CN101930668B - Processing method of road traffic OD (Optical Density) information collection system for license plate recognition - Google Patents

Processing method of road traffic OD (Optical Density) information collection system for license plate recognition Download PDF

Info

Publication number
CN101930668B
CN101930668B CN2009100502115A CN200910050211A CN101930668B CN 101930668 B CN101930668 B CN 101930668B CN 2009100502115 A CN2009100502115 A CN 2009100502115A CN 200910050211 A CN200910050211 A CN 200910050211A CN 101930668 B CN101930668 B CN 101930668B
Authority
CN
China
Prior art keywords
od
license plate
information
vehicle
subsystem
Prior art date
Application number
CN2009100502115A
Other languages
Chinese (zh)
Other versions
CN101930668A (en
Inventor
林瑜
沈峰
沈建惠
肖永来
陈平
陈红洁
Original Assignee
上海电器科学研究所(集团)有限公司
上海电科智能系统股份有限公司
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 上海电器科学研究所(集团)有限公司, 上海电科智能系统股份有限公司 filed Critical 上海电器科学研究所(集团)有限公司
Priority to CN2009100502115A priority Critical patent/CN101930668B/en
Publication of CN101930668A publication Critical patent/CN101930668A/en
Application granted granted Critical
Publication of CN101930668B publication Critical patent/CN101930668B/en

Links

Abstract

The invention relates to a road traffic OD information collection system for license plate recognition data, which comprises three subsystems of a roadside data collection subsystem, a roadside communication subsystem and a license plate recognition collection and processing platform subsystem; high-speed communication, transmission and data invocation between the data collection subsystem and the license plate recognition subsystem are finished by utilizing the roadside communication subsystem. The processing method comprises the following working steps from step 1 to step 7: storing and uploading the license plate recognition information of an external field collection section; secondary cleaning and secondary recognition; tracing and updating vehicle routes in real time; processing single-vehicle track tracing and revising single-vehicle OD errors caused by errors generated during the license plate information collection and transmission; counting the number of specific vehicles with the same types, the same starting points and destinations and in the same period when passing through the destinations; classifying and storing the revised OD values according to the information ofthe statistics periods, the vehicle types, the destinations and the like; and displaying an OD distribution matrix in forms of reports and graphs.

Description

车牌识别的道路交通OD信息采集系统的处理方法 OD approach road traffic information collection system, license plate recognition

技术领域 FIELD

[0001] 本发明涉及一种城市交通信息采集系统及其处理方法,特别是一种车牌识别数据的道路交通OD信息采集系统及其处理方法,属于智能交通技术领域。 [0001] The present invention relates to an urban transport system and information collection processing method, road traffic information collection system OD and more particularly to a method of processing data in license plate recognition, intelligent transportation technical field belongs.

背景技术 Background technique

[0002]自进入21世纪以来,城市道路建设、机动车辆保有量增加均进入了快速发展阶段。 [0002] Since entering the 21st century, urban road construction, motor vehicle ownership increases have entered a stage of rapid development. 与此同时,通过城市交通信息采集系统的建设,在道路交通管理、支持道路交通信息应用和服务等方面,取得了较好的社会、经济和环境效益。 At the same time, through the construction of urban traffic information collection system, road traffic management, road traffic information applications and support services, etc., and achieved good social, economic and environmental benefits. 但是,已有的道路交通信息采集系统,随着覆盖地域的扩张和不同区域道路网的交通流特征的不同,也逐步呈现出其内在的不足,或表现在所获取的交通信息内容不够、或表现在获取交通信息的成本过高等缺点。 However, the existing road traffic information collection system, with different flow characteristics of traffic and the expansion of geographical coverage of the different areas of the road network, but also gradually showing its inherent shortcomings, or the performance of the traffic information acquired content is not enough, or reflected in the high cost of acquiring traffic information and other shortcomings.

[0003] 道路交通0D(0rigin Destination),是指道路起终点之间的出行交通量分布,其结果通常用二维表格或矩阵来表示。 [0003] Road Traffic 0D (0rigin Destination), it refers to the amount of traffic from road to travel between the end of the distribution, the result is usually a two-dimensional table or matrix to represent. 路网交通OD分布信息采集,能够完善城市交通模型,为交通战略研究、交通规划、交通管理提供定量分析基础,为缓解城市紧张的交通状况提供数据支撑;也可以分析城市居民出行源强度、分布、交通结构、出行目的等基本出行特征的变化情况;能够从居民出行的角度出发,分析交通需求供给矛盾的根本原因,为交通需求管理提供理论依据;在掌握居民出行规律的基础上,构筑交通模型,为城市进行科学的交通规划服务。 Road network traffic OD distribution information collection, can improve the urban transport model for the transport strategy study, traffic planning, traffic management provides a quantitative basis for analysis, to provide data support to relieve urban traffic tension condition; can also analyze travel source intensity urban distribution change the basic characteristics of the trips of the traffic structure, trip purpose and so on; can proceed from residents travel point of view, analyze the root causes of traffic demand supply conflict, provide a theoretical basis for the transportation demand management; on the basis of master travel law residents on building traffic model for urban transport planning for scientific services.

[0004] 传统的OD信息获取的方法之一是OD抽样调查,要求所有的被调查对象提供出行时间、出行起点、出行目的地、出行方式等信息。 [0004] One of the traditional methods of information acquisition is OD OD sample survey, requires that all respondents provide travel time, travel starting point, travel destination information, travel mode and so on. 在专利申请号为200710024832. 7,公开号为CN101127040的OD信息获取方法中,采用问卷表单的形式对公路网上的部分车辆出行信息进行抽样并推算全样本的OD分布矩阵,但是人工抽样调查需要耗费大量的人力、物力、财力,频繁进行大规模的交通调查显然不可行,并且由于基础建设和交通结均的经常变化,因此其获取的OD矩阵的可信度也较低;在专利申请号为200610118461. 4,公开号为CN1959759的OD分析方法中,利用GPS车辆定位系统对出租车群体的出行轨迹进行记录并生成出租车的OD分布矩阵,但是其获取的OD矩阵仅仅是具备GPS定位装置的出租车这单独一个群体,具有较大的局限性;在专利申请号为200810038365. 8,公开号为CN101295326的OD推算方法中,利用少量抽样的出租车GPS定位数据计算车辆OD出行分布的概率并应用交通仿真手段对路网上的实际交通流量进行随机分配 In Patent Application No. 200710024832.7, Publication No. CN101127040 OD information acquisition method, in the form of a questionnaire form part of the travel information of the vehicle on the road network and sampling the full sample distribution matrix calculated OD, it takes a sample of artificial a lot of manpower, material and financial resources, frequent large-scale traffic survey is clearly not feasible, and because of infrastructure and traffic junctions are frequently change, so their credibility acquired OD matrix is ​​low; in the patent application No. 200610118461.4, Publication No. CN1959759 OD analytical method, the vehicle positioning system using GPS travel track groups recorded and taxi cab OD generate distribution matrix, but the matrix is ​​only acquired OD GPS positioning device comprising this taxi single group, with great limitations; OD estimation method in Patent application No. 200810038365.8, Publication No. CN101295326, the probability of a small sample using GPS location data calculating taxi vehicle OD trip and LO actual application traffic simulation tools were randomly assigned traffic online 算全样本的OD分布矩阵,但是出租车的出行分布具有群体的特殊性,对路网整体的交通出行行为不具有代表性。 OD sample distribution matrix full count, but the taxi trip distribution groups have the particularity of the road network traffic overall travel behavior is not representative.

[0005] 另一种OD信息采集的方法是OD反推,通过路段交通流数据,根据各种模型算法,反推交通OD分布矩阵。 [0005] Another method of information collection OD OD matrix is, road traffic stream data, according to various model algorithm, backstepping OD Distribution Matrix. 在专利申请号为200810084396. 7,公开号为CN101261768的路网OD推算方法中,利用公路网的历史OD数据和路段实际流量观测值推算得到当前的OD分布矩阵,由于路段流量检测存在的误差以及OD反推模型在算法假设和参数标定方面存在的局限性,往往导致OD反推的结果与真实情况存在较大偏差。 In Patent Application No. 200810084396.7, Publication No. CN101261768 OD estimating method of the road network, the obtained current distribution matrix OD OD using historical data and the observation value calculated actual flow sections of the road network, due to the presence of an error detection, and link flow OD estimation model has limitations in terms of calibration parameters and the algorithm assumes, often resulting in a big deviation of OD results with the real situation.

发明内容:[0006] 本发明的目的在于提供一种车牌识别数据的道路交通OD信息采集系统及其处理方法,该系统及其处理方法,能够自动采集城市道路交通网络长期、动态、高质量的OD信息,为交通需求管理提供理论依据;并掌握居民出行规律的基础上,构筑交通模型,为城市进行科学的交通规划服务。 SUMMARY OF THE INVENTION: [0006] The object of the present invention is to provide a license plate recognition data OD road traffic information collection system and processing method, the system and its processing method, it is possible to automatically collect long urban road network, dynamic quality. OD information, provide a theoretical basis for the transportation demand management; and master the basic law of the residents travel on to build a traffic model for the city's transportation planning for scientific services.

[0007] 为达到上述目的,本发明的技术方案是,所述车牌识别数据的道路交通OD信息采集系统及其处理方法,包括路边数据采集子系统、路边通信子系统、车牌识别采集处理平台子系统;所述的路边数据采集子系统,又包含布设在外场的摄像机、交通参数检测单元、工业控制处理计算机、龙门架和支架、供电设备等,承担车辆与牌照图片的捕获、牌照识别处理、交通参数检测、原始数据存储管理;所述的车牌识别采集处理平台子系统,又包含主干交换机、数据库服务器、应用服务器、计算机终端以及所需要的存储设备、安全设备、网络管理设备等,完成路边数据采集子系统获取各种原始数据、进行所需要的各种处理、数据存储、以及对所存储数据的管理。 [0007] To achieve the above object, the technical solution of the present invention is that the road traffic information collection system OD and license plate recognition data processing method, a data acquisition subsystem including roadside, roadside communication subsystem, license plate recognition acquisition and processing platform subsystem; data acquisition subsystem of the street, and laid outside the camera's field comprising, parameter detecting means traffic, industrial process control computer, and the support gantry, the power supply apparatus and the like, and the vehicle license plate captured bear picture, licenses recognition processing traffic parameter detection, the original data storage management; the license plate recognition subsystem acquisition and processing platform, and comprising a backbone switch, database servers, application servers, computer terminals and storage devices required, the security device, the network management equipment to complete the data acquisition subsystem acquires various roadside raw data required for various processing, data storage, and management of the stored data.

[0008] 所述的路边数据采集子系统与系统车牌识别采集处理平台子系统之间通过路边通信子系统完成高速通信,传输路边数据采集子系统所采集的各种原始数据,满足系统车牌识别采集处理平台子系统对分布式存储在各路边数据采集子系统内的各种数据的调用。 [0008] The complete high-speed data acquisition roadside communication via a roadside communication between the subsystem and the subsystem license plate recognition system acquisition and processing platform subsystem, roadside transmission data acquisition subsystem various raw data collected to meet the system license plate recognition subsystem invokes various internet acquisition and processing of data within a distributed storage subsystem data collected at each side of the road.

[0009] 另一目的是所述的车牌识别数据的道路交通OD信息采集系统及其处理方法的处理方法,通过下述的信息采集模块、预处理模块、单车轨迹跟踪模块、OD修正模块、单车OD生成模块、OD统计模块、OD分布展示模块、基础配置模块来实现,其工作步骤为: [0009] Another object of the acquisition system and processing method is a method of processing information according to road traffic OD LPR data, by the following information collection module, the preprocessing module, bicycle tracking module, OD correction module, cycling OD generation module, statistics module OD, OD distribution display module, basic configuration module to implement its work steps:

[0010] 步骤I :信息采集模块对外场采集各个断面的车牌特写图片、车辆号牌、车牌颜色、车型、经过时间、经过地点、车辆颜色等信息,并对这些车牌识别信息进行存储和上传; [0010] Step I: External field information collection module collecting plate close-up pictures of each section, the vehicle license plate, the license plate color, model, elapsed time, place and after, vehicle color, and license plate recognition which information is stored and uploaded;

[0011] 步骤2 :将步骤I的识别信息,通过预处理通过模块,对外场采集的无法识别的车牌特写图片进行二次清洗和再识别,进一步提高车牌信息采集的精度; [0011] Step 2: Step I of the identification information, for cleaning and re-identified by the second module, the external field does not recognize the license plate captured by close-up pictures pretreatment, further improve the accuracy of the information collection plate;

[0012] 步骤3 :单车轨迹跟踪模块对车牌识别数据的道路交通OD信息采集系统捕捉到的车辆路径进行实时的跟踪与更新,其最新的路径信息包括车辆的基础信息(属地、车型等)和之前出行通过的车牌识别断面的状态(时间、地点等); [0012] Step 3: Cycle tracking module road traffic OD information collection system license plate identification data captured vehicle path in real time tracking and updating the latest path information includes a vehicle base information (territorial, models, etc.) and license plate recognition state prior to travel through the section (time, location, etc.);

[0013] 单车轨迹跟踪的关键环节是将最新上传的车牌经过的断面记录与车辆之前的断面记录队列匹配比对,若车牌匹配成功则对车辆当前出行最新的路径进行更新,反之则重新生成另外一条实时更新的出行路径; Key [0013] bicycle tracking is the section before the section record vehicle latest uploads license plate after recording queue matching comparison, if the license plate matched the success of the current vehicle traveling the latest route to be updated, and vice versa regenerate another a real-time update of the travel path;

[0014] 步骤4 :0D修正模块对车牌信息采集和传输过程产生的误差造成的单车OD误差进行修正。 [0014] Step 4: 0D OD error correction module cycling license information acquisition and transmission error generated during resulting corrected. 具体的修正方法包括:(I)车牌模糊匹配。 Specific correction method comprising: (I) plate fuzzy matches. 由于车牌识别可能存在的识别误差,采用车牌完全精确匹配的方法可能会带来一定的匹配误差,造成单车轨迹跟踪过程中部分路径节点的缺失。 Since the identification plate recognition error may exist, the method entirely accurate match plate may bring some matching error, resulting in deletion of part of the path tracking cycling process node. 因此制定了相应的模糊匹配规则,在两个车牌间出现特定字符的匹配误差但满足其他匹配条件的前提下,认定该两个车牌号码匹配成功,这样可以在车牌个别字符识别错误的情况下仍然保证车辆出行路径的完整性和准确性;(2)路径节点缺失修补。 Therefore developed a corresponding fuzzy matching rules appear to match the specific character of the error between the two plates, but the premise of meeting the conditions of the other matches, finds that the two license plate number matching success, so you can still license plate character recognition errors of individual circumstances ensure the integrity and accuracy of the travel path of the vehicle; (2) repair path node deletion. 模糊匹配可以大大减少路径中节点缺失的概率,但对于超出模糊匹配规则而产生的路径节点缺失的情况,需要对路径进行修补,以保证路径的完整性。 Fuzzy matches can greatly reduce the probability of missing nodes in the path, but the path node for the case of fuzzy matching rule generated beyond missing, it is necessary to repair the path, to ensure the integrity of the path. 当单车出行路径中某些节点缺失,根据道路的连通性和上下游关系,根据最短路径的原则对丢失部分的节点进行修补并记录相关修补信息; Where a single travel path deletion certain nodes, according to the relationship between the upstream and downstream communication and road repair node missing portion based on the principle of shortest paths and associated patch records information;

[0015] 步骤5 :单车OD生成模块是在单车轨迹跟踪的基础上进行处理,当设定的时间范围内下游未出现该车牌新的出行断面信息时,可以确认该车牌的出行OD已经完成并写入单车OD的数据库;反之该车牌的继续保持轨迹跟踪,直至车牌新的出行断面记录出现并对实时轨迹进行更新; [0015] Step 5: Cycle OD generation module is processed on the basis of the tracking bicycle, a new travel when the downstream section of the license information does not appear within the set time, it was confirmed that the license has been completed and travel OD written bicycle OD database; on the contrary continue to maintain the license plate tracking, until a new travel section and real-time track record license plates appear to be updated;

[0016] 单车OD记录是OD分布计算中的最小粒度单位的信息,通过对单车出行的统计可以进一步生成小区间的OD分布。 [0016] Cycling is the information recording OD OD distributed computing unit in the minimum particle size, can further generate OD distribution between cells through cycling travel statistics. 单车OD记录存储的信息包括车辆在一次出行中经过的车牌识别断面序列、断面经过时间等信息; OD cycling recording information storing section comprises a license plate recognition sequence of a trip in the vehicle passes, a cross section through the time information;

[0017] 步骤6 :0D统计模块对对路网中符合指定车辆类型、出行起讫点相同、且经过讫点时间位于同一个周期内的单车OD记录个数进行统计,生成各个车牌采集断面之间的指定车辆类型的断面间OD分布矩阵。 [0017] Step 6: 0D statistics module to conform to the road network in the specified vehicle type, the same start and end point of travel, and after a time point located finished bicycle OD record number within the same cycle count, each generating section between the collecting plate the discontinuity of the type specified vehicle OD distribution matrix. 由于部分车辆OD的起点或终点的缺失,会导致断面间OD观测量的统计值与真实值之间存在一定偏差,通过分析OD观测量误差产生的机理,建立观测OD量与断面流量之间的校验方程组对起终点缺失引起的OD误差进行修正,并根据断面与小区的隶属关系合成小区间OD分布矩阵; Due to missing start or end portions of the OD of the vehicle, there is a certain deviation between the cause discontinuity OD statistics for the measurements and the true values, by analyzing the observables OD error generation mechanism is established between the observed and the amount of cross-sectional flow OD equations for parity errors from the end point OD due to deletion is corrected, and according to inter-OD affiliation distribution matrix section cells and cell synthesis;

[0018] 步骤7 :0D存储与展示模块将修正后的OD值按照统计周期、车辆类型和起讫点等信息分类存储,并对路网动态OD分布矩阵进行报表和图形展示。 [0018] Step 7: 0D memory module and display the corrected OD values ​​in accordance with statistical period, and vehicle type classification origin-destination information storage, and distribution of road network Dynamic OD matrix display reports and graphs.

[0019] 本发明的有益效果:所述的车牌识别数据的道路交通OD信息采集系统及其处理方法,具有以下优点=(I)OD数据种类更加丰富,实现精细化交通信息采集⑵具有更高精度和可靠性(3)平均获取成本较低(4)获取难度较低,相对于人工调查大量的前期准备、现场调查和后期数据处理,以及模型推算需要的复杂的算法设计、算法验证和调试工作,车牌识别数据的道路交通OD信息采集系统及其处理方法,在算法设计和数据处理等诸多环节都更加容易实现,具有明显的经济效益和社会效益。 [0019] Advantageous effects of the present invention are: OD road traffic information collection system and data processing method according to the license plate recognition, has the following advantages = (I) OD richer data type, to achieve fine having a higher traffic information collection ⑵ accuracy and reliability (3) the average acquisition cost less (4) get less difficult than manually investigating a spate of preparation, site investigation and post-processing data, and model projections require complex algorithm design, verification and debug algorithms work, OD road traffic information collection system and method for processing license plate recognition data, the data processing algorithm design and many other aspects are easier to implement, has obvious economic and social benefits.

[0020] 以下结合附图和实施例对本发明作比较详细的说明。 [0020] Hereinafter, the present invention will be described in more detail in conjunction with the accompanying drawings and embodiments.

附图说明: BRIEF DESCRIPTION OF:

[0021] 图I是本发明的车牌识别数据的道路交通OD信息采集系统总体构成框图; [0021] Figure I is a general road traffic information acquisition system OD license plate identification data is a block diagram of the present invention;

[0022] 图2是本发明的车牌识别数据的道路交通OD信息采集系统的处理方法流程图。 [0022] FIG. 2 is a processing method OD road traffic information collection system of the present invention LPR data flowchart.

具体实施方式 detailed description

[0023] 参照图1,这是本发明的车牌识别数据的道路交通OD信息采集系统总体构成框图。 [0023] Referring to FIG. 1, which is a block diagram showing an overall OD road traffic information data acquisition system of the present invention, the license plate recognition.

[0024] 如图所示,它包括路边数据采集子系统I、路边通信子系统2、车牌识别采集处理平台子系统3。 [0024] As shown, the data acquisition subsystem including roadside I, roadside communication subsystem 2, license plate recognition platform acquisition and processing subsystem 3.

[0025] 路边数据采集子系统1,包括布设在外场的摄像机、交通参数检测单元、工业控制处理计算机、龙门架和支架、供电设备设施等。 [0025] Roadside data acquisition subsystem 1, including laid outside the field of the camera, the traffic parameter-detecting means, the control process computer industry, gantry and bracket, power supply and other equipment and facilities. 路边数据采集子系统,承担车辆与牌照图片的捕获、牌照识别处理、交通参数检测、原始数据存储管理等功能。 Roadside data acquisition subsystem, and vehicle license picture bear capture, license plate recognition processing, traffic parameter detection, raw data storage management.

[0026] 车牌识别采集处理平台子系统3,包括主干交换机、数据库服务器、应用服务器、计算机终端以及所需要的存储设备、安全设备、网络管理设备等。 [0026] acquisition and processing platform license plate recognition subsystem 3, comprising a backbone switch, database servers, application servers, computer terminals and storage devices required for the security device, the network management equipment. 车牌识别采集处理平台子系统,完成路边数据采集子系统获取各种原始数据、进行所需要的各种处理、数据存储、以及对所存储数据的管理,其中所述的主干交换机分别连接数据库服务器和应用服务器。 License plate recognition subsystem acquisition and processing platform, completion of the data acquisition subsystem acquires various roadside raw data required for various processes, data storage, and management of the stored data, wherein said switch is connected to the database server backbone and application servers.

[0027] 所述的路边数据采集子系统I与系统车牌识别采集处理平台子系统3之间通过路边通信子系统2完成高速通信,传输路边数据采集子系统所采集的各种原始数据,满足系统车牌识别采集处理平台子系统对分布式存储在各路边数据采集子系统内的各种数据的调用。 [0027] The data acquisition subsystem I roadside license plate recognition system acquisition and processing subsystem internet roadside communication subsystem 2 to complete high-speed communication, data transmission between 3 roadside collection various raw data acquisition subsystem , license plate recognition system to meet the acquisition and processing subsystem of the distributed memory platform invokes various data in the respective roadside data acquisition subsystem.

[0028] 本发明的车牌识别数据的道路交通OD信息采集系统所采用模块构成。 [0028] OD road traffic information data acquisition system of the present invention, the license plate recognition modules employed.

[0029] 信息采集模块4、预处理模块5、单车轨迹跟踪模块6、OD修正模块7、单车OD生成模块8、OD统计模块9、OD分布展示模块10、基础配置模块11 ; [0029] The information collection module 4, the preprocessing module 5, cycling tracking module. 6, 7 OD correction module, cycling. 8 OD generation module, a statistics module. 9 OD, OD distribution display module 10, basic configuration module 11;

[0030] 本发明信息采集系统所采用处理方法是通过下述模块来实现的。 The method of processing is realized by the following modules [0030] The information collection system of the present invention is employed. 它包括信息采集模块4、预处理模块5、单车轨迹跟踪模块6、OD修正模块7、单车OD生成模块8、OD统计模块9、OD分布展示模块10、基础配置模块11)。 It includes information collection module 4, the preprocessing module 5, cycling tracking module. 6, 7 OD correction module, cycling. 8 OD generation module, a statistics module. 9 OD, OD distribution display module 10, basic configuration module 11).

[0031] 参照图2,这是本发明的车牌识别数据的道路交通OD信息采集系统处理方法流程图。 [0031] Referring to FIG 2, which is a road traffic information collection OD license plate recognition system, data processing method of the present invention. FIG. 如图所示,本发明的工作步骤为: As shown, the work steps of the present invention are:

[0032] 步骤I :信息采集模块对外场采集各个断面的车牌特写图片、车辆号牌、车牌颜色、车型、经过时间、经过地点等车牌识别信息进行存储并上传; [0032] Step I: External field information collection module collecting plate close-up pictures of each section, the vehicle license plate, the license plate color, model, elapsed time, place and the like through the identification plate and upload stores;

[0033] 步骤2 :预处理通过模块对外场采集的无法识别的车牌特写图片进行二次清洗和再识别,进一步提高车牌信息采集的精度; [0033] Step 2: preprocessing module by the external field plate unrecognized capture close-up images and re-washed secondary identification, further improve the accuracy of the information collection plate;

[0034] 步骤3 :单车轨迹跟踪模块对车牌识别数据的道路交通OD信息采集系统捕捉到的车辆路径进行实时的跟踪与更新,其最新的路径信息包括车辆的基础信息(属地、车型等)和之前出行通过的车牌识别断面的状态(时间、地点等); [0034] Step 3: Cycle tracking module road traffic OD information collection system license plate identification data captured vehicle path in real time tracking and updating the latest path information includes a vehicle base information (territorial, models, etc.) and license plate recognition state prior to travel through the section (time, location, etc.);

[0035] 单车轨迹跟踪的关键环节是将最新上传的车牌经过的断面记录与车辆之前的断面记录队列匹配比对,若车牌匹配成功则对车辆当前出行最新的路径进行更新,反之则重新生成另外一条实时更新的出行路径; Key [0035] bicycle tracking is the section before the section record vehicle latest uploads license plate after recording queue matching comparison, if the license plate matched the success of the current vehicle traveling the latest route to be updated, and vice versa regenerate another a real-time update of the travel path;

[0036] 步骤4 :0D修正模块对车牌信息采集和传输过程产生的误差造成的单车OD误差进行修正。 [0036] Step 4: 0D OD error correction module cycling license information acquisition and transmission error generated during resulting corrected. 具体的修正方法包括:(I)车牌模糊匹配。 Specific correction method comprising: (I) plate fuzzy matches. 由于车牌识别可能存在的识别误差,采用车牌完全精确匹配的方法可能会带来一定的匹配误差,造成单车轨迹跟踪过程中部分路径节点的缺失。 Since the identification plate recognition error may exist, the method entirely accurate match plate may bring some matching error, resulting in deletion of part of the path tracking cycling process node. 因此制定了相应的模糊匹配规则,在两个车牌间出现特定字符的匹配误差但满足其他匹配条件的前提下,认定该两个车牌号码匹配成功,这样可以在车牌个别字符识别错误的情况下仍然保证车辆出行路径的完整性和准确性;(2)路径节点缺失修补。 Therefore developed a corresponding fuzzy matching rules appear to match the specific character of the error between the two plates, but the premise of meeting the conditions of the other matches, finds that the two license plate number matching success, so you can still license plate character recognition errors of individual circumstances ensure the integrity and accuracy of the travel path of the vehicle; (2) repair path node deletion. 模糊匹配可以大大减少路径中节点缺失的概率,但对于超出模糊匹配规则而产生的路径节点缺失的情况,需要对路径进行修补,以保证路径的完整性。 Fuzzy matches can greatly reduce the probability of missing nodes in the path, but the path node for the case of fuzzy matching rule generated beyond missing, it is necessary to repair the path, to ensure the integrity of the path. 当单车出行路径中某些节点缺失,根据道路的连通性和上下游关系,根据最短路径的原则对丢失部分的节点进行修补并记录相关修补信息; Where a single travel path deletion certain nodes, according to the relationship between the upstream and downstream communication and road repair node missing portion based on the principle of shortest paths and associated patch records information;

[0037] 步骤5 :单车OD生成模块是在单车轨迹跟踪的基础上进行处理,当设定的时间范围内下游未出现该车牌新的出行断面信息时,可以确认该车牌的出行OD已经完成并写入单车OD的数据库;反之该车牌的继续保持轨迹跟踪,直至车牌新的出行断面记录出现并对实时轨迹进行更新; [0037] Step 5: Cycle OD generation module is processed on the basis of the tracking bicycle, a new travel when the downstream section of the license information does not appear within the set time, it was confirmed that the license has been completed and travel OD written bicycle OD database; on the contrary continue to maintain the license plate tracking, until a new travel section and real-time track record license plates appear to be updated;

[0038] 单车OD记录是OD分布计算中的最小粒度单位的信息,通过对单车出行的统计可以进一步生成小区间的OD分布。 [0038] Cycling is the information recording OD OD distributed computing unit in the minimum particle size, can further generate OD distribution between cells through cycling travel statistics. 单车OD记录存储的信息包括车辆在一次出行中经过的车牌识别断面序列、断面经过时间等信息; OD cycling recording information storing section comprises a license plate recognition sequence of a trip in the vehicle passes, a cross section through the time information;

[0039] 步骤6 :0D统计模块对对路网中符合指定车辆类型、出行起讫点相同、且经过讫点时间位于同一个周期内的单车OD记录个数进行统计,生成各个车牌采集断面之间的指定车辆类型的断面间OD分布矩阵。 [0039] Step 6: 0D statistics module to conform to the road network in the specified vehicle type, the same start and end point of travel, and after a time point located finished bicycle OD record number within the same cycle count, each generating section between the collecting plate the discontinuity of the type specified vehicle OD distribution matrix. 由于部分车辆OD的起点或终点的缺失,会导致断面间OD观测量的统计值与真实值之间存在一定偏差,通过分析OD观测量误差产生的机理,建立观测OD量与断面流量之间的校验方程组对起终点缺失引起的OD误差进行修正,并根据断面与小区的隶属关系合成小区间OD分布矩阵; [0040] 步骤7 :0D存储与展示模块将修正后的OD值按照统计周期、车辆类型和起讫点等信息分类存储,并对路网动态OD分布矩阵进行报表和图形展示。 Due to missing start or end portions of the OD of the vehicle, there is a certain deviation between the cause discontinuity OD statistics for the measurements and the true values, by analyzing the observables OD error generation mechanism is established between the observed and the amount of cross-sectional flow OD OD parity equations for error due to deletion from the end of the correction, and inter-OD affiliation distribution matrix according to the cell a synthetic cell section; [0040] step 7: 0D memory and display module according to the corrected OD value statistics period , vehicle type and origin-destination information classified storage, distribution and road network dynamic OD matrix reporting and graphical display.

[0041] 虽然本发明已参照上述的实施例来描述,但是本技术领域中的普通技术人员,应当认识到以上的实施例仅是用来说明本发明,应理解其中可作各种变化和修改而在广义上没有脱离本发明,所以并非作为对本发明的限定,只要在本发明的实质精神范围内,对以上所述的实施例的变化、变形都将落入本发明权利要求的保护范围。 [0041] The embodiments of the present invention has been described with reference to While, those skilled in the art of ordinary skill in the art, it will be appreciated that the above embodiments are merely to illustrate the invention, which should be understood that various changes and modifications may be without departing from the present invention in a broad sense, the present invention is not limited as long as within the true spirit of the invention, a variation of the above embodiment, the modification will fall scope of the claims of the invention.

Claims (4)

1. 一种车牌识别OD信息采集系统的处理方法,该车牌识别OD信息采集系统包括路边数据采集子系统(I)、路边通信子系统(2)、车牌识别采集处理平台子系统(3);所述的路边数据采集子系统(I)与系统车牌识别采集处理平台子系统(3)之间,通过路边通信子系统(2)完成高速通信,传输路边数据采集子系统所采集的各种原始数据,满足于系统车牌识别采集处理平台子系统对分布式存储在各路边数据采集子系统内的各种数据的调用;所述的路边数据采集子系统(I),承担车辆与牌照图片的捕获、牌照识别处理、交通参数检测、原始数据存储管理,它包括设在外场的摄像机、交通参数检测单元、工业控制处理计算机、龙门架和支架、供电设备;所述的车牌识别采集处理平台子系统(3),是完成路边数据采集子系统获取各种原始数据、进行所需要的各种处理、数 1. A method of processing information collection OD license plate recognition system, the license plate recognition OD roadside information gathering system includes a data acquisition subsystem (the I), the roadside communication subsystem (2), acquisition and processing platform license plate recognition subsystem (3 ); roadside between the data acquisition subsystem (I) and license plate recognition system acquisition and processing platform subsystem (3), high-speed communication is completed, the transmission data acquisition by the roadside roadside communication subsystem (2) subsystem various raw data collection, to meet the license plate recognition system acquisition and processing subsystem internet call various data stored in the distributed subsystems each side of the road data acquisition; roadside said data acquisition subsystem (the I), assumed vehicle license plate image capture, license plate recognition process, traffic parameter detection, the original data storage management, which includes a camera provided on the outer fields, the traffic parameter-detecting means, the control process computer industry, gantry and bracket, power supply apparatus; the license plate recognition subsystem acquisition and processing platform (3), to complete the data acquisition subsystem acquires various roadside raw data, performs various processing required, the number of 据存储、以及对所存储数据的管理,它包括主干交换机、数据库服务器、应用服务器、计算机终端以及所需要的存储设备、安全设备、网络管理设备;所述的信息采集系统,包括信息采集模块(4)、预处理模块(5)、单车轨迹跟踪模块(6)、OD修正模块(7)、单车OD生成模块(8)、OD统计模块(9)、OD分布展示模块(10)、基础配置模块(11),其特征在于:所述的处理方法,其工作步骤为: 步骤1,信息采集模块,对外场采集各个断面的车牌特写图片、车辆号牌、车牌颜色、车型、经过时间、经过地点车牌识别信息进行存储并上传; 步骤2,预处理通过模块,对外场采集的无法识别的车牌特写图片进行二次清洗和再识别; 步骤3,单车轨迹跟踪模块,对车牌识别数据的交通OD信息采集系统捕捉到的车辆路径进行实时的跟踪与更新,其最新的路径信息包括车辆的 Data storage, and management of stored data, comprising the backbone switches, database servers, application servers, computer terminals and storage devices required for the security device, the network management device; the information collection system, includes an information collection module ( 4), the preprocessing module (5), cycling tracking module (. 6), OD correction module (7), cycling OD generation module (. 8), statistics module OD (. 9), the distribution of OD display module (10), basic configuration module (11), characterized in that: the processing method, the work steps: step 1, the information collection module, the external field capture close-up images of each section of the license plate, vehicle license plate, the license plate color, model, elapsed time, elapsed Location identification information storage plate and uploading; step 2, and then the second cleaning pretreatment is performed through the identification module, the external field does not recognize the license plate acquired a close-up image; step 3, bicycle tracking module, the traffic data OD to the license plate recognition information collection system to capture real-time vehicle tracking and route updates, the latest path information includes vehicle 地、车型基础信息和之前出行通过的车牌识别断面的时间、地点状态信息; 步骤4,OD修正模块,对车牌信息采集和传输过程产生的误差造成的将一次OD误识为多次OD和OD起终点识别错误进行修正;具体的修正方法包括:(I)车牌模糊匹配:由于车牌识别可能存在的识别误差,采用车牌完全精确匹配的方法可能会带来一定的匹配误差,造成单车轨迹跟踪过程中部分路径节点的缺失;因此制定了相应的模糊匹配规则,在两个车牌间出现特定字符的匹配误差但满足其他匹配条件的前提下,认定该两个车牌号码匹配成功,这样可以在车牌个别字符识别错误的情况下仍然保证车辆出行路径的完整性和准确性;(2)路径节点缺失修补:模糊匹配可以大大减少路径中节点缺失的概率,但对于超出模糊匹配规则而产生的路径节点缺失的情况,需要对路径进行修补,以保 , The basic information and the vehicle license plate recognition section prior to travel through the time and place of the status information; Step. 4, OD correction module, and the information collection plate errors generated during transmission of the primary causes of multiple misrecognized OD OD OD and Since the end of the recognition error correction; specific correction method comprising: (the I) fuzzy matching license plate: plate recognition because the identification error may exist, the method entirely accurate match plate may bring some matching error, causing cycling trajectory tracking process deletion of part of the path node; thus develop the corresponding rule of fuzzy matching, matching error occurs under the premise of particular character, but meet other criteria matching, the identification numbers matching is successful two plates between two plates so that the plates can be individually character recognition error is still to ensure the integrity and accuracy of the path of travel of the vehicle; (2) path node missing patches: fuzzy match can greatly reduce the probability of missing nodes in the path, but the path to the node beyond the fuzzy matching rules arising from missing circumstances, need to be repaired path to security 证路径的完整性;当单车出行路径中某些节点缺失,根据道路的连通性和上下游关系,根据最短路径的原则对丢失部分的节点进行修补并记录相关修补信息; 步骤5,单车OD生成模块,是在单车轨迹跟踪的基础上进行处理,当设定的时间范围内下游未出现该车牌新的出行断面信息时,可以确认该车牌的出行OD已经完成并写入单车OD的数据库;反之该车牌的继续保持轨迹跟踪,直至车牌新的出行断面记录出现并对实时轨迹进行更新; 步骤6,OD统计模块对路网中符合指定车辆类型、出行起讫点相同、且经过讫点时间位于同一个周期内的单车OD记录个数进行统计,生成各个车牌采集断面之间的指定车辆类型的断面OD分别矩阵,由于部分车辆OD的起点或终点的缺失,会导致断面间OD观测量的统计值与真实值之间存在一定偏差,通过分析OD观测量误差产生的机理, The integrity of the certificate path; bicycle travel path when certain nodes missing, according to the relationship between the upstream and downstream communication and road repair node missing portion based on the principle of shortest paths and associated patch records information; Step 5, the OD generation cycling module, are processed on the basis of the tracking bicycle, a new travel when the downstream section of the license information does not appear within the set time, it was confirmed that the license plate travel OD OD cycling has been completed and the writing of a database; conversely the license plate continues to track the trajectory until a new travel section of track record the license plate appears and is updated in real time; step 6, OD statistics module designated road network in line with the type of vehicle, the same trip origins and destinations, and after finished in the same time point OD cycling the number of records within a statistical period, generating a respective type specified vehicle license acquisition section between the section matrix OD respectively, due to missing start or end portion of the vehicle OD, statistics can cause discontinuity OD observables there is a certain deviation between the true value, the mechanism by analyzing the Viewpoint OD measurement error generated, 建立观测OD量与断面流量之间的校验方程组对起终点缺失引起的OD误差进行修正,并根据断面与小区的隶属关系合成小区间OD分布矩阵; 步骤7,OD存储与展示模块将修正后的OD值按照统计周期、车辆类型和起讫点信息分类存储,并对路网动态OD分布矩阵进行报表和图形展示。 Establishing calibration equation between the amount of observation OD OD section of flow from the end of the error caused by missing correction, and inter-OD affiliation distribution matrix according to the cell cross-section synthetic cell; step. 7, the storage and display OD correction module OD values ​​in accordance with the statistical period, vehicle type and origin-destination information classified storage, distribution and road network dynamic OD matrix reporting and graphical display.
2.根据权利要求I所述的车牌识别OD信息采集系统的处理方法,其特征在于:所述的单车轨迹跟踪模块,其关键环节是将最新上传的车牌经过的断面记录与车辆之前的断面记录队列匹配比对,若车牌匹配成功则对车辆当前出行最新的路径进行更新,反之则重新生成另外一条实时更新的出行路径。 The license plate recognition OD processing method information collection system according to claim I, wherein: said bicycle tracking module, which is the key section and the recording section records vehicle before the plate passes recently uploaded queue matching comparison, if the license plate matched the success of the trip is the latest update of the current path of the vehicle, and vice versa regenerate the addition of a real-time updated travel path.
3.根据权利要求I所述的车牌识别OD信息采集系统的处理方法,其特征在于:所述的单车OD记录,是OD分布计算中的最小粒度单位的信息,通过对单车出行的统计,进一步生成小区间的OD分布。 The license plate recognition OD processing method information collection system according to claim I, wherein: said bicycle OD record information in distributed computing OD smallest granularity units, the statistical travel by bicycle, further OD distribution generated between cells.
4.根据权利要求I或3所述的车牌识别OD信息采集系统的处理方法,其特征在于:所述的单车OD记录,为单车OD记录存储的信息,包括车辆在一次出行中经过的车牌识别断面序列、断面经过时间信息。 4. I processing method of claim 3 OD license plate recognition information collection system according to claim, wherein: said recording OD bike, vehicle license plate recognition in a trip for cycling through the stored information record OD, comprising section sequence, the elapsed time information section.
CN2009100502115A 2009-04-29 2009-04-29 Processing method of road traffic OD (Optical Density) information collection system for license plate recognition CN101930668B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2009100502115A CN101930668B (en) 2009-04-29 2009-04-29 Processing method of road traffic OD (Optical Density) information collection system for license plate recognition

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2009100502115A CN101930668B (en) 2009-04-29 2009-04-29 Processing method of road traffic OD (Optical Density) information collection system for license plate recognition

Publications (2)

Publication Number Publication Date
CN101930668A CN101930668A (en) 2010-12-29
CN101930668B true CN101930668B (en) 2012-08-22

Family

ID=43369822

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2009100502115A CN101930668B (en) 2009-04-29 2009-04-29 Processing method of road traffic OD (Optical Density) information collection system for license plate recognition

Country Status (1)

Country Link
CN (1) CN101930668B (en)

Families Citing this family (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102521965B (en) * 2011-12-09 2013-08-14 同济大学 Effect evaluation method of traffic demand management measures based on identification data of license plates
CN102693561A (en) * 2012-05-25 2012-09-26 王骊 Method for enabling expressway vehicle over-speed monitoring camera system to realize automatic toll collection
CN102708686A (en) * 2012-05-30 2012-10-03 东南大学 Travel origin-destination and path recognizing method for motor vehicle on urban road
CN104077916B (en) * 2013-03-29 2016-12-28 上海市南电信服务中心有限公司 A kind of traffic information system based on Car license recognition
CN103700265B (en) * 2013-12-16 2016-01-06 青岛海信网络科技股份有限公司 The method of road network traffic information is obtained based on trip Origin And Destination
CN103646187B (en) * 2013-12-27 2017-02-08 中国科学院自动化研究所 Method for obtaining vehicle travel path and OD (Origin-Destination) matrix in statistic period
CN103903437B (en) * 2014-02-27 2016-06-08 中国科学院自动化研究所 The Vehicle emission OD matrix acquisition methods of data is detected based on video traffic
CN104183119B (en) * 2014-08-19 2016-08-24 中山大学 Based on the anti-arithmetic for real-time traffic flow distribution forecasting method pushed away of section OD
CN104269052B (en) * 2014-10-21 2017-02-01 南京感动科技有限公司 Perception traffic platform system and method of perception traffic platform system
CN105989731A (en) * 2015-02-15 2016-10-05 北京东土科技股份有限公司 Network-based vehicle track obtaining method and system
CN104915731B (en) * 2015-06-11 2018-04-27 同济大学 A kind of grand microcosmic integration method of vehicle driving reconstructing path based on automatic vehicle identification data
CN105279971B (en) * 2015-10-09 2017-11-14 北京易华录信息技术股份有限公司 A kind of information of vehicles matching process, system and supervising device
CN105355047B (en) * 2015-11-03 2017-06-06 吉林大学 The Data Fusion method of many Vehicle Detection source dynamic time granularities
CN106846538B (en) * 2015-12-04 2019-12-03 杭州海康威视数字技术股份有限公司 Cross vehicle record treating method and apparatus
CN106205199A (en) * 2016-08-31 2016-12-07 广东艾科智泊科技股份有限公司 A kind of car plate fault-tolerance processing system
CN106971534B (en) * 2017-02-09 2019-09-06 江苏智通交通科技有限公司 Commuter characteristic analysis method based on number plate data
CN108806271A (en) * 2018-06-22 2018-11-13 安徽科力信息产业有限责任公司 A kind of method and system of record road Vehicle emission distribution

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101017609A (en) 2007-02-07 2007-08-15 浙江工业大学 Intelligent analysis system for municipal traffic journey time
CN201035763Y (en) 2007-03-30 2008-03-12 王立国 Intelligent traffic guidance scheduling system
CN101216997A (en) 2008-01-11 2008-07-09 长安大学 A traffic information acquisition system based on linear array CCD video camera imaging

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101017609A (en) 2007-02-07 2007-08-15 浙江工业大学 Intelligent analysis system for municipal traffic journey time
CN201035763Y (en) 2007-03-30 2008-03-12 王立国 Intelligent traffic guidance scheduling system
CN101216997A (en) 2008-01-11 2008-07-09 长安大学 A traffic information acquisition system based on linear array CCD video camera imaging

Also Published As

Publication number Publication date
CN101930668A (en) 2010-12-29

Similar Documents

Publication Publication Date Title
O’Connor et al. Geo-temporal tracking and analysis of tourist movement
Zhan et al. Urban link travel time estimation using large-scale taxi data with partial information
Patnaik et al. Estimation of bus arrival times using APC data
CN103116825B (en) Intelligent city management
Park et al. Development and Evaluation of a Procedure for the Calibration of Simulation Models
CN101373559B (en) Method for evaluating city road net traffic state based on floating vehicle data
JP2011138486A (en) Method and apparatus for processing traffic information based on intersections and sections
CN101604478B (en) Method and system for processing dynamic traffic information
Toledo et al. Statistical validation of traffic simulation models
CN101794507B (en) Method for evaluating macroscopic road network traffic state based on floating car data
Strauch et al. Linking transport and land use planning: the microscopic dynamic simulation model ILUMASS
CN101783075B (en) System for forecasting traffic flow of urban ring-shaped roads
CN101751777B (en) Dynamic urban road network traffic zone partitioning method based on space cluster analysis
Frejinger Route choice analysis: data, models, algorithms and applications
Li et al. Traffic management and forecasting system based on 3d gis
CN100492434C (en) Traffic flow state analysis required detection vehicle sampling quantity obtaining method
Ho et al. Developing Singapore Driving Cycle for passenger cars to estimate fuel consumption and vehicular emissions
CN102044149B (en) City bus operation coordinating method and device based on time variant passenger flows
CN102097002A (en) Method and system for acquiring bus stop OD based on IC card data
El Esawey et al. Development of daily adjustment factors for bicycle traffic
Ma et al. Processing commercial global positioning system data to develop a web-based truck performance measures program
CN102737510A (en) Real-time traffic condition acquisition method based on mobile intelligent terminal
CN101794516B (en) Mobile phone information based section speed calculation method
Kumar et al. Short term traffic flow prediction in heterogeneous condition using artificial neural network
CN101694743B (en) Method and device for predicting road conditions

Legal Events

Date Code Title Description
C06 Publication
C10 Entry into substantive examination
C14 Grant of patent or utility model
C41 Transfer of patent application or patent right or utility model
ASS Succession or assignment of patent right

Owner name: SHANGHAI DIANKE INTELLIGENT SYSTEM CO., LTD.

Free format text: FORMER OWNER: SHANGHAI ELECTRICAL APPARATUS RESEARCH INSTITUTE (GROUP) CO., LTD.

Effective date: 20120803

Free format text: FORMER OWNER: SHANGHAI DIANKE INTELLIGENT SYSTEM CO., LTD.

Effective date: 20120803

LICC Enforcement, change and cancellation of record of contracts on the licence for exploitation of a patent or utility model