TWI774984B - Traffic incident detection system and method - Google Patents

Abstract

The invention discloses traffic incident detection system and method. First, selecting a road to be detected for traffic incident, and then detecting or extracting current and past travel time and traffic flow of the road from data source of traffic information. Second, calculating historical statistics of the past travel time and traffic flow of the road, and comparing the current travel time and traffic flow with the historical statistics of the past travel time and traffic flow to calculate a relative value of travel time and a relative value of traffic flow, and then inputting the relative values into a traffic incident analysis table. Third, analyzing or estimating a corresponding traffic event based on the relative values of the travel time and the relative values of the traffic flow from the traffic incident analysis table. Thereby, the invention can quickly analyze or estimate the traffic events of the selected road.

Description

Traffic incident detection system and method

The present invention relates to a traffic incident detection technology, in particular to a traffic incident detection system and method.

Most of the current traffic incidents are carried out manually, such as using a monitor to determine the content of the screen manually or using a passerby to notify, etc., while a few are automatically determined by the processing unit based on the image obtained by the ordinary camera lens. However, when using manual methods to determine or report traffic incidents, there may be a shortage of monitoring manpower or delays in reporting. In addition, when judging traffic incidents through images, misjudgment of traffic incidents may be caused by disturbances such as dislocation of images, shadows, and road surface color.

In the prior art, a real-time road traffic accident risk control method is proposed, which is a real-time road traffic accident risk prediction and control method based on a multi-class Support Vector Machine (SVM), which can be used to predict and detect the risk of traffic accidents in road sections. possibility. An accident prediction model based on multi-class support vector machines is established for the detected road sections, and the collected real-time traffic characteristic parameters are brought into the accident prediction model to judge whether there is a risk of a traffic accident. However, this method needs to use a multi-class support vector machine to learn the classification model of traffic accidents to discriminate the accident risk. and classification, so that it takes a lot of time and cost to build multi-class support vector machines and traffic accident learning classification models.

Therefore, how to provide a novel or innovative traffic incident detection technology has become a major research topic for those skilled in the art.

The present invention provides a novel or innovative traffic incident detection system and method, which can analyze or estimate traffic incidents on selected roads economically or quickly.

The traffic incident detection system of the present invention includes: a selection module, a detection module, a calculation module, and an analysis module, wherein the selection module selects a road to detect traffic incidents, and the detection The module detects or retrieves the current and past travel time and traffic flow of the road selected by the selection module through the data source of traffic information. The calculation module calculates the past of the road detected or retrieved by the detection module. The historical statistical values of travel time and traffic flow, and then compare the current travel time and traffic flow of the road detected by the detection module with the historical statistical values of the road's past travel time and traffic flow calculated by the calculation module. , to calculate the relative value of the travel time and the relative value of the traffic flow, so that the relative value of the travel time and the relative value of the traffic flow are respectively input into the traffic event analysis table by the calculation module, and the analysis module is derived from the traffic event analysis In the table, the corresponding traffic events are analyzed or estimated based on the relative value of travel time and the relative value of traffic flow.

The traffic incident detection method of the present invention includes: selecting a road to detect a traffic incident, detecting or retrieving the current and past travel time and traffic flow of the selected road from a data source of traffic information; calculating the detected road Or the historical statistical values of the past travel time and traffic flow of the road are extracted, and then the current travel time and traffic flow of the detected road are compared to calculate the The historical statistic values of the past travel time and traffic flow of the road, to calculate the relative value of the travel time and the relative value of the traffic flow, so as to input the relative value of the travel time and the relative value of the traffic flow into the traffic event analysis table respectively; and From the traffic event analysis table, the corresponding traffic events are analyzed or estimated based on the relative value of travel time and the relative value of traffic flow.

In order to make the above-mentioned features and advantages of the present invention more obvious and easy to understand, the following specific embodiments are given and described in detail with the accompanying drawings. Additional features and advantages of the present invention will be set forth in part in the description that follows, and in part will be apparent from the description, or may be learned by practice of the invention. The features and advantages of the invention will be realized and attained by means of the elements and combinations particularly pointed out in the scope of the patent application. It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not intended to limit the scope of the invention as claimed.

1‧‧‧Traffic Incident Detection System

10‧‧‧Select Module

11‧‧‧Traffic Incident

20‧‧‧Classification Module

21‧‧‧Road classification

30‧‧‧Detection Module

30'‧‧‧Collection Module

31‧‧‧Travel time

32‧‧‧Traffic flow

40‧‧‧Storage Module

41‧‧‧Traffic Information

50‧‧‧Computing Modules

51‧‧‧Historical Statistics

52‧‧‧ Relative value of travel time

53‧‧‧Relative value of traffic flow

60‧‧‧Analysis Module

61‧‧‧Analysis of traffic incidents

70‧‧‧Link Module

80‧‧‧Sort module

90‧‧‧Send module

Steps from S11 to S16, S21 to S26‧‧‧

Figure 1 is a schematic diagram of the structure of the traffic incident detection system of the present invention; Figure 2 is a traffic incident analysis table of the present invention; Figure 3 is a schematic flowchart of the traffic incident detection method of the present invention; A schematic diagram of the flow chart of the method of establishing the traffic incident analysis table of the invention.

The embodiments of the present invention will be described below by means of specific embodiments. Those who are familiar with the technology can understand other advantages and effects of the present invention from the contents disclosed in this specification. Therefore, it can also be implemented or applied by other different specific equivalent embodiments.

FIG. 1 is a schematic diagram of the structure of a traffic incident detection system 1 of the present invention, and FIG. 2 is a traffic incident analysis table 61 of the present invention. Meanwhile, the main technical contents of the traffic incident detection system 1 in FIG. 1 are as follows, and the remaining technical contents are the same as those described in FIGS. 3 to 4, and will not be repeated here.

As shown in FIG. 1, the traffic incident detection system 1 may include a selection module 10, a classification module 20, a detection module 30, a collection module 30', a storage module 40, and a calculation module A group 50 , an analysis module 60 , a connection module 70 , a sorting module 80 and a sending module 90 . In some embodiments, the selection module 10 may be a selector or selection software, the classification module 20 may be a classifier or classification software, the detection module 30 may be a detector or detection software, etc. The group 30' can be a collector or a collection software, etc., and the storage module 40 can be a database, a memory (such as a memory card), a hard disk (such as a cloud hard disk, a network hard disk), a CD or a pen drive, etc. The module 50 may be an arithmetic logic unit (ALU), computing software or statistical software, etc., the analysis module 60 may be an analyzer or analysis software, etc., the linking module 70 may be linking software, etc., and the sorting module 80 may be sorting software etc., the sending module 90 is a transmitter, a transceiver, a sending software or a sending and receiving software, and the like. However, the present invention is not limited to this.

The selection module 10 can select the road to detect the traffic event 11 , the classification module 20 can classify the road according to the characteristics of the road, and the detection module 30 can detect or retrieve the selection module 10 through the data source of the traffic information 41 . Present and past travel time 31 and traffic flow 32 of the selected road, and the storage module 40 can store traffic events 11 and traffic information 41 from data sources. In some embodiments, the traffic information 41 may include travel time 31 , traffic flow 32 , vehicle speed, occupancy rate, stop length and/or stop time for the detection module 30 , the storage module 40 , the calculation The module 50 and the analysis module 60 respectively detect, store, calculate and analyze the traffic information 41 .

The calculation module 50 can calculate the past travel time 31 of the road detected or captured by the detection module 30 and the historical statistical value 51 of the traffic flow 32 , and then calculate the current travel time of the road detected by the detection module 30 31 and the traffic flow 32 are respectively compared with the past travel time 31 and the historical statistical value 51 of the traffic flow 32 calculated by the calculation module 50, so as to calculate the relative value 52 of the travel time and the relative value 53 of the traffic flow. The calculation module 50 inputs the relative value 52 of the travel time and the relative value 53 of the traffic flow into the traffic event analysis table 61 respectively. The analysis module 60 can analyze or estimate the corresponding traffic event 11 from the traffic event analysis table 61 according to the relative value 52 of the travel time and the relative value 53 of the traffic flow.

The linking module 70 can establish a corresponding link between the relative value 52 of travel time, the relative value 53 of traffic flow and the traffic event 11 , so as to establish a correlation according to the relative value 52 of travel time and the relative value 53 of traffic flow in each road classification 21 . Linked Traffic Incident 11. The sorting module 80 can sequentially arrange the relative value 52 of the travel time and the relative value 53 of the traffic flow in each road classification 21 to establish or generate the traffic event analysis table 61 . The sending module 90 can send the traffic event 11 analyzed or estimated by the analyzing module 60 to the vehicle on the road, so as to notify or warn the driver of the vehicle about the traffic event 11 .

Therefore, the present invention can utilize the relationship between the travel time 31 of the road, the traffic flow 32 and the historical statistical value 51 and other traffic information 41 to quickly detect the traffic situation and analyze or estimate the traffic event 11 of the selected road. That is, the present invention can observe the traffic conditions according to the changes of the traffic information 41 such as the travel time 31 and the traffic flow 32 of the road, so as to quickly determine or estimate the traffic conditions and the traffic events 11 , such as serious traffic accidents, falling objects, congestion, etc. . In addition, when the traffic event 11 occurs, the vehicles behind are often blocked, so the present invention can automatically detect the duration of the traffic event 11, thereby providing necessary traffic information 41 to the driver behind.

In addition, the present invention can collect traffic events 11 on the road through, for example, the traffic database of the Highway Bureau of the Ministry of Communications, the police broadcast real-time road condition information of the government data open platform, etc. -Based Vehicle Probe; CVP), Fixed Vehicle Detector (Vehicle Detector; VD), Electronic Toll Collection (ETC)-based Vehicle Detection (ETC-Based Vehicle Probe; EVP), Global Positioning The GPS-Based Vehicle Probe (GVP) of the system obtains the relevant travel time 31 and traffic flow 32 and other traffic information 41, and then stores the traffic event 11 and the traffic information 41 in the storage module 40 (such as a database) , and then use big data analysis technology to automatically analyze and determine the traffic event 11 according to the traffic information 41 .

FIG. 3 is a schematic flowchart of the traffic incident detection method of the present invention, and please refer to FIG. 1 to FIG. 2 . Meanwhile, the traffic incident detection method in FIG. 3 mainly includes the following technical contents of steps S11 to S16 , and the remaining technical contents are the same as those described in FIGS. 1 and 4 , and will not be repeated here.

In step S11 of FIG. 3 , the selection module 10 shown in FIG. 1 selects the road to detect the traffic event 11 , and the classification module 20 classifies the road according to the characteristics of the road. For example, the road in the area where the traffic event 11 is to be detected is selected by the selection module 10, and the classification module 20 analyzes or determines which road classification 21 the road belongs to according to the road classification 21 of the administrative system, for example, the road belongs to a national highway, Road classification such as provincial, municipal, county, district or township roads21.

For example, the road selected by the selection module 10 to detect the traffic event 11 on April 2, 2019 is a national highway, such as National Highway No. 1 (Zhongshan Expressway).

In step S12 of FIG. 3 , the current and past travel times 31 of the road selected by the selection module 10 are detected or retrieved by the detection module 30 through at least one (eg, multiple) data sources. Traffic information 41 such as traffic flow 32 . For example, at least one data source of the traffic information 41 may include vehicle detection (CVP) based on cell phone base station, stationary vehicle detector (VD), vehicle detection (EVP) based on electronic road pricing system (ETC) ), GPS Detective Vehicle (GVP), the traffic database of the Highway Bureau of the Ministry of Communications, the police broadcast real-time road condition information of the government data open platform, etc., and the traffic information 41 can be stored in the storage module 40 (such as a database) and classified by administrative system according to the road classification21.

For example, at 10:00 on April 02, 2019, the detection module 30 obtains real-time road condition information through the police broadcast of the government data open platform, and travels 71 kilometers (km) to 83 kilometers south of National Highway No. 1 (Zhongshan Expressway). Within the range, the average travel time 31 of the road is 43 minutes, and the traffic flow 32 every five minutes is 146 vehicles (for example, the total amount of traffic passing through a detection station on National Highway No. 1 within this time range) ).

In step S13 of FIG. 3 , the calculation module 50 calculates the historical statistics 51 of the traffic information 41 such as the past travel time 31 and the traffic flow 32 of the road detected or captured by the detection module 30 . For example, the calculation module 50 calculates the historical statistical value 51 of the traffic information 41 such as the travel time 31 and the traffic flow 32 of the road in the past according to the traffic information 41 in the storage module 40 (such as a database), The historical statistical value 51 can be an arithmetic mean, median, mode, or standard deviation.

For example, the calculation module 50 calculates the historical statistical value 51 of the traffic information 41 such as the past travel time 31 and the traffic flow 32 of the road in the traffic information 41 of the storage module 40 (such as a database), for example, in March 2019 From 10:00 to 10:05, within the range of 71 kilometers to 83 kilometers south of National Highway No. 1 (Zhongshan Expressway), the average travel time 31 of this road in the past is 8 minutes, and the five-minute traffic flow 32 is 520 vehicle.

In step S14 in FIG. 3 , the calculation module 50 compares the current travel time 31 and the traffic flow 32 of the road detected by the detection module 50 in step S12 with those calculated by the calculation module 50 in step S13 respectively. The past travel time 31 and the historical statistical value 51 of the traffic flow 32 of the road are used to calculate the relative value 52 (including the proportional relationship) of the travel time and the relative value 53 (including the proportional relationship) of the traffic flow.

For example, the relative value 52 of the travel time is the current travel time 31 (such as 43 minutes) of the road in step S12 divided by the past travel time 31 (such as 8 minutes) of the road in step S13, so the travel time in this embodiment is The relative value 52 of time is 5.375 (ie 43/8=5.375). At the same time, the relative value 53 of the traffic flow is the current traffic flow 32 (eg 146 vehicles) of the road in step S12 divided by the past traffic flow 32 (eg 520 vehicles) of the road in step S13, so the traffic flow in this embodiment is The relative value 53 is 0.28 (ie 146/520=0.28).

In step S15 in FIG. 3, the calculation module 50 inputs the relative value 52 of the travel time and the relative value 53 of the traffic flow into the traffic event analysis table 61 shown in FIG. In the event analysis table 61, the corresponding traffic event 11 is analyzed or estimated according to the relative value 52 of the travel time and the relative value 53 of the traffic flow. For example, the analysis module 60 first analyzes the road classification (such as national road, provincial road, city road, county road, district road or township road) from the traffic event analysis table 61, and then analyzes it from the traffic event analysis table. In 61, the corresponding traffic event 11 is extracted according to the location of the closest relative value 52 of the travel time and the relative value 53 of the traffic flow.

For example, the calculation module 50 inputs the relative value 52 of the travel time (eg, the value 5.375) and the relative value 53 (eg, the value 0.28) of the traffic flow in step S14 into the traffic event analysis table 61 shown in FIG. 2, respectively, Then, the analysis module 60 compares the relative value of the closest travel time according to the relative value 52 of the travel time (eg, a value of 5.375) and the relative value of the traffic flow 53 (eg, a value of 0.28). The relative value of 52 (eg value 5) and the relative value of traffic flow 53 (eg value 0.3), according to the relative value of the closest travel time 52 (eg value 5), the relative value of traffic flow 53 (eg value 0.3) and the road. The road classification 21 (such as a national road) analyzes or estimates the corresponding traffic event 11 as a "collision accident".

In step S16 of FIG. 3 , the sending module 90 sends the traffic event 11 analyzed or estimated by the analysis module 60 to the vehicle on the road, so as to notify or warn the driver of the vehicle about the traffic event 11 .

For example, the sending module 90 sends the vehicle to the vehicle at 10:00 on April 2, 2019, within the range of 71 kilometers to 83 kilometers south of National Highway No. 1 (Zhongshan Expressway), the traffic incident 11 on the road is analyzed or It is estimated that the impact of a "collision accident" will cause severe back-blocking to notify or warn the driver of the vehicle.

FIG. 4 is a schematic flowchart of the method of establishing the traffic incident analysis table 61 shown in FIG. 2 of the present invention, and please refer to FIG. 1 . Meanwhile, the establishment of the traffic event analysis table 61 may include the technical contents of the following steps S21 to S26.

In step S21 of FIG. 4 , the classification module 20 shown in FIG. 1 performs road classification according to the characteristics of the roads.

For example, the classification module 20 stores the road classification 21 of the road in the storage module 40 (such as a database) according to the characteristics of the road, and classifies the roads in the region into national highways, provincial highways, and municipal highways according to the administrative system. , county road, district road or township road, etc.

In step S22 in FIG. 4 , the collection module 30 ′ collects traffic information such as road-related travel time 31 and traffic flow 32 of the traffic event 11 through at least one (eg, plural) data source of the traffic information 41 according to the road classification 21 . 41. For example, the collection module 30' broadcasts real-time road condition information through the traffic database of the Highway Bureau of the Ministry of Communications and the police on the government data open platform. Data sources collect various traffic events on the road11, and use mobile phone base station-based vehicle detection (CVP), fixed vehicle detection (VD), electronic road pricing system (ETC)-based vehicle detection (EVP) ), GPS Detective Vehicle (GVP), etc. to obtain the relevant travel time 31 and traffic flow 32 and other traffic information 41 to store the traffic event 11 and the traffic information 41 in the storage module 40 (eg database).

For example, the collection module 30 ′ collects the traffic information 41 such as the travel time 31 and the traffic flow 32 related to the road (eg, national highway) of the traffic event 11 . For example, in the traffic incident 11 at 08:00 on April 01, 2019, it is an outside drop, and the national highway No. 1 (Zhongshan Expressway) is within a range of 71 kilometers to 83 kilometers south, the average travel time 31 related to the road 30 minutes, and the five-minute traffic flow 32 is 208 vehicles.

In step S23 of FIG. 4 , the calculation module 50 calculates the historical statistics 51 of the traffic information 41 such as the past travel time 31 of the road and the traffic flow 32 of the traffic event 11 . For example, the calculation module 50 calculates the historical statistical value 51 of the traffic information 41 such as the travel time 31 and the traffic flow 32 of the road in the past according to the traffic information 41 in the storage module 40 (such as a database), The historical statistical value 51 can be an arithmetic mean, median, mode, or standard deviation.

For example, at 08:00 in March 2019, the computing module 50 obtains real-time road condition information from the police broadcast on the government data open platform or the traffic database of the Highway Bureau of the Ministry of Communications, and goes south on National Highway No. 1 (Zhongshan Expressway) 71 Within the range from km to 83 km, the average travel time 31 of the road in the past is 7.2 minutes, and the five-minute traffic flow 32 is 520 vehicles (for example, a detection station on National Highway No. 1 passes by within this range). Therefore, the historical statistical value of travel time 31 is 7.2 minutes, and the historical statistical value of traffic flow 32 is 520 vehicles.

In step S24 in FIG. 4 , the calculation module 50 compares the road-related travel time 31 and the traffic flow 32 (see step S22 ) of the traffic events 11 collected by the collection module 30 ′ with the data collected by the calculation module 50 , respectively. The calculated historical statistical value 51 of the past travel time 31 and the traffic flow 32 of the road of the traffic event 11 (see step S23), to calculate the relative value 52 of the travel time (including the proportional relationship) and the relative value 53 of the traffic flow (including ratio).

For example, the relative value 52 of the travel time is the travel time 31 (eg, 30 minutes) related to the road in step S22 divided by the past travel time 31 (eg, 7.2 minutes) of the road in step S23 , so the travel time in this embodiment is The relative value 52 of time is 4.16 (ie, 30/7.2=4.16). At the same time, the relative value 53 of the traffic flow is the traffic flow 32 (such as 208 vehicles) related to the road in step S22 divided by the past traffic flow 32 (such as 520 vehicles) of the road in step S23, so the traffic flow in this embodiment is The relative value 53 is 0.4 (ie 208/520=0.4).

In step S25 in FIG. 4 , the relative value 52 of travel time, the relative value 53 of traffic flow and the corresponding connection of the traffic event 11 are established by the link module 70 , so that the relative value 52 of travel time is used in each road classification 21 The traffic event 11 is linked to the relative value 53 of the traffic flow.

In step S26 in FIG. 4 , the relative value 52 of the travel time and the relative value 53 of the traffic flow in each road classification 21 are sequentially arranged by the sorting module 80 to establish or generate the traffic event analysis shown in FIG. 2 Table 61.

In conclusion, the traffic incident detection system and method of the present invention can at least have the following features, advantages or technical effects.

1. The present invention can greatly reduce the construction cost of traffic infrastructure without the need for additional road testing equipment, so as to analyze or estimate the traffic events of the selected road economically or quickly.

2. The present invention can analyze or estimate the traffic events of the selected road by using the relationship between the travel time of the road, the traffic flow and the historical statistical value and other traffic information to quickly detect the traffic situation. That is, the present invention can observe traffic conditions according to changes in traffic information such as road travel time and traffic flow, so as to quickly determine or estimate what traffic conditions are traffic events, such as serious traffic accidents, falling objects, and congestion.

3. When a traffic incident occurs, the rear vehicles are often blocked, so the present invention can automatically detect the duration of the traffic incident, thereby providing necessary traffic information for the rear driver.

Fourth, the present invention can use big data analysis technology to automatically analyze and judge traffic events. That is, the present invention can collect various traffic events on the road through the traffic database of the Expressway Bureau of the Ministry of Communications, the police broadcast real-time traffic information on the government data open platform, etc. , stationary vehicle detector (VD), vehicle detection (EVP) based on electronic road pricing system (ETC), global positioning system detection vehicle (GVP), etc. to obtain relevant travel time and traffic flow and other traffic information, And the traffic information is stored in a storage module (such as a database), so as to use big data analysis technology to automatically analyze and determine traffic events according to the traffic information.

Fifth, the present invention can be applied to, for example, intelligent transportation, traffic control center or vehicle navigation system.

The above-mentioned embodiments are only illustrative of the principles, features and effects of the present invention, and are not intended to limit the applicable scope of the present invention. Modifications and changes are made to the implementation form. Any equivalent changes and modifications made by using the contents disclosed in the present invention should still be covered by the scope of the patent application. Therefore, the scope of protection of the right of the present invention should be listed in the scope of the patent application.

1‧‧‧Traffic Incident Detection System

10‧‧‧Select Module

11‧‧‧Traffic Incident

20‧‧‧Classification Module

21‧‧‧Road classification

30‧‧‧Detection Module

30'‧‧‧Collection Module

31‧‧‧Travel time

32‧‧‧Traffic flow

40‧‧‧Storage Module

41‧‧‧Traffic Information

50‧‧‧Computing Modules

51‧‧‧Historical Statistics

52‧‧‧ Relative value of travel time

53‧‧‧Relative value of traffic flow

60‧‧‧Analysis Module

61‧‧‧Analysis of traffic incidents

70‧‧‧Link Module

80‧‧‧Sort module

90‧‧‧Send module

Claims (18)

A traffic event detection system, comprising: a selection module for selecting a road to detect traffic events; a detection module for detecting or retrieving the road selected by the selection module through a data source of traffic information The current and past travel time and traffic flow of the road; a calculation module, which calculates the historical statistical values of the past travel time and traffic flow of the road detected or captured by the detection module, and then the detection module The current travel time and traffic flow of the road detected by the group are respectively compared with the historical statistical values of the past travel time and traffic flow of the road calculated by the calculation module, so as to calculate the relative value of the travel time and the vehicle flow. The relative value of the flow, so that the relative value of the travel time and the relative value of the traffic flow are respectively input into the traffic incident analysis table by the calculation module; and an analysis module is based on the travel time from the traffic incident analysis table. The relative value of the relative value and the relative value of the traffic flow are analyzed or estimated to correspond to the traffic event. The traffic incident detection system as described in Item 1 of the patent application scope, wherein the traffic information includes travel time, traffic flow, vehicle speed, occupancy rate, waiting time or waiting time, and the historical statistical value is an arithmetic mean , median, mode, or standard deviation. The traffic incident detection system as described in item 1 of the patent application scope, wherein the data sources of the traffic information include vehicle detection (CVP) based on mobile phone base stations, fixed vehicle detectors (VD), electronic Vehicle detection (EVP) based on road pricing system (ETC), detection vehicle (GVP) of global positioning system, traffic database of Highway Bureau of Ministry of Communications, or police broadcast real-time traffic information on government data open platform. The traffic incident detection system as described in item 1 of the scope of the application further comprises a classification module and a storage module, wherein the classification module is used to classify roads according to the characteristics of the road, and the storage module is Stores traffic events and traffic information from this data source. The traffic incident detection system described in item 1 of the scope of the patent application further includes a collection module, which collects the road-related travel time and traffic flow of the traffic incident through the data source of the traffic information according to the road classification. The traffic incident detection system as described in item 5 of the scope of the application, wherein the calculation module further calculates the historical statistical values of the past travel time and traffic flow on the road of the traffic incident, and then collects the data collected by the collection module. The road-related travel time and traffic flow of the traffic event are compared with the historical statistics of the past travel time and traffic flow of the road of the traffic event calculated by the calculation module, so as to calculate the relative value of the travel time and the traffic flow. Relative value of traffic flow. The traffic incident detection system described in item 1 of the scope of the application further includes a link module for establishing the corresponding link between the relative value of the travel time, the relative value of the traffic flow and the traffic incident, so as to be The traffic events that are linked based on the relative value of the travel time and the relative value of the traffic flow in the classification. The traffic incident detection system as described in item 1 of the scope of the patent application further comprises a sorting module, which sequentially sorts the relative value of the travel time and the relative value of the traffic flow in each road classification to establish or generate The traffic incident analysis table. The traffic incident detection system as described in item 1 of the scope of the application further includes a sending module, which sends the traffic incident analyzed or estimated by the analysis module to the vehicle on the road, so as to transmit the traffic incident to the vehicle on the road. Notify or warn the driver of the vehicle. A traffic incident detection method, comprising: Select a road to detect traffic events, and detect or retrieve the current and past travel time and traffic flow of the selected road from the data source of traffic information; calculate the detected or retrieved past travel of the road The historical statistic value of time and traffic flow, and then compare the detected current travel time and traffic flow of the road with the calculated historical statistic value of the past travel time and traffic flow of the road to calculate the travel time The relative value of the travel time and the relative value of the traffic flow are respectively entered into the traffic incident analysis table; The relative value of traffic flow is analyzed or estimated to correspond to traffic events. The traffic incident detection method as described in Item 10 of the scope of application, wherein the traffic information includes travel time, traffic flow, vehicle speed, occupancy rate, waiting length or waiting time, and the historical statistical value is an arithmetic mean , median, mode, or standard deviation. The traffic incident detection method as described in item 10 of the scope of application, wherein the data sources of the traffic information include vehicle detection (CVP) based on mobile phone base stations, fixed vehicle detectors (VD), electronic Vehicle detection (EVP) based on road pricing system (ETC), detection vehicle (GVP) of global positioning system, traffic database of Highway Bureau of Ministry of Communications, or police broadcast real-time traffic information on government data open platform. The traffic incident detection method described in Item 10 of the scope of the patent application further includes classifying the road according to the characteristics of the road, and storing the traffic incident and traffic information from the data source. The traffic incident detection method described in Item 10 of the scope of the patent application further includes collecting the road-related travel time and traffic flow of the traffic incident through the data source of the traffic information according to the road classification. The traffic incident detection method described in Item 14 of the scope of the application further includes calculating the historical statistics of the past travel time and traffic flow of the road of the traffic incident, and then calculating the collected travel time related to the road of the traffic incident. Compare the calculated historical statistic value of the past travel time and the traffic flow on the road of the traffic event with the traffic flow, so as to calculate the relative value of the travel time and the relative value of the traffic flow. The traffic incident detection method as described in item 10 of the scope of the patent application further comprises establishing a corresponding link between the relative value of the travel time, the relative value of the traffic flow and the traffic incident, so as to classify the roads according to the travel time The relative value of the traffic event is linked to the relative value of the traffic flow. The traffic incident detection method as described in item 10 of the scope of the patent application further comprises sequentially arranging the relative value of the travel time and the relative value of the traffic flow in each road classification to establish or generate the traffic incident analysis table. The traffic incident detection method described in item 10 of the scope of the patent application further comprises sending the analyzed or estimated traffic incident to the vehicle on the road, so as to notify or warn the driver of the vehicle about the traffic incident.

Images