WO2019001044A1 - 基于手机数据的车辆类型识别方法和设备 - Google Patents

基于手机数据的车辆类型识别方法和设备 Download PDF

Info

Publication number
WO2019001044A1
WO2019001044A1 PCT/CN2018/080945 CN2018080945W WO2019001044A1 WO 2019001044 A1 WO2019001044 A1 WO 2019001044A1 CN 2018080945 W CN2018080945 W CN 2018080945W WO 2019001044 A1 WO2019001044 A1 WO 2019001044A1
Authority
WO
WIPO (PCT)
Prior art keywords
vehicle
trajectory
data
service area
obtaining
Prior art date
Application number
PCT/CN2018/080945
Other languages
English (en)
French (fr)
Inventor
吴伟令
刘伟
赵凯
李勇
金德鹏
牟涛
李贻武
毕玉峰
郑建辉
常颖
许孝滨
邵晓明
Original Assignee
山东省交通规划设计院
清华大学
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 山东省交通规划设计院, 清华大学 filed Critical 山东省交通规划设计院
Priority to US16/322,095 priority Critical patent/US10573173B2/en
Publication of WO2019001044A1 publication Critical patent/WO2019001044A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/029—Location-based management or tracking services
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G06N20/20—Ensemble learning
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
    • G08G1/0112—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from the vehicle, e.g. floating car data [FCD]
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
    • G08G1/012—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from other sources than vehicle or roadside beacons, e.g. mobile networks
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125—Traffic data processing
    • G08G1/0133—Traffic data processing for classifying traffic situation
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/015—Detecting movement of traffic to be counted or controlled with provision for distinguishing between two or more types of vehicles, e.g. between motor-cars and cycles
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/023—Services making use of location information using mutual or relative location information between multiple location based services [LBS] targets or of distance thresholds
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/025—Services making use of location information using location based information parameters
    • H04W4/027—Services making use of location information using location based information parameters using movement velocity, acceleration information

Definitions

  • the present invention relates to computer technology, and in particular to a vehicle type identification method and apparatus based on mobile phone data.
  • the intelligent transportation system is a system that effectively combines computer technology, communication technology and control technology into transportation management. It can manage traffic in real time, accurately and efficiently, and is the main direction of traffic development in the 21st century. Among them, the automatic identification and type detection of vehicles is an important part of the intelligent transportation system, which can greatly improve the efficiency of traffic management and provide an important reference for transportation planning.
  • Intrusive detection technology requires the installation of sensors under the road surface, such as pneumatic tubes, loop detectors, magnetic sensors and piezoelectric sensors. When the vehicle passes by, the sensor can collect corresponding information, such as speed, weight, body length, etc. Thereby determining the type of vehicle.
  • the disadvantages of this type of method are the short service life and inconvenient installation.
  • Non-intrusive methods are mainly used to install sensors outside the road, such as radar, infrared sensors, acoustic sensors and cameras. Information such as body contour, height and wheelbase is received to classify the vehicle.
  • This type of method has the advantage of high recognition efficiency, but the cost of large-scale deployment is high.
  • the method of identifying a vehicle model by analyzing an image in a video signal is the fastest growing in recent years, and the accuracy rate is also the highest.
  • its relatively large drawback is that the installation cost is relatively high and is affected by the weather.
  • the present invention proposes a mobile phone data based vehicle type identification method and apparatus that overcomes the above problems or at least partially solves the above problems.
  • the present invention provides a vehicle type identification method based on mobile phone data, optionally including the steps of:
  • the vehicle classifier is a vehicle classifier obtained by training a machine learning method and a feature vector of a sample.
  • the determining, according to the movement trajectory data, the mobile phone user who is riding the same vehicle, obtaining the number of passengers corresponding to the vehicle trajectory and the vehicle trajectory including:
  • the intersection has an intersection with each other; according to the different first geographic location and the second geographic location, the trajectory data points are moved one by one. For each collection;
  • the degree of matching between the movement trajectories in each set is calculated one by one, and if it matches, the movement trajectory belongs to the passengers on the same vehicle, thereby determining the number of users corresponding to the vehicle trajectory and the vehicle trajectory.
  • the starting point type includes a residential area, an industrial area, a commercial area, an attraction, an entertainment area, and a bus station.
  • the obtaining vehicle driving data according to the vehicle trajectory includes:
  • One or more kinds of driving data including the average speed, the maximum speed, the speed standard deviation, and the speed distribution are calculated according to the vehicle trajectory after deducting the service area sub-track and the traffic jam trajectory.
  • the service area data includes: a number of service area stays, a total service area stay time, and an average time spent in each service area.
  • the machine learning method is a random forest method.
  • the present invention provides a computer readable storage medium having stored thereon a computer program that, when executed by a processor, implements the steps of any of the methods described above.
  • the present invention provides a computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the program to implement any of the above The steps of the method.
  • the mobile phone big data proposed by the present invention uses the mobile phone signaling data collected and provided by the operator, and the cost of obtaining the data is low.
  • the mobile phone signaling data covers almost all inter-city peers, so the user coverage is high, making the calculation of the final output of intercity traffic travel data more trustworthy.
  • the mobile phone signaling data corresponds to a continuous trajectory, and the acquired mobile phone signaling data has a long time span and wide spatial coverage. Therefore, in addition to being used for studying current highway vehicle types, it can also be used to study historical highways. The type of vehicle is used to study the demand for highways in different areas.
  • FIG. 1 is a schematic flow chart of an execution method in an embodiment of the present invention.
  • the vehicle type identification method based on the mobile phone data is used to determine whether the vehicle on the highway is a car type, a passenger car, a bus or a truck, and the method comprises the steps of:
  • S101 acquiring, by using, the mobile trajectory data recorded by the mobile phone of the user in the base station in the first preset time period;
  • S102 determining, according to the movement trajectory data, the mobile phone user who is riding the same vehicle, and obtaining the number of passengers corresponding to the vehicle trajectory and the vehicle trajectory;
  • S103 obtains a vehicle starting point according to the vehicle trajectory, and obtains a starting point type of the vehicle trajectory according to the geographic data;
  • S104 obtains vehicle driving data according to the vehicle trajectory
  • S105 obtains service area data of the vehicle staying according to the vehicle trajectory
  • S106 constructs a feature vector according to at least a number of passengers corresponding to the vehicle trajectory, a starting point type, a driving speed data of the vehicle, and a service area data of the vehicle staying;
  • S107 processes the feature vector by the vehicle classifier to obtain the vehicle type recognition result
  • the vehicle classifier is a vehicle classifier obtained by training a machine learning method and a feature vector of a sample.
  • the use of mobile phone big data proposed by the present invention that is, using mobile phone signaling data collected and provided by an operator, has low cost of obtaining data.
  • the mobile phone signaling data covers almost all inter-city peers, so the user coverage is high, making the calculation of the final output of intercity traffic travel data more trustworthy.
  • the mobile phone signaling data corresponds to a continuous trajectory, and the acquired mobile phone signaling data has a long time span and wide spatial coverage. Therefore, in addition to being used for studying current highway vehicle types, it can also be used to study historical highways. The type of vehicle is used to study the demand for highways in different areas.
  • the user's mobile phone actually refers to the calling card, that is, the user's mobile phone is mainly distinguished by the calling card used by the user, and the signs of different calling cards correspond to different moving track data in the base station.
  • the movement trajectory data is composed of a movement trajectory corresponding to a large number of users.
  • the movement trajectory means that the base station has the positioning data of the mobile phone after the mobile phone is connected to the base station, and the positioning data has the spatial domain feature.
  • the trajectory data of the user a and the user b are respectively among them Is collecting Moment, Yes
  • the location information of the user's mobile phone at any time
  • the code of the handset base station can be used to represent the location of the handset, ie It can be a base station number or a base station coverage area or a mobile network location of a mobile phone.
  • Determining, according to the movement trajectory data, the mobile phone user who is riding the same vehicle, and obtaining the number of passengers corresponding to the vehicle trajectory and the vehicle trajectory including:
  • the intersection has an intersection with each other; according to the different first geographic location and the second geographic location, the trajectory data points are moved one by one. For each collection;
  • the first geographic location and the second geographic location may be two points on the road, such as the start and end points of the segment road.
  • the time coincidence degree is greater than the first time coincidence degree, then user a and user b may be on the same vehicle, user a and user b belong to the same set, otherwise a and b belong to different sets; the first time coincides
  • the degree is a preset value, for example, 80% in one embodiment, then the above calculations are respectively performed versus Whether the coincidence degree is greater than 80%
  • this step the movement trajectories that are apparently not in the same vehicle are separated from each other into different sets, and this step simplifies the calculation of the degree of matching between the subsequent calculations of the movement trajectories.
  • the degree of matching between the movement trajectories in each set is calculated one by one, and if it matches, the movement trajectory belongs to the passengers on the same vehicle, thereby determining the number of users corresponding to the vehicle trajectory and the vehicle trajectory.
  • the spatiotemporal coincidence degree determination includes the steps of:
  • the time intersection can be initialized to according to with Complement user A and user b trajectory data, so that have Corresponding to it, among them
  • the ratio of Tb Location sequence ⁇ l 1 ?? l i ?? l n ' ⁇ can be determined and the location of the sequence Ta ⁇ l 1 ?? l i ?? l m' ⁇ and a user whether the user is in the same b On the car.
  • the average of the Euclidean distances determines whether user a and user b are in the same car.
  • a user data track completion at time t i the user can point a track in the vicinity of the time t i with And GPS accuracy and the road matching algorithm to calculate the geographical coordinates of a user l i at time t i.
  • trajectory data of user x does not coincide with any other trajectory data in the set (ie, they are not coincident by the sequence of locations), then the user x will be considered to be riding a car alone.
  • Each set is processed one by one until all vehicle trajectories are obtained, it being understood that the trajectory of the vehicle is the trajectory of the user riding the vehicle. If there are multiple users riding the car, a suitable vehicle trajectory can be obtained according to the calculation of each user trajectory. At this point, the trajectory of the car obtained by the human trajectory is completed.
  • the starting point type includes a residential area, an industrial area, a commercial area, an attraction, an entertainment area, and a bus station to extract a starting point (ls, le) of all moving tracks, and a type of location obtained by combining the map.
  • the starting point can take one of the following six types, a total of 36 cases.
  • the starting point of different types of vehicles generally has a big difference.
  • the starting point of a truck is generally an industrial area
  • the starting point of a bus is generally a bus stop, so that the characteristics of the starting point are used in machine learning, and can be identified.
  • the type of vehicle that is, the type of vehicle can be distinguished by the type of starting point.
  • two matrices of the starting point and the ending point can be obtained.
  • the two dimensions of the matrix are the track number and the location type.
  • a moving track may have multiple starting points (ls, le), and multiple starting points (ls, le) divide the moving track into several segments.
  • the obtaining the vehicle driving data according to the vehicle trajectory includes: calculating a time interval in which the speed is less than the first speed V un according to all the vehicle trajectories Set, calculate a time interval The number of vehicle trajectories coincident in the third geographical position; when the number of vehicle trajectories coincident in the third geographical position is greater than the traffic jam threshold, it is determined that the traffic jam is obtained during the period of time, and the traffic jam trajectory is obtained; the third location does not include the service area ;
  • One or more kinds of driving data including the average speed, the maximum speed, the speed standard deviation, and the speed distribution are calculated according to the vehicle trajectory after deducting the service area sub-track and the traffic jam trajectory.
  • the speed set of all the trajectory segments of the vehicle (the trajectory between the two sampling moments), find all the speed sets less than 40/km h, and then smooth the trajectory segments to find continuous a period of time During this time period, the user's speed is less than V un and the vehicle location is not in the service area collection. Perform the same operation for all users to get the corresponding corresponding to each track When the number of coincident vehicles reaches a certain threshold for a certain period of time, it is judged that there is a traffic jam. The trajectory of this time period is deleted from the movement trajectories of these vehicles when calculating the travel data.
  • the service area data can also be calculated synchronously.
  • the service area data includes: the number of stays in the service area, the total time of stay in the service area, and the average time spent in each service area.
  • reference table 2 can be related to the formula, where Indicates the stop point of the vehicle in the kth service area, and the set of stay points of the moving track is recorded as
  • the machine learning method is a random forest method.
  • the following sufficient conditions are used to generate the training data corresponding to the sample track of the vehicle. If the number of vehicles in the vehicle is greater than 5; b) the bus stop is the bus; if the a) the average speed exceeds 100km/h and the starting point is a residential area or an entertainment area, then a small passenger car; if the a) average speed does not exceed 100km/h and the starting point is an industrial area, it is a truck.
  • the present invention also provides a computer readable storage medium having stored thereon a computer program that, when executed by a processor, implements the steps of any of the methods described above.
  • the present invention also provides a computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the program to implement any of the methods described above A step of.
  • At least one As used herein, “at least one”, “one or more”, and “and/or” are an open-ended expression that can be combined and separated in use. For example, “at least one of A, B, and C”, “at least one of A, B, or C", “one or more of A, B, and C” and “one of A, B, or C, or “Multiple” means only A, only B, only C, A and B together, A and C together, B and C together or A, B and C together.
  • automated refers to any process or operation that is performed without substantial human input when performing a process or operation. However, the process or operation may be automated even if substantial or insubstantial human input received prior to performing the process or operation is used in performing the process or operation. If the input affects how the process or operation will proceed, then the input is considered to be substantial. Human input that does not affect the processing or operation is not considered to be substantial.
  • computer readable medium refers to any tangible storage device and/or transmission medium that participates in providing instructions to a processor for execution.
  • the computer readable medium can be a serial set of instructions encoded in a network transmission (e.g., SOAP) over an IP network.
  • a network transmission e.g., SOAP
  • Such a medium may take many forms, including but not limited to, non-volatile media, volatile media, and transmission media.
  • Non-volatile media includes, for example, NVRAM or magnetic or optical disks.
  • Volatile media includes dynamic memory such as RAM, such as main memory.
  • Computer readable media include, for example, floppy disks, flexible disks, hard disks, magnetic tape or any other magnetic media, magneto-optical media, CD-ROM, any other optical media, perforated cards, paper tape, any other physics having a hole pattern Medium, RAM, PROM, EPROM, FLASH-EPROM, solid state media such as memory cards, any other memory chip or tape cartridge, carrier wave described later, or any other medium readable by a computer.
  • a digital file attachment or other self-contained information file or collection of emails is considered a distribution medium equivalent to a tangible storage medium.
  • the computer readable medium is configured as a database
  • the database can be any type of database, such as a relational database, a hierarchical database, an object oriented database, and the like. Accordingly, the present invention is considered to include a tangible storage medium or distribution medium and equivalents well known in the art, and a medium developed in the future, in which the software implementation of the present invention is stored.
  • module or “tool” as used herein refers to any known or later developed hardware, software, firmware, artificial intelligence, fuzzy logic, or a combination of hardware and software capable of performing the functions associated with the element. Additionally, while the invention has been described in terms of exemplary embodiments, it is understood that aspects of the invention may be

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Software Systems (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Analytical Chemistry (AREA)
  • Chemical & Material Sciences (AREA)
  • Medical Informatics (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Traffic Control Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

本发明提供基于手机数据的车辆类型识别方法和设备用于解决提供一种方便、低成本的车型识别方法问题。其中方法包括：获取第一预设时间段内的用户的手机在基站中记录的移动轨迹数据；根据移动轨迹数据判断搭乘同一车辆的手机用户，获得车辆轨迹和车辆轨迹对应的乘车用户数；根据车辆轨迹获得车辆起讫点，结合地理数据获得车辆轨迹的起讫点类型；根据车辆轨迹获得车辆行驶数据；根据车辆轨迹获得车辆停留的服务区数据；本发明提出的利用手机大数据，使用由运营商采集和提供的手机信令数据，获得数据的成本低。

Description

基于手机数据的车辆类型识别方法和设备 技术领域
本发明涉及计算机技术，具体涉及基于手机数据的车辆类型识别方法和设备。
背景技术
智能交通系统是将计算机技术、通信技术以及控制技术等有效地结合运用于交通运输管理的体系，能够实时、准确、高效地管理交通，是21世纪交通发展的主要方向。其中，车辆的自动识别和类型检测是智能交通系统中一个重要的组成部分，可以大大提高交通管理的效率，为交通规划提供重要参考依据。
目前国内外主流的车型识别方法主要可以分为两大类，侵入性分类方法和非侵入性识别方法。侵入性检测技术需要在公路路面下安装传感器，比如气动管，环路检测器，磁传感器和压电传感器等，当车辆经过时传感器能够收集相应的信息，比如车速，车重，车身长度等，从而判断车辆类型。这一类方法的缺点是使用寿命短，安装不方便。非侵入式方法主要是在道路以外的地方安装传感器，比如雷达，红外传感器，声传感器和摄像头。接收到车体轮廓，高度和轴距等信息来给车辆分类。这一类方法的有优点是识别效率高，但是大规模部署的成本较高。其中，随着图像处理技术的发展，通过分析视频信号中的图像而来识别车型的方法是近年来发展最快的，准确率也是最高的。但是它存在比较大的缺陷是安装成本比较高，并且受天气的影响比较大。
因此需要提供一种方便、低成本的车型识别方法。
发明内容
鉴于上述问题，本发明提出了克服上述问题或者至少部分地解决上述问题的基于手机数据的车辆类型识别方法和设备。
为此目的，第一方面，本发明提供一种基于手机数据的车辆类型识别方法，可选地，包括步骤：
获取第一预设时间段内的用户的手机在基站中记录的移动轨迹数据；
根据移动轨迹数据判断搭乘同一车辆的手机用户，获得车辆轨迹和车辆轨迹对应的乘车用户数；
根据车辆轨迹获得车辆起讫点，结合地理数据获得车辆轨迹的起讫点类型；
根据车辆轨迹获得车辆行驶数据；
根据车辆轨迹获得车辆停留的服务区数据；
至少根据车辆轨迹对应的乘车用户数、起讫点类型、车辆的行驶速度数据和车辆停留的服务区数据构建特征向量；
用车辆分类器处理特征向量，获得车辆类型识别结果；
车辆分类器是用机器学习方法和样本的特征向量训练获得的车辆分类器。
可选地，所述根据移动轨迹数据判断搭乘同一车辆的手机用户，获得车辆轨迹和车辆轨迹对应的乘车用户数，包括：
若移动轨迹在经过第一地理位置和第二地理位置的时间重合度大于第一时间重合度，则互相之间具有交集；根据不同的第一地理位置和第二地理位置，逐一移动轨迹数据分为各集合；
逐一计算各集合中各移动轨迹之间的匹配度，若匹配则这个移动轨迹属于同一车辆上的乘客，进而确定车辆轨迹和车辆轨迹对应的用户数。
可选地，所述起讫点类型包括住宅区、工业区、商业区、景点、 娱乐区和汽车站。
可选地，所述根据车辆轨迹获得车辆行驶数据，包括：
根据所有的车辆轨迹计算速度小于第一速度V un的时间区间
Figure PCTCN2018080945-appb-000001
Figure PCTCN2018080945-appb-000002
的集合，
计算某一时间区间
Figure PCTCN2018080945-appb-000003
内，在第三地理位置重合的车辆轨迹的数量；当在第三地理位置重合的车辆轨迹的数量大于堵车阈值时，判断该段时间内堵车，获得堵车子轨迹；第三位置不包括服务区；
根据服务区的地理位置信息，计算车辆轨迹中处于服务区且车速小于第二速度V sa 的服务区子轨迹，
根据扣除服务区子轨迹和堵车子轨迹后的车辆轨迹，计算车辆的包括平均速度、最大速度、速度标准差、速度分布中的一种或多种行驶数据。
可选地，所述服务区数据包括：服务区停留次数、服务区停留总时间、平均在每个服务区停留的时间。
可选地，所述机器学习方法为随机森林方法。
第二方面，本发明提供一种计算机可读存储介质，其上存储有计算机程序，该程序被处理器执行时实现如上任一所述方法的步骤。
第三方面，本发明提供一种计算机设备，包括存储器、处理器以及存储在所述存储器上并可在所述处理器上执行的计算机程序，所述处理器执行所述程序时实现如上任一所述方法的步骤。
由上述技术方案可知，本发明提出的利用手机大数据，使用由运营商采集和提供的手机信令数据，获得数据的成本低。而手机信令数据几乎覆盖所有城际间同行的用户，因此用户覆盖率高，使计算最终输出的城际交通出行的数据更值得信赖。而手机信令数据对应的移动轨迹连续，且能获取到的手机信令数据的时间跨度长、空间覆盖广，因此除了可以用于研究当前高速公路车辆类型，还可以用于研究历史上高速公路车辆类型，从而研究人们对不同区域的高速公路的需求。
前面是提供对本发明一些方面的理解的简要发明内容。这个部分既不是本发明及其各种实施例的详尽表述也不是穷举的表述。它既不用于识别本发明的重要或关键特征也不限定本发明的范围，而是以一种简化形式给出本发明的所选原理，作为对下面给出的更具体的描述的简介。应当理解，单独地或者组合地利用上面阐述或下面具体描述的一个或多个特征，本发明的其它实施例也是可能的。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图是本发明的一些实施例，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为本发明的一个实施例中执行方法流程示意图。
具体实施方式
下面将结合示例性的实施例描述本发明。
在发明的一个实施例中，基于手机数据的车辆类型识别方法用于判断高速公路上车辆是汽车类型，小客车、大客车或者货车，方法包括步骤：
S101获取第一预设时间段内的用户的手机在基站中记录的移动轨迹数据；
S102根据移动轨迹数据判断搭乘同一车辆的手机用户，获得车辆轨迹和车辆轨迹对应的乘车用户数；
S103根据车辆轨迹获得车辆起讫点，结合地理数据获得车辆轨迹的起讫点类型；
S104根据车辆轨迹获得车辆行驶数据；
S105根据车辆轨迹获得车辆停留的服务区数据；
S106至少根据车辆轨迹对应的乘车用户数、起讫点类型、车辆的行驶速度数据和车辆停留的服务区数据构建特征向量；
S107用车辆分类器处理特征向量，获得车辆类型识别结果；
车辆分类器是用机器学习方法和样本的特征向量训练获得的车辆分类器。
由上述技术方案可知，本发明提出的利用手机大数据，即使用由运营商采集和提供的手机信令数据，获得数据的成本低。而手机信令数据几乎覆盖所有城际间同行的用户，因此用户覆盖率高，使计算最终输出的城际交通出行的数据更值得信赖。而手机信令数据对应的移动轨迹连续，且能获取到的手机信令数据的时间跨度长、空间覆盖广，因此除了可以用于研究当前高速公路车辆类型，还可以用于研究历史上高速公路车辆类型，从而研究人们对不同区域的高速公路的需求。
用户的手机实际指电话卡，即主要通过用户所使用的电话卡区分用户手机，不同的电话卡的标志在基站中对应不同的移动轨迹数据。移动轨迹数据是由大量用户对应的移动轨迹组成，移动轨迹是指手机连接基站后基站就具有了手机的定位数据，该定位数据是具有空域特征的，例如用户a和用户b的轨迹数据分别为
Figure PCTCN2018080945-appb-000004
Figure PCTCN2018080945-appb-000005
Figure PCTCN2018080945-appb-000006
其中
Figure PCTCN2018080945-appb-000007
是采集
Figure PCTCN2018080945-appb-000008
的时刻，
Figure PCTCN2018080945-appb-000009
是
Figure PCTCN2018080945-appb-000010
时刻用户手机的位置信息，
Figure PCTCN2018080945-appb-000011
在一些实施例中，可以用手机基站的编码代表手机位置，即
Figure PCTCN2018080945-appb-000012
可以是基站编号或基站覆盖区域或者是手机的移动网络定位。
所述根据移动轨迹数据判断搭乘同一车辆的手机用户，获得车辆轨迹和车辆轨迹对应的乘车用户数，包括：
若移动轨迹在经过第一地理位置和第二地理位置的时间重合度大于第一时间重合度，则互相之间具有交集；根据不同的第一地理位置和第二地理位置，逐一移动轨迹数据分为各集合；
第一地理位置和第二地理位置可以是道路上两点，例如段道路的起点和终点。
以用户a和用户b为例，说明如何将移动轨迹数据分到各集合：假设有两个用户a和b，根据他们对应的移动轨迹，计算经过道路起点s和道路终点e的时间分别为
Figure PCTCN2018080945-appb-000013
若
Figure PCTCN2018080945-appb-000014
Figure PCTCN2018080945-appb-000015
之间的时间重合度大于第一时间重合度，则用户a和用户b可能是在同一车辆上的，用户a和用户b归入同一集合，否则a和b归入不同集合；第一时间重合度为预设值，例如在一个实施例中为80％，则上述分别计算
Figure PCTCN2018080945-appb-000016
与
Figure PCTCN2018080945-appb-000017
Figure PCTCN2018080945-appb-000018
之间的重合度是否大于80％
在该步骤把明显不在同一辆车的移动轨迹互相分离到不同的集合，该步骤简化了后续的计算乘坐各移动轨迹之间的匹配度的计算量。
逐一计算各集合中各移动轨迹之间的匹配度，若匹配则这个移动轨迹属于同一车辆上的乘客，进而确定车辆轨迹和车辆轨迹对应的用户数。
判断属于同一集合的两条移动轨迹是否是乘坐同一辆车上的用户产生的，主要根据移动轨迹的时空重合度判断。例如在本发明的一个实施例中，时空重合度判断包括步骤：
首先取两条轨迹的时间交集，时间交集可以初始化为
Figure PCTCN2018080945-appb-000019
Figure PCTCN2018080945-appb-000020
根据
Figure PCTCN2018080945-appb-000021
Figure PCTCN2018080945-appb-000022
和
Figure PCTCN2018080945-appb-000023
Figure PCTCN2018080945-appb-000024
补全用户a和用户b轨迹数据，使得对于
Figure PCTCN2018080945-appb-000025
Figure PCTCN2018080945-appb-000026
具有
Figure PCTCN2018080945-appb-000027
与之对应，其中
Figure PCTCN2018080945-appb-000028
因此比对Tb的地点序列{l 1……l i……l n’}和Ta的的地点序列{l 1……l i……l m’}即可判断用户a与用户b是否处于同一辆车上。在一个实施例中，根据Tb的地点序列{l 1……l i……l n’}和Ta的的地点序列{l 1……l i……l m’}之间的对应点之间的欧几里得距离的平均值判断用户a与用户b是否处于同一辆车上。
补全用户a在t i时刻的轨迹数据，可以根据用户a在t i时刻附近的轨迹点
Figure PCTCN2018080945-appb-000029
和
Figure PCTCN2018080945-appb-000030
以及GPS精度和道路匹配算法，计算用户a在t i时刻的地理坐标l i。
若用户x的轨迹数据与集合中任一其他轨迹数据的都没有重合(即通过地点序列判断他们不重合)，则将认为该用户x单独乘坐一辆车。逐一处理各集合，直到获得所有车辆轨迹，可以理解的是车辆的轨迹即乘坐该车的用户的轨迹。若有多个用户乘坐该车，可以根据各用户轨迹的计算获得较为合适的车辆轨迹。到此，完成了由人的轨迹获得车的轨迹。
所述起讫点类型包括住宅区、工业区、商业区、景点、娱乐区和汽车站提取出所有移动轨迹的起讫点(ls,le)，结合地图获得地点的类型。
起讫点都可以取以下6种类型中的一种，一共有36种情况。不同类型的车辆的起讫点一般有比较大的区别，比如货车的起讫点一般都是工业区，而客车的起讫点一般都是公交站，从而将起讫点的特征 用于机器学习中，可以识别车辆的类型，即可以通过起讫点的类型来对车辆的类别进行区分。通过提取轨迹起讫点类型，可以得到起点和终点两个矩阵，矩阵的两个维度分别是轨迹编号和地点类型。一段移动轨迹可以有多个起讫点(ls,le)，多个起讫点(ls,le)将移动轨迹分为若干段。
所述根据车辆轨迹获得车辆行驶数据，包括：根据所有的车辆轨迹计算速度小于第一速度V un的时间区间
Figure PCTCN2018080945-appb-000031
的集合，计算某一时间区间
Figure PCTCN2018080945-appb-000032
内，在第三地理位置重合的车辆轨迹的数量；当在第三地理位置重合的车辆轨迹的数量大于堵车阈值时，判断该段时间内堵车，获得堵车子轨迹；第三位置不包括服务区；
根据服务区的地理位置信息，计算车辆轨迹中处于服务区且车速小于第二速度V sa 的服务区子轨迹，
根据扣除服务区子轨迹和堵车子轨迹后的车辆轨迹，计算车辆的包括平均速度、最大速度、速度标准差、速度分布中的一种或多种行驶数据。
在高速公路上，不同类型的汽车限速不同，所以速度特征在我们的问题中是比较重要的特征。在计算速度特征之前我们也要对数据做预处理，来排除车辆停止不动的情况，比如堵车，进入服务区等。
对于堵车情况，计算出车辆的所有轨迹段(两个采样时刻之间的轨迹)的速度集合，找出所有小于40/km h的速度集，然后对这些轨迹段做平滑处理，找出连续的一段时间
Figure PCTCN2018080945-appb-000033
在该时间段内用户的速度都小于V un，并且车辆位置不在服务区合集。对于所有用户都进行相同操作获得每段轨迹对应的
Figure PCTCN2018080945-appb-000034
当某段时间的重合的车辆数量到达一定阈值时，判断存在堵车的情况。在计算行驶数据时将该时段的轨迹从这些车辆的移动轨迹中删除。
对于我们要研究的高速公路，我们首先找到高速公路中所有服务区位置集合Lservices＝{l s1,……l si……l sk}，即其中包含K个不同的服 务区。计算出车辆的所有轨迹段(两个采样时刻之间的轨迹)的速度集合V，判断速度小于预设值，且位置与服务区位置集合中某一点的的距离小于预设值的轨迹，在计算行驶数据时将该轨迹从这些车辆的移动轨迹中删除。
在计算行驶数据时，可以选取计算下列表1中的一种或多种：
表1行驶数据表
Figure PCTCN2018080945-appb-000035
在计算判断移动轨迹是否包括服务区轨迹时，还可以同步计算服务区数据。所述服务区数据包括：服务区停留次数、服务区停留总时间、平均在每个服务区停留的时间。对于一个移动轨迹，具有k个停留点，参考表2可以相关公式，其中
Figure PCTCN2018080945-appb-000036
表示车辆在第k个服务区的停留点，移动轨迹的停留点集合记为
Figure PCTCN2018080945-appb-000037
表2服务区数据表
Figure PCTCN2018080945-appb-000038
所述机器学习方法为随机森林方法。在构建样本数据时，采用下列充分条件生成车辆的样本轨迹对应的训练数据，若满足a)车上人数大于5人；b)起讫点都是公交站则是客车；若满足a)平均车速超过100km/h且起讫点是住宅区或者娱乐区，则小客车；若满足a)平均车速不超过100km/h且起讫点是工业区，则是货车。
本发明还提供一种计算机可读存储介质，其上存储有计算机程序，该程序被处理器执行时实现如上任一所述方法的步骤。
本发明还提供一种计算机设备，包括存储器、处理器以及存储在所述存储器上并可在所述处理器上执行的计算机程序，所述处理器执行所述程序时实现如上任一所述方法的步骤。
本文中使用的“至少一个”、“一个或多个”以及“和/或”是开放式的表述，在使用时可以是联合的和分离的。例如，“A、B和C中的至少一个”，“A、B或C中的至少一个”，“A、B和C中的一个或多个”以及“A、B或C中的一个或多个”指仅有A、仅有B、仅有C、A和B一起、A和C一起、B和C一起或A、B和C一起。
术语“一个”实体是指一个或多个所述实体。由此术语“一个”、“一个或多个”和“至少一个”在本文中是可以互换使用的。还应注意到术语“包括”、“包含”和“具有”也是可以互换使用的。
本文中使用的术语“自动的”及其变型是指在执行处理或操作时没有实质的人为输入的情况下完成的任何处理或操作。然而，即使在 执行处理或操作时使用了执行所述处理或操作前接收到的实质的或非实质的人为输入，所述处理或操作也可以是自动的。如果输入影响所述处理或操作将怎样进行，则视该人为输入是实质的。不影响所述处理或操作进行的人为输入不视为是实质的。
本文中使用的术语“计算机可读介质”是指参与将指令提供给处理器执行的任何有形存储设备和/或传输介质。计算机可读介质可以是在IP网络上的网络传输(如SOAP)中编码的串行指令集。这样的介质可以采取很多形式，包括但不限于非易失性介质、易失性介质和传输介质。非易失性介质包括例如NVRAM或者磁或光盘。易失性介质包括诸如主存储器的动态存储器(如RAM)。计算机可读介质的常见形式包括例如软盘、柔性盘、硬盘、磁带或任何其它磁介质、磁光介质、CD-ROM、任何其它光介质、穿孔卡、纸带、任何其它具有孔形图案的物理介质、RAM、PROM、EPROM、FLASH-EPROM、诸如存储卡的固态介质、任何其它存储芯片或磁带盒、后面描述的载波、或计算机可以读取的任何其它介质。电子邮件的数字文件附件或其它自含信息档案或档案集被认为是相当于有形存储介质的分发介质。当计算机可读介质被配置为数据库时，应该理解该数据库可以是任何类型的数据库，例如关系数据库、层级数据库、面向对象的数据库等等。相应地，认为本发明包括有形存储介质或分发介质和现有技术公知的等同物以及未来开发的介质，在这些介质中存储本发明的软件实施。
本文中使用的术语“确定”、“运算”和“计算”及其变型可以互换使用，并且包括任何类型的方法、处理、数学运算或技术。更具体地，这样的术语可以包括诸如BPEL的解释规则或规则语言，其中逻辑不是硬编码的而是在可以被读、解释、编译和执行的规则文件中表示。
本文中使用的术语“模块”或“工具”是指任何已知的或以后发 展的硬件、软件、固件、人工智能、模糊逻辑或能够执行与该元件相关的功能的硬件和软件的组合。另外，虽然用示例性实施方式来描述本发明，但应当理解本发明的各方面可以单独要求保护。
需要说明的是，在本文中，诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者终端设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者终端设备所固有的要素。在没有更多限制的情况下，由语句“包括……”或“包含……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者终端设备中还存在另外的要素。此外，在本文中，“大于”、“小于”、“超过”等理解为不包括本数；“以上”、“以下”、“以内”等理解为包括本数。
尽管已经对上述各实施例进行了描述，但本领域内的技术人员一旦得知了基本创造性概念，则可对这些实施例做出另外的变更和修改，所以以上所述仅为本发明的实施例，并非因此限制本发明的专利保护范围，凡是利用本发明说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本发明的专利保护范围之内。

Claims (8)

  1. 基于手机数据的车辆类型识别方法，其特征在于，包括步骤：
    获取第一预设时间段内的用户的手机在基站中记录的移动轨迹数据；
    根据移动轨迹数据判断搭乘同一车辆的手机用户，获得车辆轨迹和车辆轨迹对应的乘车用户数；
    根据车辆轨迹获得车辆起讫点，结合地理数据获得车辆轨迹的起讫点类型；
    根据车辆轨迹获得车辆行驶数据；
    根据车辆轨迹获得车辆停留的服务区数据；
    至少根据车辆轨迹对应的乘车用户数、起讫点类型、车辆的行驶速度数据和车辆停留的服务区数据构建特征向量；
    用车辆分类器处理特征向量，获得车辆类型识别结果；
    车辆分类器是用机器学习方法和样本的特征向量训练获得的车辆分类器。
  2. 根据权利要求1所述的方法，其特征在于，所述根据移动轨迹数据判断搭乘同一车辆的手机用户，获得车辆轨迹和车辆轨迹对应的乘车用户数，包括：
    若移动轨迹在经过第一地理位置和第二地理位置的时间重合度大于第一时间重合度，则互相之间具有交集；根据不同的第一地理位置和第二地理位置，逐一移动轨迹数据分为各集合；
    逐一计算各集合中各移动轨迹之间的匹配度，若匹配则这个移动轨迹属于同一车辆上的乘客，进而确定车辆轨迹和车辆轨迹对应的用户数。
  3. 根据权利要求1所述的方法，其特征在于，所述起讫点类型包括住宅区、工业区、商业区、景点、娱乐区和汽车站。
  4. 根据权利要求1所述的方法，其特征在于，所述根据车辆轨迹获得车辆行驶数据，包括：
    根据所有的车辆轨迹计算速度小于第一速度V un的时间区间
    Figure PCTCN2018080945-appb-100001
    Figure PCTCN2018080945-appb-100002
    的集合，
    计算某一时间区间
    Figure PCTCN2018080945-appb-100003
    内，在第三地理位置重合的车辆轨迹的数量；当在第三地理位置重合的车辆轨迹的数量大于堵车阈值时，判断该段时间内堵车，获得堵车子轨迹；第三位置不包括服务区；
    根据服务区的地理位置信息，计算车辆轨迹中处于服务区且车速小于第二速度V sa 的服务区子轨迹，
    根据扣除服务区子轨迹和堵车子轨迹后的车辆轨迹，计算车辆的包括平均速度、最大速度、速度标准差、速度分布中的一种或多种行驶数据。
  5. 根据权利要求1所述的方法，其特征在于，所述服务区数据包括：服务区停留次数、服务区停留总时间、平均在每个服务区停留的时间。
  6. 根据权利要求1所述的方法，其特征在于，所述机器学习方法为随机森林方法。
  7. 一种计算机可读存储介质，其上存储有计算机程序，其特征在于，该程序被处理器执行时实现如权利要求1至6任一所述方法的步骤。
  8. 一种计算机设备，包括存储器、处理器以及存储在所述存储器上并可在所述处理器上执行的计算机程序，其特征在于，所述处理器执行所述程序时实现如权利要求1至6任一所述方法的步骤。
PCT/CN2018/080945 2017-06-29 2018-03-28 基于手机数据的车辆类型识别方法和设备 WO2019001044A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US16/322,095 US10573173B2 (en) 2017-06-29 2018-03-28 Vehicle type identification method and device based on mobile phone data

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710517910.0A CN107463940B (zh) 2017-06-29 2017-06-29 基于手机数据的车辆类型识别方法和设备
CN201710517910.0 2017-06-29

Publications (1)

Publication Number Publication Date
WO2019001044A1 true WO2019001044A1 (zh) 2019-01-03

Family

ID=60544063

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/080945 WO2019001044A1 (zh) 2017-06-29 2018-03-28 基于手机数据的车辆类型识别方法和设备

Country Status (3)

Country Link
US (1) US10573173B2 (zh)
CN (1) CN107463940B (zh)
WO (1) WO2019001044A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109961106A (zh) * 2019-04-18 2019-07-02 北京百度网讯科技有限公司 轨迹分类模型的训练方法和装置、电子设备
CN111859112A (zh) * 2020-06-16 2020-10-30 北京嘀嘀无限科技发展有限公司 消息推送方法、装置及服务器
CN112905578A (zh) * 2021-03-03 2021-06-04 西南交通大学 一种货车gps轨迹停留点识别方法
CN113762315A (zh) * 2021-02-04 2021-12-07 北京京东振世信息技术有限公司 图像检测方法、装置、电子设备和计算机可读介质

Families Citing this family (28)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB201503855D0 (en) * 2015-03-06 2015-04-22 Q Free Asa Vehicle detection
CN107463940B (zh) * 2017-06-29 2020-02-21 清华大学 基于手机数据的车辆类型识别方法和设备
CN110164134B (zh) * 2018-02-12 2021-08-13 华为技术有限公司 信息处理的方法和装置
CN108346289A (zh) * 2018-02-13 2018-07-31 重庆交通大学 一种高速公路人车关联系统及方法
US20190311289A1 (en) 2018-04-09 2019-10-10 Cambridge Mobile Telematics Inc. Vehicle classification based on telematics data
CN110517500B (zh) * 2018-05-21 2021-04-13 上海大唐移动通信设备有限公司 一种人车关联处理方法及装置
KR102656655B1 (ko) * 2018-08-16 2024-04-12 삼성전자 주식회사 외부 전자 장치의 이동 방향에 기반하여 동작을 수행하는 전자 장치 및 전자 장치의 동작 방법
CN111125167B (zh) * 2018-10-31 2022-08-23 腾讯科技（深圳）有限公司 一种车辆匹配方法、装置及计算设备
CN110196891B (zh) * 2018-11-15 2024-03-15 腾讯大地通途（北京）科技有限公司 街区类型的确定方法、装置、存储介质及电子装置
CN111326016A (zh) * 2018-12-14 2020-06-23 哈尔滨工业大学 基于车辆定位信息的高速公路驾驶安全监测系统及方法
CN111723835A (zh) * 2019-03-21 2020-09-29 北京嘀嘀无限科技发展有限公司 车辆移动轨迹区分方法、装置和电子设备
CN110505583B (zh) * 2019-07-23 2021-01-22 中山大学 一种基于卡口数据与信令数据的轨迹匹配方法
CN110418291B (zh) * 2019-07-30 2021-07-20 中国联合网络通信集团有限公司 乘车站点识别方法、售票管理系统及电信运营商平台
CN110517495A (zh) * 2019-09-05 2019-11-29 四川东方网力科技有限公司 车辆轨迹类别的确认方法、装置、设备和存储介质
CN110852354A (zh) * 2019-10-22 2020-02-28 上海中旖能源科技有限公司 车辆轨迹点识别方法及设备
CN110969861B (zh) * 2019-12-20 2022-10-14 中国移动通信集团黑龙江有限公司 一种车辆识别方法、装置、设备及计算机存储介质
CN111159254B (zh) * 2019-12-30 2023-07-25 武汉长江通信产业集团股份有限公司 一种基于大数据处理的车辆与人员的关联方法
CN111521191A (zh) * 2020-04-20 2020-08-11 中国农业科学院农业信息研究所 一种基于信令数据的移动电话用户移动路径地图匹配方法
CN111757270A (zh) * 2020-06-19 2020-10-09 中国联合网络通信集团有限公司 超员车辆识别方法及装置
CN112542043B (zh) * 2020-12-01 2021-10-26 江苏欣网视讯软件技术有限公司 基于手机信令以及大数据分析识别公交线网覆盖盲区的方法与系统
CN112634489B (zh) * 2020-12-09 2022-12-09 众安在线财产保险股份有限公司 一种基于移动终端的车辆状态确定方法、装置及系统
US11482099B2 (en) 2021-02-02 2022-10-25 Here Global B.V. Method and apparatus for preventing traffic over-reporting via identifying misleading probe data
CN113282638B (zh) * 2021-04-23 2024-05-07 中寰卫星导航通信有限公司 一种城建用车识别方法和装置
CN114297323B (zh) * 2021-08-31 2023-05-09 北京九栖科技有限责任公司 一种一机多号识别方法、装置及其存储介质
CN114466328B (zh) * 2022-04-13 2022-06-28 北京融信数联科技有限公司 渣土车轨迹还原方法、系统和可读存储介质
EP4322131A1 (de) * 2022-08-12 2024-02-14 iunera GmbH & Co. KG Eine erfindung als system, verfahren oder anordnungen zum zuordnen von fahrdaten
CN116434529B (zh) * 2022-12-12 2023-10-24 交通运输部规划研究院 城际公路货运特征分析方法、装置和电子设备
CN116092037B (zh) * 2023-02-13 2023-07-28 长沙理工大学 融合轨迹空间-语义特征的车辆类型识别方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101136141A (zh) * 2007-10-12 2008-03-05 清华大学 基于单频连续波雷达的车型分类方法
CN104008576A (zh) * 2013-07-10 2014-08-27 易通星云（北京）科技发展有限公司 基于北斗的高速公路车辆自由流电子收费方法、系统及装置
CN107463940A (zh) * 2017-06-29 2017-12-12 清华大学 基于手机数据的车辆类型识别方法和设备

Family Cites Families (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7469827B2 (en) * 2005-11-17 2008-12-30 Google Inc. Vehicle information systems and methods
US9261376B2 (en) * 2010-02-24 2016-02-16 Microsoft Technology Licensing, Llc Route computation based on route-oriented vehicle trajectories
CN103918247B (zh) * 2011-09-23 2016-08-24 数字标记公司 基于背景环境的智能手机传感器逻辑
CN102496280B (zh) * 2011-12-13 2014-04-23 北京航空航天大学 一种路况信息实时获取方法
US10270642B2 (en) * 2012-12-05 2019-04-23 Origin Wireless, Inc. Method, apparatus, and system for object tracking and navigation
US9801027B2 (en) * 2013-07-26 2017-10-24 Anagog Ltd. Associating external devices to vehicles and usage of said association
US9859998B2 (en) * 2015-08-20 2018-01-02 Samsung Electronics Co., Ltd. Apparatus and method for identifying and localizing vehicle occupant and on-demand personalization
US9763055B2 (en) * 2014-08-26 2017-09-12 Regents Of The University Of Minnesota Travel and activity capturing
US9451077B2 (en) * 2015-02-24 2016-09-20 Patrick Duvaut Device-based safeguard systems and methods for enhancing driver safety
US9786177B2 (en) * 2015-04-10 2017-10-10 Honda Motor Co., Ltd. Pedestrian path predictions
US9558664B1 (en) * 2015-08-13 2017-01-31 Here Global B.V. Method and apparatus for providing parking availability detection based on vehicle trajectory information
US9818239B2 (en) * 2015-08-20 2017-11-14 Zendrive, Inc. Method for smartphone-based accident detection
CN105788263B (zh) * 2016-04-27 2017-12-26 大连理工大学 一种通过手机信息预测道路拥堵的方法
CN106197458B (zh) * 2016-08-10 2018-11-13 重庆邮电大学 一种基于手机信令数据和导航路线数据的手机用户出行方式识别方法
CN106314438B (zh) * 2016-08-15 2018-09-25 西北工业大学 一种司机驾驶轨迹中异常轨迹的检测方法和系统
US10319224B2 (en) * 2016-08-19 2019-06-11 Veniam, Inc. Adaptive road management in the network of moving things
WO2018099574A1 (en) * 2016-12-02 2018-06-07 Fleetmatics Ireland Limited System and method for determining a vehicle classification from gps tracks.
CN106781479B (zh) * 2016-12-23 2019-03-22 重庆邮电大学 一种基于手机信令数据实时获取高速公路运行状态的方法
US9900747B1 (en) * 2017-05-16 2018-02-20 Cambridge Mobile Telematics, Inc. Using telematics data to identify a type of a trip

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101136141A (zh) * 2007-10-12 2008-03-05 清华大学 基于单频连续波雷达的车型分类方法
CN104008576A (zh) * 2013-07-10 2014-08-27 易通星云（北京）科技发展有限公司 基于北斗的高速公路车辆自由流电子收费方法、系统及装置
CN107463940A (zh) * 2017-06-29 2017-12-12 清华大学 基于手机数据的车辆类型识别方法和设备

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
BAI, YU: "Identifying Travel Mode Using GPS Based Travel Survey Data", SCIENCE -ENGINEERING (B), CHINA MASTER'S THESES FULL - TEXT DATABASE, 15 September 2016 (2016-09-15), ISSN: 1674-0246 *
SONG, LU: "Research on Traffic Origin-Destination Distribution Based on Cell Phone Data", SCIENCE -ENGINEERING (B), CHINA MASTER'S THESES FULL - TEXT DATABASE, 15 August 2016 (2016-08-15), ISSN: 1674-0246 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109961106A (zh) * 2019-04-18 2019-07-02 北京百度网讯科技有限公司 轨迹分类模型的训练方法和装置、电子设备
CN111859112A (zh) * 2020-06-16 2020-10-30 北京嘀嘀无限科技发展有限公司 消息推送方法、装置及服务器
CN113762315A (zh) * 2021-02-04 2021-12-07 北京京东振世信息技术有限公司 图像检测方法、装置、电子设备和计算机可读介质
CN112905578A (zh) * 2021-03-03 2021-06-04 西南交通大学 一种货车gps轨迹停留点识别方法

Also Published As

Publication number Publication date
CN107463940B (zh) 2020-02-21
US20190180610A1 (en) 2019-06-13
CN107463940A (zh) 2017-12-12
US10573173B2 (en) 2020-02-25

Similar Documents

Publication Publication Date Title
WO2019001044A1 (zh) 基于手机数据的车辆类型识别方法和设备
WO2021237812A1 (zh) 一种基于手机信令数据且含个人属性修正的城市出行方式综合识别方法
CN111862606B (zh) 一种基于多源数据的非法营运车辆识别方法
CN107103775B (zh) 一种基于群智计算的道路质量检测方法
WO2019001045A1 (zh) 基于手机数据的城际交通出行方式判断方法和设备
WO2019085807A1 (zh) 一种路况信息获取方法及其设备、存储介质
CN106197458A (zh) 一种基于手机信令数据和导航路线数据的手机用户出行方式识别方法
CN108761509B (zh) 一种基于历史数据的汽车行驶轨迹及里程预测方法
CN111710170B (zh) 一种高速路口辅助进行温度检测的方法及装置
CN103700174A (zh) 一种基于wifi身份识别的公交客流数据采集及od分析方法
CN104020751A (zh) 基于物联网的校园安全监测系统及方法
CN108806248B (zh) 一种针对rfid电子车牌数据的车辆出行轨迹划分方法
CN104318781B (zh) 基于rfid技术的行程速度获取方法
CN111653096A (zh) 一种基于手机信令数据的城市出行方式识别方法
WO2012024976A1 (zh) 一种交通信息处理方法及装置
CN108848460A (zh) 基于rfid和gps数据的人车关联方法
WO2023123616A1 (zh) 一种快速路车辆od位置提取方法及系统
CN112036757A (zh) 基于手机信令和浮动车数据的停车换乘停车场的选址方法
CN111369824B (zh) 一种基于图像识别定位的引导泊车方法及系统
CN109489679A (zh) 一种导航路径中的到达时间计算方法
CN106297301A (zh) 一种车辆套牌检查方法及使用该方法的车辆套牌检测系统
US20210397187A1 (en) Method and system for operating a mobile robot
CN111105619A (zh) 一种路侧逆向停车的判断方法及装置
WO2024098992A1 (zh) 倒车检测方法及装置
CN104616495A (zh) 一种基于公交gps探测车的城市交通状态识别方法

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: 18825515

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: 18825515

Country of ref document: EP

Kind code of ref document: A1