passed, the residual information report is generated according to the suspected legacy data packet in a predetermined format.
  • FIG. 1 is a structural diagram of a rapid detection system for road debris according to an embodiment of the present application
  • FIG. 2 is a structural diagram of a road debris rapid detection system according to another embodiment of the present application.
  • FIG. 3 is a structural diagram of a method for quickly detecting a road remnant according to an embodiment of the present application.
  • an embodiment means that the specific features, structures, materials, or characteristics described in connection with the embodiments or examples are included in the present application. At least one embodiment or example.
  • the schematic representation of the above terms does not necessarily mean the same embodiment or example.
  • the particular features, structures, materials, or characteristics described may be combined in a suitable manner in any one or more embodiments or examples.
  • the order of the steps involved in the embodiments is used to schematically illustrate the implementation of the present application, and the order of the steps is not limited, and may be appropriately adjusted as needed.
  • a video image monitoring terminal and an in-vehicle monitoring terminal having a navigation function are installed in the operating vehicle described in the present application.
  • FIG. 1 is a structural diagram of a rapid detection system for highway debris according to an embodiment of the present application.
  • the system can shorten the discovery time, reduce the frequency of manual inspection, and improve the safety degree and traffic efficiency of the road.
  • the rapid detection system for road debris includes:
  • the data receiving module 101 is configured to receive and store a plurality of sets of operating vehicle monitoring information, and push the operating vehicle monitoring information to the data preprocessing module 102, wherein the operating vehicle monitoring information includes a vehicle speed, a monitoring time, a monitoring location, and a monitoring image.
  • the operational vehicle monitoring information is sent by the ministerial and provincial vehicle monitoring platforms, and the data receiving module stores the operating vehicle monitoring data for a certain period of time, such as half an hour, for subsequent data analysis.
  • the data pre-processing module 102 is configured to receive the monitoring information of the operating vehicle pushed by the data receiving module 101, process the monitoring image in the monitoring information of each operating vehicle, and roughly determine whether there is a suspected relic, and find the monitoring with the suspected surviving object.
  • Image analyzing the surveillance image with suspected remains to obtain suspected survivor information; generating suspected residual data packets based on suspected survivor information and corresponding operational vehicle monitoring information, suspected residual data packets including suspected residual information and corresponding operational vehicle monitoring information;
  • the suspected legacy packet is pushed to the suspected analysis module 103.
  • the suspected residual analysis module 103 is configured to store the suspected legacy data packet in the database; for each suspected legacy data packet, extract the first predetermined space and time in the database according to the monitoring time and the monitoring location in the suspected legacy data packet. (such as a radius of 50m, 10 minutes before and after the time) suspected data packet; compare the suspected data packet with the extracted suspected data packet, calculate the suspected probability of the suspected legacy data packet based on the comparison result (ie suspected The data packet indicates the probability of the real object.
  • the suspected legacy data packet with the suspected probability greater than the predetermined threshold is pushed to the legacy confirmation module 104, and the suspected legacy data packet whose suspect probability is less than or equal to the predetermined threshold is stored in the database for subsequent analysis. .
  • the confirmation module 104 sends the suspected legacy data packet sent by the suspected residual analysis module 103 to the corresponding operational vehicle for confirmation or manual confirmation. If the confirmation is passed, the suspected residual data packet is pressed according to the suspected legacy data packet.
  • the predetermined format generates a report information report, and the report information report includes all the contents of the legacy data package.
  • the corresponding operating vehicle refers to an operating vehicle with a camera or a call function in the area where the object is to be passed, and the operating vehicle responds to the confirmation result by collecting images or voices.
  • the legacy confirmation module 104 can transmit the received suspected data packet to the operational vehicle monitoring platform, and the operational vehicle monitoring platform sends the confirmation information to the operating vehicle based on the suspected legacy data packet.
  • the road debris rapid detection system further includes: an information release module 105, configured to issue a report information report, receive the information about the disposal of the object returned by the maintenance unit, and decide whether to withdraw the residue according to the disposal situation of the object.
  • Information report form configured to issue a report information report, receive the information about the disposal of the object returned by the maintenance unit, and decide whether to withdraw the residue according to the disposal situation of the object. Information report form.
  • the information release module can selectively publish the information report to the road management and maintenance unit according to the pre-set authority, and the road management and maintenance unit can quickly locate the exact location of the suspected object according to the information report.
  • the information release module may also send the report of the lost information to the operating vehicle or individual who is going to pass the road where the object is located to remind the corresponding operating vehicle to pay attention.
  • the information release module can also publish the information report to the navigation software to remind the road surface of the situation.
  • the present application utilizes the existing operational vehicle monitoring information to detect whether there are any survivors on the road in a timely, accurate and low-cost manner, which can reduce the risk of traffic accidents and reduce the impact on traffic efficiency.
  • the data receiving module 101 is further configured to perform cleaning processing on the operating vehicle monitoring information according to a predetermined rule, such as automatically integrating monitoring data of the same operating vehicle from different vehicle monitoring platforms, so as to improve the subsequent data preprocessing module. Analysis speed.
  • the suspected survivor information includes a suspected type of suspense, an area distribution of the suspected surviving object, and gray scale information of the suspected surviving object.
  • the data pre-processing module processes the monitoring image in the operating vehicle monitoring information according to the receiving sequence of the operating vehicle monitoring information.
  • the processing and analysis process of the monitoring image in each operating vehicle monitoring information includes :
  • Step 301 extracting a road surface image of the monitoring image according to the grayscale contrast.
  • Step 302 removing lane lines in the road surface image according to the continuous rectangular feature of the lane line.
  • Step 303 Extract a road surface image that satisfies a second predetermined space and time (for example, a space of 50 meters, a time within the first 10 minutes) according to the monitoring time and the monitoring position in the monitoring information of the operating vehicle, according to the extracted road surface.
  • the image calculates the road surface gradation of the monitored image.
  • the average gradation of the extracted road surface image is calculated, the average gradation is used as the road surface gradation of the monitoring image. This step can accurately determine the gray scale of the road surface of the monitored image.
  • Step 304 Determine a block-shaped area in the road surface image after the lane line is removed by comparing with the gray scale of the road surface of the monitoring image. If there is no block area, it indicates that no-spray occurs. If there is a block-shaped area, perform the following step 305. .
  • step 305 the area distribution and gray scale information of the suspected object are obtained according to the block region statistics.
  • Step 306 matching the area distribution and the gray scale information of the suspected residual object with the real residual area distribution and the gray scale information in the model of the surviving object, and determining the suspected type of the missing.
  • the model library of prey is a pre-established model for image analysis of different types of remains (such as soil remains, small stones, weaving, and tires). Degree information representation.
  • the suspected legacy data package includes suspected survivor information, corresponding operational vehicle monitoring information, and road surface images of corresponding monitored images.
  • the process of obtaining the area distribution and gray scale information of the suspected object based on the block region statistics in the above step 305 includes:
  • Step 401 determining the area of the block area according to the lane line width.
  • the standard width of the lane line is approximately 15 cm, and the area of the block area is roughly determined by comparing the block area with the lane line width.
  • step 402 the block regions in the image are grouped according to a predetermined area group, and the number of block regions under each group is counted.
  • the predetermined area grouping can be divided by an equal division method, such as 0 to 0.1 square meters, 0.1 to 0.2 square meters.
  • the predetermined area grouping can also be divided into non-uniform (depending on the size of the real thing), such as 0-0.3 square meters, 0.3-0.8 square meters, 0.8-1.2 square meters, 1.2 square meters or more.
  • Table 1 the statistical results of this step are shown in Table 1:
  • Step 403 Calculate the area ratio of the block-shaped area under the corresponding group according to the number of the block-shaped areas in each group and the total number of the block-shaped areas in the image, and combine the area ratio of the block-shaped area under each group to obtain the area distribution of the suspected objects.
  • the ratio of the area of the block corresponding to 2 and >8m 2 is 0, and the area distribution of the suspected remains is shown in Table 2:
  • Step 404 Calculate the gradation information of the block region under the corresponding group according to the gradation and the number of the block regions under each group, and combine the gradation information of the block region under each group to obtain the gradation information of the suspected object.
  • the gray scale information of the block region under each group is calculated by the following formula:
  • G(i) is the gradation information of the block region of the i-th group
  • n i is the number of block regions of the i-th group
  • a ij is the area of the j-th block-shaped region of the i-th group.
  • the suspected legacy analysis module 103 compares the suspected legacy data packet with the extracted suspected legacy data packet, and calculates a suspected probability of the suspected legacy data packet according to the comparison result. :
  • Step 501 comparing the suspected legacy data packet with the extracted suspected data packet from four dimensions: a suspected type of the suspect, an area distribution of the suspected object, a grayscale information of the suspected object, and a vehicle speed, to obtain a comparison of the dimensions.
  • the comparison result is expressed by the degree of consistency (similarity probability).
  • the suspected legacy packet A For example, suppose that for the suspected legacy packet A, a total of five suspected legacy packets (ie, five related suspected legacy packets) are extracted. By comparison, the suspected scatter type in the suspected scatter packet A is consistent with the suspected severance type in the five related suspected scatter packets, and the comparison result corresponding to the suspected scatter type is 1. Suspected scattered data packet A in the area of suspected remains and four suspected legacy in the suspected legacy data packet The area distribution is consistent, and the comparison result of the area distribution of the suspected remains is 0.8. The gray scale information of the suspected survivor in the suspected scattered data packet A is consistent with the gray scale information of the suspected survivor in the two related suspected scattered data packets, and the comparison result corresponding to the gray matter information of the suspected survivor is 0.4. If the speed of the vehicle in the suspected scattered data packet A coincides with the speed of the vehicle in the three related suspected scattered data packets, the comparison result of the vehicle speed is 0.6.
  • Step 502 Perform weighted summation calculation on each comparison result to obtain a suspected probability of the suspected legacy data packet.
  • the weight of each dimension comparison result can be set according to requirements, and the specific value of the application is not limited. Generally, it can be determined in two ways: one is based on the accuracy of association between different information and the remaining; the second is determined according to the degree of consistency. The larger the number of samples, the higher the consistency and the greater the weight.
  • the first-level weight value when the daytime lighting condition is good, may be: the suspected surviving type corresponds to the weight 1, the area distribution corresponds to the weight 0.7, the gray information corresponds to the weight 0.9, and the vehicle speed corresponds to the weight. 1.2.
  • the secondary weight can be set to the square of the comparison result multiplied by the number of samples.
  • the suspected residual analysis packet is filtered out by the suspected residual analysis module, and the suspected legacy data packet with the suspected probability is pushed to the legacy confirmation module, which can improve the processing efficiency of the confirmation module, avoid unnecessary confirmation, and improve the user experience.
  • the process of the confirmation confirmation module 104 transmitting the suspected lost data packet to the corresponding operational vehicle for confirmation or manual confirmation includes:
  • the present invention can obtain the suspected survivor information, generate the suspected residual data packet according to the suspected survivor information and the operating vehicle monitoring information, calculate the suspected probability of the suspected lost data packet, and filter out the suspected probability.
  • the discarded data packet larger than the predetermined threshold is sent to the operation vehicle for confirmation or manual confirmation. If the confirmation is passed, the residual information report form is generated according to the suspected legacy data packet according to the predetermined format, which can shorten the road discovery time and reduce the manual inspection frequency. To improve the safety and traffic efficiency of the road.
  • the present application also provides a method for quickly detecting road debris, as described in the following embodiments. Since the method for solving the problem is similar to the system, the implementation of the method can be referred to the implementation of the system, and the repeated description is not repeated.
  • the rapid detection method of highway remnants includes:
  • Step 601 Receive multiple sets of operating vehicle monitoring information, where the operating vehicle monitoring information includes vehicle speed, monitoring time, monitoring location, and monitoring image;
  • Step 602 processing the monitoring image in the monitoring information of each operating vehicle, analyzing and obtaining the suspected survivor information, and generating the suspected residual data packet according to the suspected survivor information and the corresponding operating vehicle monitoring information;
  • Step 603 For each suspected legacy data packet, extract a suspected legacy data packet that satisfies the first predetermined space and time according to the monitoring time and the monitoring location in the suspected legacy data packet; compare the suspected legacy data packet with the extracted suspected data packet. The data packet is discarded, and the suspected probability of the suspected lost data packet is calculated according to the comparison result;
  • Step 604 Send the suspected legacy data packet whose suspect probability is greater than the predetermined threshold to the corresponding operational vehicle for confirmation or hand-in-confirm, and if the confirmation is passed, generate the residual information report according to the suspected legacy data packet in a predetermined format.
  • the present application further provides a computer device, including a memory, a processor, and a computer program stored on the memory and operable on the processor, and the method for quickly detecting the road remnant described in the foregoing embodiment when the processor executes the computer program .
  • the present application also provides a computer readable storage medium storing a computer program for performing the method for quickly detecting a road debris according to the above embodiments.
  • This application may take the form of an entirely hardware embodiment, an entirely software embodiment, or a combination of software and hardware.
  • the application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) including computer usable program code.
  • the computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device.
  • the apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.

Abstract

本申请提供了一种公路遗撒物快速检测系统及方法，系统包括：数据接收模块，用于接收多组运营车辆监测信息；数据预处理模块，用于对各运营车辆监测信息中的监测图像进行处理，分析得到疑似遗撒物信息，生成疑似遗撒数据包；疑似遗撒分析模块，用于分析每一疑似遗撒数据包，根据该疑似遗撒数据包，提取满足第一预定空间及时间的疑似遗撒数据包，比较该疑似遗撒数据包与提取出的疑似遗撒数据包，根据比较结果计算该疑似遗撒数据包的疑似概率；遗撒确认模块，用于将疑似概率大于预定阈值的疑似遗撒数据包发送给相应运营车辆确认或交由人工确认，生成遗撒信息报告单。本申请可缩短公路遗撒发现时间，降低人工巡检频度，提高公路的安全程度和通行效率。

Description

一种公路遗撒物快速检测系统及方法 技术领域
本申请属于公路检测领域，尤其涉及一种公路遗撒物快速检测系统及方法。
背景技术
目前交通运输部等部委联合发文要求重点营运车辆必须配备具有卫星导航功能的车载定位终端，要求在客运车辆中安装视频图像监控终端，并将位置信息数据和部分图像数据发送至监控平台，使得监控平台汇集了大量的实时位置和图像数据。目前接入监控平台的车辆数据已经达到500万台以上。未来几年车载终端将进一步提升定位精度和图像质量，覆盖车辆范围也将进一步扩大，预计2020将达到800万台以上。
公路遗撒是行驶在公路上的车辆由于颠簸、碰撞或包装损坏、固定物失效等原因导致运输的货物或车辆零件从车辆上散落到公路路面。较大或尖锐的遗撒物对来往车辆正常行驶构成了严重威胁，尤其是在速度较快的高速公路上，影响公路通行效率，且会威胁车辆行驶安全。目前公路遗撒的发现手段非常有限，在有视频监测的路段可采用人工观察或图像分析方法监测，在没有视频监测的路段只能采用道路巡检人员人工发现或来往人员车辆报告的方式。2016年底我国公路通车里程已经接近470万公里，高速公路达到13万公里，其中绝大部分路段没有视频监控，这些公路上的遗撒如何及时发现目前尚没有有效手段。
发明内容
本申请用于解决现有技术中没有视频监测的路段上的公路遗撒监测存在浪费人力、不易实时监测及监测效率低的问题。
为了解决上述问题，本申请的一技术方案为提供一种公路遗撒物快速检测系统，包括：数据接收模块，用于接收并存储多组运营车辆的车辆监测信息，其中，运营车辆监测信息包括车速、监测时间、监测位置和监测图像；
数据预处理模块，用于对各运营车辆监测信息中的监测图像进行处理，分析得到疑似遗撒物信息，根据疑似遗撒物信息和相应运营车辆监测信息生成疑似遗撒数据包；
疑似遗撒分析模块，用于存储疑似遗撒数据包；对于每一疑似遗撒数据包，根据该疑似遗撒数据包中的监测时间及监测位置，提取满足第一预定空间及时间的疑似遗撒数 据包；比较该疑似遗撒数据包与提取出的疑似遗撒数据包，根据比较结果计算该疑似遗撒数据包的疑似概率；将疑似概率大于预定阈值的疑似遗撒数据包推送至遗撒信息确认模块；
遗撒确认模块，用于接收疑似遗撒分析模块推送的疑似遗撒数据包，将疑似遗撒数据包发送给相应运营车辆确认或交由人工确认，若确认通过，则根据疑似遗撒数据包按预定格式生成遗撒信息报告单。
本申请另一技术方案中，还提供一种公路遗散物快速检测方法，包括：
接收多组运营车辆监测信息，其中，运营车辆监测信息包括车速、监测时间、监测位置和监测图像；
对各运营车辆监测信息中的监测图像进行处理，分析得到疑似遗撒物信息，根据疑似遗撒物信息和相应运营车辆监测信息生成疑似遗撒数据包；
对于每一疑似遗撒数据包，根据该疑似遗撒数据包中的监测时间及监测位置，提取满足第一预定空间及时间的疑似遗撒数据包；比较该疑似遗撒数据包与提取出的疑似遗撒数据包，根据比较结果计算该疑似遗撒数据包的疑似概率；
将疑似概率大于预定阈值的疑似遗撒数据包发送给相应运营车辆确认或交由人工确认，若确认通过，则根据疑似遗撒数据包按预定格式生成遗撒信息报告单。
本申请再一技术方案中，还提供一种计算机设备，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，处理器执行计算机程序时实现上述实施例所述的公路遗散物快速检测方法。
本申请又一技术方案中，还提供一种计算机可读存储介质，计算机可读存储介质存储有执行上述实施例所述的公路遗散物快速检测方法的计算机程序。
本申请通过对运营车辆监测信息中的监测图像进行分析能够得到疑似遗撒物信息，根据疑似遗撒物信息及运营车辆监测信息生成疑似遗撒数据包，计算疑似遗撒数据包的疑似概率，筛选出疑似概率大于预定阈值的遗撒数据包发送给运营车辆确认或交由人工确认，若确认通过，则根据疑似遗撒数据包按预定格式生成遗撒信息报告单。本申请可以缩短公路遗撒发现时间，降低人工巡检频度，从而提高公路的安全程度和通行效率。
附图说明
为了更清楚地说明本申请实施例的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于 本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为本申请一实施例的公路遗撒物快速检测系统的结构图；
图2为本申请另一实施例的公路遗撒物快速检测系统的结构图；
图3为本申请一实施例的公路遗散物快速检测方法的结构图。
具体实施方式
为了使本申请的技术特点及效果更加明显，下面结合附图对本申请的技术方案做进一步说明，本申请也可有其他不同的具体实例来加以说明或实施，任何本领域技术人员在权利要求范围内做的等同变换均属于本申请的保护范畴。
在本说明书的描述中，参考术语“一实施例”、“一具体实施例”、“例如”等的描述意指结合该实施例或示例描述的具体特征、结构、材料或者特点包含于本申请的至少一个实施例或示例中。在本说明书中，对上述术语的示意性表述不一定指的是相同的实施例或示例。而且，描述的具体特征、结构、材料或者特点可以在任何的一个或多个实施例或示例中以合适的方式结合。各实施例中涉及的步骤顺序用于示意性说明本申请的实施，其中的步骤顺序不作限定，可根据需要作适当调整。
本申请所述的运营车辆中安装有视频图像监控终端及具有导航功能的车载监控终端。
如图1所示，图1为本申请实施例的公路遗撒物快速检测系统的结构图，通过本系统能够缩短遗撒发现时间，降低人工巡检频度，提高公路的安全程度和通行效率。具体的，公路遗撒物快速检测系统包括：
数据接收模块101，用于接收并存储多组运营车辆监测信息，将运营车辆监测信息推送到数据预处理模块102，其中，运营车辆监测信息包括车速、监测时间、监测位置和监测图像。实施时，运营车辆监控信息由部级和省级车辆监控平台发送，数据接收模块储存一定时间的运营车辆监测数据，如半小时，以便后续数据分析。
数据预处理模块102，用于接收数据接收模块101推送的运营车辆监测信息，对各运营车辆监测信息中的监测图像进行处理，粗判断是否有疑似遗散物，找出具有疑似遗撒物的监测图像，分析具有疑似遗撒物的监测图像得到疑似遗撒物信息；根据疑似遗撒物信息和相应运营车辆监测信息生成疑似遗撒数据包，疑似遗撒数据包包括疑似遗撒物信息和相应运营车辆监测信息；将疑似遗撒数据包推送至疑似遗撒分析模块103。
疑似遗撒分析模块103，用于在数据库中存储疑似遗撒数据包；对于每一疑似遗撒数据包，根据该疑似遗撒数据包中的监测时间及监测位置，在数据库中提取满足第一预定空间及时间(如半径50m的空间，前后10分钟的时间)的疑似遗撒数据包；比较该疑似遗撒数据包与提取出的疑似遗撒数据包，根据比较结果计算该疑似遗撒数据包的疑似概率(即疑似遗撒数据包表示真实遗撒物的概率)，将疑似概率大于预定阈值的疑似遗撒数据包推送至遗撒确认模块104，对于疑似概率小于或等于预定阈值的疑似遗撒数据包存储至数据库中，以备后续分析。
遗撒确认模块104，用接收疑似遗撒分析模块103推送的疑似遗撒数据包，将接收到的疑似遗撒数据包发送给相应运营车辆确认或交由人工确认，若确认通过，则根据疑似遗撒数据包按预定格式生成遗撒信息报告单，遗撒信息报告单包括遗撒数据包的中全部内容。详细的说，相应运营车辆指的是即将经过遗撒物所在区域的具备摄像头或通话功能的运营车辆，运营车辆通过采集图像或语音的方式来回复确认结果。
实施时，遗撒确认模块104可将接收到的疑似遗撒数据包发送至运营车辆监控平台，由运营车辆监测平台根据疑似遗撒数据包发送确认信息至运营车辆。
进一步的，如图2所示，公路遗撒物快速检测系统还包括：信息发布模块105，用于发布遗撒信息报告单，接收养护单位反馈的遗撒物处理情况，根据遗撒物处理情况决定是否撤回遗撒信息报告单。
实施时，信息发布模块可根据事先设定的权限，有选择的将遗撒信息报告单发布给公路管理养护单位，公路管理养护单位根据遗撒信息报告单能够快速定位到疑似遗撒物的准确位置。信息发布模块还可将遗撒信息报告发送给即将经过遗撒物所在路段的运营车辆或个人，以提醒相应运营车辆注意。信息发布模块还可将遗撒信息报告发布至导航软件，以提醒路面遗撒情况。
本申请利用现有运营车辆监测信息，能够及时、准确、低成本的发现公路是否存在遗撒物，能够降低交通事故风险，降低对通行效率的影响。
本申请一实施例中，数据接收模块101还用于按预定规则对运营车辆监测信息进行清洗处理，如自动融合来自不同车辆监控平台的同一运营车辆的监测数据，以便提高后续数据预处理模块的分析速度。
本申请一实施例中，疑似遗撒物信息包括疑似遗撒类型、疑似遗撒物的面积分布及疑似遗撒物的灰度信息。
本申请一实施例中，数据预处理模块按照运营车辆监测信息的接收顺序处理运营车辆监测信息中的监测图像，具体的，对于每一运营车辆监测信息中的监测图像，其处理及分析过程包括：
步骤301，根据灰度对比度抽取该监测图像的道路路面图像。
步骤302，根据车道线的连续矩形特征去除道路路面图像中的车道线。
步骤303，根据该运营车辆监测信息中的监测时间及监测位置，提取满足第二预定空间及时间(如50米的空间，前10分钟内的时间)的道路路面图像，根据提取出的道路路面图像计算该监测图像的路面基础灰度。实施时，如计算提取出的道路路面图像的平均灰度，将该平均灰度作为该监测图像的路面基础灰度。本步骤能够精确确定监测图像的路面基础灰度。
步骤304，通过与该监测图像的路面基础灰度对比确定去除车道线后道路路面图像中的块状区域，若无块状区域，说明无遗撒发生，若有块状区域，则执行如下步骤305。
步骤305，根据块状区域统计得到疑似遗撒物的面积分布及灰度信息。
步骤306，将疑似遗撒物的面积分布及灰度信息分别与遗撒物模型库中真实遗撒物面积分布及灰度信息进行匹配，确定疑似遗撒类型。详细的说，遗撒物模型库是通过对不同类型遗撒(如土质遗撒、小石块遗撒、编织物遗撒、轮胎遗撒)进行图像分析预先建立的模型，遗撒物模型由遗撒物的面积分布及灰度信息表示。
一些实施方式中，为了便于人工确认疑似遗撒物，疑似遗撒数据包包括疑似遗撒物信息、相应运营车辆监测信息及相应监测图像的道路路面图像。
一些实施方式中，上述步骤305中根据块状区域统计得到疑似遗撒物的面积分布及灰度信息的过程包括：
步骤401，根据车道线宽度确定块状区域面积。车道线标准宽度约为15cm，通过将块状区域与车道线宽度进行对比粗略确定块状区域面积。
步骤402，按预定面积分组对图像中的块状区域进行分组，统计各分组下块状区域的数量。
实施时，预定面积分组可采用等分法划分，如0～0.1平米，0.1～0.2平米…。预定面积分组还可采用非均匀划分(可根据真实遗撒物大小而定)，如0～0.3平米，0.3～0.8平米，0.8～1.2平米，1.2平米以上等。举例来说，本步骤的统计结果如表一：
表一：
分组(m2) A<＝0.5 0.5<A<＝1 1<A<＝8 >8
数量(个) 40 10 0 0
步骤403，根据各分组下块状区域的数量及图像中块状区域总数量计算相应分组下块状区域面积比例，组合各分组下块状区域面积比例得到疑似遗撒物的面积分布。对于步骤402的举例，A<＝0.5m2对应的区块面积比例为4/5，0.5m2<A<＝1m2对应的区块面积比例为1/5，1m2<A<＝8m2及>8m2对应的区块面积比例为0，疑似遗撒物面积分布如表二：
表二：
分组(m2) A<＝0.5 0.5<A<＝1 1<A<＝8 >8
数量(个) 40 10 0 0
比例 4/5 1/5 0 0
步骤404，根据各分组下块状区域的灰度及数量计算相应分组下块状区域的灰度信息，组合各分组下块状区域的灰度信息得到疑似遗撒物的灰度信息。具体实施时，通过如下公式计算各分组下块状区域的灰度信息：
Figure PCTCN2017107460-appb-000001
其中，G(i)为第i组的区块区域的灰度信息，ni为第i组的块状区域数量，Aij为第i组的第j个块状区域的面积。
本申请一实施例中，对于每一疑似遗撒数据包，疑似遗撒分析模块103比较该疑似遗撒数据包与提取出的疑似遗撒数据包，根据比较结果计算该疑似遗撒数据包的疑似概率的过程包括：
步骤501，将该疑似遗撒数据包与提取出的疑似遗撒数据包从疑似遗撒类型、疑似遗撒物的面积分布、疑似遗撒物的灰度信息、车辆速度四个维度进行比较，得到各维度的比较结果，比较结果用一致性程度(相似度概率)来表示。
举例而言，假设对于疑似遗撒数据包A而言，共提取出5个疑似遗撒数据包(即5个相关疑似遗撒数据包)。通过比较，疑似遗散数据包A中的疑似遗散类型与5个相关疑似遗散数据包中的疑似遗散类型都一致，则疑似遗散类型对应的比较结果为1。疑似遗散数据包A中的疑似遗撒物面积分布与其中4个相关疑似遗撒数据包中的疑似遗撒物 面积分布一致，则疑似遗撒物面积分布对应的比较结果为0.8。疑似遗散数据包A中的疑似遗撒物灰度信息与其中2个相关疑似遗散数据包中的疑似遗撒物灰度信息一致，则疑似遗撒物灰度信息对应的比较结果为0.4。疑似遗散数据包A中的车辆速度与其中3个相关疑似遗散数据包中的车辆速度一致，则车辆速度对应的比较结果为0.6。
步骤502，对各比较结果进行加权求和计算，得到该疑似遗撒数据包的疑似概率。
详细的说，各维度比较结果的权值可根据需求进行设定，本申请对其具体取值不做限定。通常情况下，可通过如下两种方式确定：一是依据不同信息与遗撒的关联准确度来确定；二是依据一致性程度来确定，样本数量越大，一致性越高，权值越大。
一些具体实施方式中，在白天光照条件较好时，一级权值可以为：疑似遗撒类型对应权值1，面积分布对应权值0.7，灰度信息对应权值0.9，车辆速度对应权值为1.2。二级权值可以设定为比较结果的平方乘以样本数量。
通过疑似遗撒分析模块滤除疑似概率小的疑似遗撒数据包，将疑似概率大的疑似遗撒数据包推送至遗撒确认模块，能够提高遗撒确认模块的处理效率，避免不必要的确认，提高用户体验。
本申请一实施例中，为了减少人力，提高遗撒确认效率，遗撒确认模块104将疑似遗撒数据包发送给相应运营车辆确认或交由人工确认的过程包括：
根据疑似遗撒数据包中的监测时间及监测位置判断预定时间内是否有运营车辆即将经过该监测位置；若有，则将疑似遗撒数据包发送至运营车辆确认，若无，则将疑似遗撒数据包交由人工确认。
本申请通过对运营车辆监测信息中的监测图像进行分析能够得到疑似遗撒物信息，根据疑似遗撒物信息及运营车辆监测信息生成疑似遗撒数据包，计算疑似遗撒数据包的疑似概率，筛选出疑似概率大于预定阈值的遗撒数据包发送给运营车辆确认或交由人工确认，若确认通过，则根据疑似遗撒数据包按预定格式生成遗撒信息报告单，能够缩短公路遗撒发现时间，降低人工巡检频度，从而提高公路的安全程度和通行效率。
基于同一发明构思，本申请还提供一种公路遗散物快速检测方法，如下面的实施例所述。由于该方法解决问题的原理与系统相似，因此该方法的实施可以参见系统的实施，重复之处不再赘述。
如图3所示，公路遗散物快速检测方法包括：
步骤601，接收多组运营车辆监测信息，其中，运营车辆监测信息包括车速、监测时间、监测位置和监测图像；
步骤602，对各运营车辆监测信息中的监测图像进行处理，分析得到疑似遗撒物信息，根据疑似遗撒物信息和相应运营车辆监测信息生成疑似遗撒数据包；
步骤603，对于每一疑似遗撒数据包，根据该疑似遗撒数据包中的监测时间及监测位置，提取满足第一预定空间及时间的疑似遗撒数据包；比较该疑似遗撒数据包与提取出的疑似遗撒数据包，根据比较结果计算该疑似遗撒数据包的疑似概率；
步骤604，将疑似概率大于预定阈值的疑似遗撒数据包发送给相应运营车辆确认或交由人工确认，若确认通过，则根据疑似遗撒数据包按预定格式生成遗撒信息报告单。
本申请还提供一种计算机设备，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，处理器执行计算机程序时实现上述实施例所述的公路遗散物快速检测方法。
本申请还提供一种计算机可读存储介质，计算机可读存储介质存储有执行上述实施例所述的公路遗散物快速检测方法的计算机程序。
本申请可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且，本申请可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本申请是参照根据本申请实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或 其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
以上所述仅用于说明本申请的技术方案，任何本领域普通技术人员均可在不违背本申请的精神及范畴下，对上述实施例进行修饰与改变。因此，本申请的权利保护范围应视权利要求范围为准。

Claims (12)

  1. 一种公路遗撒物快速检测系统，其特征在于，包括：
    数据接收模块，用于接收多组运营车辆监测信息，其中，运营车辆监测信息包括车速、监测时间、监测位置和监测图像；
    数据预处理模块，用于对各运营车辆监测信息中的监测图像进行处理，分析得到疑似遗撒物信息，根据疑似遗撒物信息和相应运营车辆监测信息生成疑似遗撒数据包；
    疑似遗撒分析模块，用于存储疑似遗撒数据包；对于每一疑似遗撒数据包，根据该疑似遗撒数据包中的监测时间及监测位置，提取满足第一预定空间及时间的疑似遗撒数据包；比较该疑似遗撒数据包与提取出的疑似遗撒数据包，根据比较结果计算该疑似遗撒数据包的疑似概率；将疑似概率大于预定阈值的疑似遗撒数据包推送至遗撒信息确认模块；
    遗撒确认模块，用于将接收到的疑似遗撒数据包发送给相应运营车辆确认或交由人工确认，若确认通过，则根据疑似遗撒数据包按预定格式生成遗撒信息报告单。
  2. 如权利要求1所述的系统，其特征在于，还包括信息发布模块，用于发布遗撒信息报告单，接收养护单位反馈的遗撒物处理情况，根据遗撒物处理情况决定是否撤回遗撒信息报告单。
  3. 如权利要求1所述的系统，其特征在于，运营车辆监测信息由至少一个运营车辆监控平台发送。
  4. 如权利要求1所述的系统，其特征在于，数据接收模块还用于按预定规则对运营车辆监测信息进行清洗处理。
  5. 如权利要求1所述的系统，其特征在于，疑似遗撒物信息包括疑似遗撒类型、疑似遗撒物的面积分布及疑似遗撒物的灰度信息。
  6. 如权利要求5所述的系统，其特征在于，数据预处理模块对各运营车辆监测信息中的监测图像进行处理，分析得到疑似遗撒物信息的过程包括：
    对于每一运营车辆监测信息中的监测图像，根据灰度对比度抽取道路路面图像；
    根据车道线的连续矩形特征去除道路路面图像中的车道线；
    根据该运营车辆监测信息中的监测时间及监测位置，提取满足第二预定空间及时间的道路路面图像，根据提取出的道路路面图像计算该监测图像的路面基础灰度；
    根据该监测图像的路面基础灰度确定去除车道线后道路路面图像中的块状区域；
    根据块状区域统计得到疑似遗撒物的面积分布及灰度信息；
    将疑似遗撒物的面积分布及灰度信息分别与遗撒物模型库中真实遗撒物面积分布及灰度信息进行匹配，确定疑似遗撒类型。
  7. 如权利要求6所述的系统，其特征在于，根据疑似遗撒物信息和相应运营车辆监测信息生成疑似遗撒数据包进一步为：
    根据疑似遗撒物信息、相应运营车辆监测信息及相应监测图像的道路路面图像生成疑似遗撒数据包。
  8. 如权利要求6所述的系统，其特征在于，根据块状区域统计得到疑似遗撒物的面积分布及灰度信息进一步包括：
    根据车道线宽度确定块状区域面积；
    按预定面积分组对块状区域进行分组，统计各分组下块状区域的数量；
    根据各分组下块状区域的数量及块状区域总数量计算相应分组下块状区域面积比例，组合各分组下块状区域面积比例得到疑似遗散物的面积分布；
    根据各分组下块状区域的灰度及数量计算相应分组下块状区域的灰度信息，组合各分组下块状区域的灰度信息得到疑似遗撒物的灰度信息。
  9. 如权利要求5所述的系统，其特征在于，疑似遗撒分析模块比较该疑似遗撒数据包与提取出的疑似遗撒数据包，根据比较结果计算该疑似遗撒数据包的疑似概率的过程包括：
    将该疑似遗撒数据包与提取出的疑似遗撒数据包从疑似遗撒类型、疑似遗撒物的面积分布、疑似遗撒物的灰度信息、车辆速度四个维度进行比较，得到各维度的比较结果；
    对各比较结果进行加权求和计算，得到该疑似遗撒数据包的疑似概率。
  10. 如权利要求1所述的系统，其特征在于，遗撒确认模块将疑似概率大于预定阈值的疑似遗撒数据包发送给相应运营车辆确认或交由人工确认的过程包括：
    根据疑似遗撒数据包中的监测时间及监测位置判断预定时间内是否有运营车辆即将经过该监测位置；若有，则将疑似遗撒数据包发送至运营车辆确认，若无，则将疑似遗撒数据包交由人工确认。
  11. 一种公路遗撒物快速检测方法，其特征在于，包括：
    接收多组运营车辆监测信息，其中，运营车辆监测信息包括车速、监测时间、监测位置和监测图像；
    对各运营车辆监测信息中的监测图像进行处理，分析得到疑似遗撒物信息，根据疑似遗撒物信息和相应运营车辆监测信息生成疑似遗撒数据包；
    对于每一疑似遗撒数据包，根据该疑似遗撒数据包中的监测时间及监测位置，提取满足第一预定空间及时间的疑似遗撒数据包；比较该疑似遗撒数据包与提取出的疑似遗撒数据包，根据比较结果计算该疑似遗撒数据包的疑似概率；
    将疑似概率大于预定阈值的疑似遗撒数据包发送给相应运营车辆确认或交由人工确认，若确认通过，则根据疑似遗撒数据包按预定格式生成遗撒信息报告单。
  12. 一种计算机设备，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，其特征在于，所述处理器执行所述计算机程序时实现权利要求11所述方法。
PCT/CN2017/107460 2017-10-24 2017-10-24 一种公路遗撒物快速检测系统及方法 WO2019079965A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/107460 WO2019079965A1 (zh) 2017-10-24 2017-10-24 一种公路遗撒物快速检测系统及方法

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/107460 WO2019079965A1 (zh) 2017-10-24 2017-10-24 一种公路遗撒物快速检测系统及方法

Publications (1)

Publication Number Publication Date
WO2019079965A1 true WO2019079965A1 (zh) 2019-05-02

Family

ID=66247168

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/107460 WO2019079965A1 (zh) 2017-10-24 2017-10-24 一种公路遗撒物快速检测系统及方法

Country Status (1)

Country Link
WO (1) WO2019079965A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112508345A (zh) * 2020-11-12 2021-03-16 聚通智用(重庆)信息科技有限公司 一种整改工作管理系统及方法

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN202042052U (zh) * 2010-12-14 2011-11-16 广东鑫程电子科技有限公司 智能视频事件检测系统
US20120274764A1 (en) * 2011-04-29 2012-11-01 Hon Hai Precision Industry Co., Ltd. Monitoring system and method for storing video
CN102867183A (zh) * 2012-08-23 2013-01-09 安科智慧城市技术（中国）有限公司 一种车辆遗撒物检测方法、装置及智能交通监控系统
CN104103172A (zh) * 2014-07-25 2014-10-15 昆明联诚科技股份有限公司 一种高速公路监控方法及系统
WO2015049352A1 (en) * 2013-10-03 2015-04-09 Kapsch Trafficcom Ab System for traffic behaviour surveillance
CN205247621U (zh) * 2015-12-24 2016-05-18 北京万集科技股份有限公司 一种基于tof相机的车辆遗撒物检测装置
CN106406287A (zh) * 2016-11-08 2017-02-15 思建科技有限公司 一种车辆安全状态监控预警方法及系统

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN202042052U (zh) * 2010-12-14 2011-11-16 广东鑫程电子科技有限公司 智能视频事件检测系统
US20120274764A1 (en) * 2011-04-29 2012-11-01 Hon Hai Precision Industry Co., Ltd. Monitoring system and method for storing video
CN102867183A (zh) * 2012-08-23 2013-01-09 安科智慧城市技术（中国）有限公司 一种车辆遗撒物检测方法、装置及智能交通监控系统
WO2015049352A1 (en) * 2013-10-03 2015-04-09 Kapsch Trafficcom Ab System for traffic behaviour surveillance
CN104103172A (zh) * 2014-07-25 2014-10-15 昆明联诚科技股份有限公司 一种高速公路监控方法及系统
CN205247621U (zh) * 2015-12-24 2016-05-18 北京万集科技股份有限公司 一种基于tof相机的车辆遗撒物检测装置
CN106406287A (zh) * 2016-11-08 2017-02-15 思建科技有限公司 一种车辆安全状态监控预警方法及系统

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112508345A (zh) * 2020-11-12 2021-03-16 聚通智用(重庆)信息科技有限公司 一种整改工作管理系统及方法

Similar Documents

Publication Publication Date Title
CN107918762B (zh) 一种公路遗撒物快速检测系统及方法
CN109686088B (zh) 一种交通视频告警方法、设备及系统
US9365217B2 (en) Mobile pothole detection system and method
US9111442B2 (en) Estimating incident duration
CN104392610A (zh) 基于分布式视频的高速路交通事件影响范围动态监控方法
CN108986448B (zh) 一种交通设施管理方法及终端设备
CN105260832A (zh) 基于订单数据的出租车驾驶员绩效评估方法
CN114248819B (zh) 基于深度学习的铁路侵限异物无人机检测方法、装置及系统
CN109767618B (zh) 一种公安交管业务异常数据综合研判方法及系统
CN112687099B (zh) 一种超载嫌疑车辆判定方法和装置
CN114399744A (zh) 一种车型识别方法、装置、电子设备和存储介质
CN113487877A (zh) 一种道路车辆违规停车监控方法
CN111429729A (zh) 一种基于flink架构的实时车速监测方法
CN107393304A (zh) 一种基于卡口通行的交通状态检测系统
CN109272760B (zh) 一种scats系统检测器数据异常值的在线检测方法
CN102622883A (zh) 判定交通事件解除的方法及装置
WO2019079965A1 (zh) 一种公路遗撒物快速检测系统及方法
CN109934161B (zh) 基于卷积神经网络的车辆识别与检测方法及系统
CN116168356B (zh) 一种基于计算机视觉的车辆损伤判别方法
KR102297801B1 (ko) 사업용 차량을 이용하여 도로 위험정보를 탐지하는 방법 및 장치
CN105046952A (zh) 稀疏道路交通事件地空联合检测方法
Mehajabin et al. Real-time deep learning based road deterioration detection for smart cities
CN115273474B (zh) 一种基于人工智能的rpa巡检机器人巡检管理系统
CN112989069B (zh) 一种基于知识图谱与区块链的交通违章分析方法
CN114202903A (zh) 基于dem数据的车辆预警方法、装置、设备及存储介质

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: 17929560

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: 17929560

Country of ref document: EP

Kind code of ref document: A1