TW201800289A - System and method for analyzing driving behavior regarding traffic accidents by integrating a GPS analysis module, G-sensor analysis module, image analysis module, and vehicular data analysis module as well as the reference to road-network info-database and traffic data info-base - Google Patents

Abstract

The invention is about a system and a method for analyzing driving behavior regarding traffic accidents. A GPS (Global Positioning System) analysis module, a G-sensor analysis module, an image analysis module, and a vehicular data analysis module in analyzing the data transmitted from a driving recorder, as well as the reference to objective information from a road-network info-database and a traffic data info-base: those are integrated by a driving behavior analysis module to signal an alarm and produce a comprehensive accident report while a vehicle accident happens.

Description

Driving behavior analysis system and method for accident

The present invention relates to a traffic monitoring system and method, and more particularly to a driving behavior analysis system and method that collects various active and passive information to report accidents.

According to the Statistics Report of the Republic of China on Road Traffic Accidents of the Republic of China in 2003, the number of road accidents in that year was as high as more than 295,000, and it can be observed that the number of accidents has increased year by year. Once an unfortunate accident occurs, clarify how Occurring to blame afterwards or claim settlement, etc., in today's common situation, most police units or appraisers do accident identification at the scene of the accident and then cooperate with the pictures of driving record images to generate accident appraisal reports. This method considers The evidence is still not comprehensive enough, and the evidence that is difficult to directly inspect may be inferred and judged by the appraisal personnel, which often causes the two parties to the accident to hold their own words or cause resentment. Therefore, if we can develop a trackable driving behavior data and Combined with all the available objective information in the surrounding area for comprehensive analysis and judgment, a more scientific and complete analysis of fair traffic accident data can be provided, which is more likely to be accepted by relevant persons or police units.

Of course, there are several previous technologies that propose various types of traffic accident monitoring systems designed to solve the problem of accident identification similar to the above. In the prior technology, there are a large number of cameras equipped with radio equipment, digital image recognition or digital The network-type digital image monitoring system of audio recognition attempts to use the observation facilities located in the surrounding environment to detect the characteristics or process of traffic accidents by identifying methods, and further attempts to visualize the location of traffic accidents to assist on-site processing personnel. Judge.

However, due to the high cost of such systems, most of them can only be applied to police units, and the evidence of observation is still limited to the monitoring systems, radio equipment, and cameras installed, and the amount of data collected is not yet comprehensive. And it is difficult to provide ordinary users with the ability to submit objective identification reports as soon as an accident occurs, so it is known that their applications still have their limitations.

For the technicians in this field, in order to solve the above problems, a technology that can widely collect all the data that affect the accident and automatically make a more objective third-party report is a technology that can quickly understand the cause of the accident and save The necessary technology for the follow-up people's waste of legal proceedings.

The present invention proposes a driving behavior analysis system and method for accidents, which is because the existing systems and methods are difficult to cover all third-party reports that may affect the accident or make a more fair and recognized report. The present invention came into being.

The system of the present invention mainly includes at least three databases, which are: a one-line vehicle data database, a road network information database, and a traffic data information database.

Among them, the driving data database is used to receive and store a line of driving data transmitted from an external plurality of driving record devices, and the driving data includes a corresponding vehicle provided with each of the driving recorders. Positioning data, vehicle speed data, gravity data, driving image data, and vehicle peripheral equipment data, etc., and the external driving recorder can be any of today's commonly used to record and collect user driving vehicles, images, various parameters The driving data device or driving computer, etc. However, it must be able to upload the searched information to the driving data database regularly or at any time through wireless transmission.

Among them, the road network database stores a plurality of public road section data, and each road section data is mainly a large or small range of road section data provided by a government unit or a professional unit. The data subdivision includes the road section number and road section positioning. Collected data, directional data of feasible vehicles on the road section or regulated vehicles, speed limit data of each road section, and special or general traffic regulations data that must be followed on each road section.

Wherein, the traffic data information database is connected with an external vehicle-connected big data database, and it can receive the external vehicle-connected big data database and store a plurality of real-time traffic information. Each of the real-time traffic information described includes Real-time speed data, real-time road condition data, individual surveillance image data, and electronic toll data for expressways or highways on a wide range of roads or road networks.

The system of the present invention mainly includes at least five analysis modules, respectively: a GPS (Global Positioning System) analysis module, a gravity sensor (G-sensor) analysis module, an image analysis module, and a vehicle Data analysis module and a driving behavior analysis module.

The positioning system analysis module is used to analyze the positioning information of the vehicle obtained from the driving data database. The positioning system analysis module can be used to make overspeed judgment based on the driving data. The positioning data of the vehicle determines the road segment of the vehicle, and then the speed limit data of the road segment of the vehicle is obtained from each of the real-time traffic information in the driving data of the road network information database. The positioning system analysis module then uses the The driving data The vehicle speed data of the vehicle determines whether the vehicle is overspeed, and then tracks the vehicle's overspeed positioning data and calculates the distance traveled by the vehicle's overspeed positioning data. This is an overspeed determination method.

In addition, the positioning system analysis module can also perform rapid acceleration and rapid braking judgments. The positioning system analysis module estimates the acceleration value of the vehicle based on changes in the positioning data of the vehicle in continuous time in the driving data ( (Or instantaneous acceleration value), the positioning system analysis module then determines whether the acceleration value of the vehicle exceeds the pre-set rapid acceleration or rapid braking threshold value set in the system. If a rapid acceleration or rapid braking situation occurs, the positioning system analyzes The module is about to mark the positioning data of the sudden acceleration and sudden braking of the vehicle corresponding to the driving data, and further count the number of occurrences of the rapid acceleration and sudden braking as a parameter to observe the driving behavior of the vehicle.

The positioning system analysis module can further compare the driving trajectory and speed data formed by the positioning data of the vehicle in the driving data data with the relevant data of the road section passed by the vehicle in the road network information database to Analyze whether the driving behavior of the vehicle's rapid acceleration or braking is reasonable, or it is just general driving habits.

The gravity sensing analysis module included in the system of the present invention is used to track the value of the gravity change of the vehicle in the driving data data, to analyze the road conditions of the section that the vehicle passes, and to perform rapid acceleration, braking or braking of the vehicle. The judgment of the behavior of sharp turns is mainly based on the change of the values of the three axes (X, Y, Z axes) carrying the gravity data in the driving data data for a continuous period of time in the driving data analysis module to calculate the gravity change. The gravity sensing analysis module then judges whether the gravity change value of the vehicle exceeds the preset threshold for rapid acceleration, braking or turning, to indicate the gravity data of the vehicle undergoing rapid acceleration, braking or turning, and statistics. Number of sudden accelerations and sudden brakings.

In addition, the gravity sensing analysis module tracks the gravitational change of the vehicle in the driving data, and can also determine the state of the road surface driven by the vehicle based on the average change of the three-axis values. The instantaneous changes in the gravity of the X, Y, and Z axes are used to understand the direction, location, or estimated severity of the vehicle being impacted or impacted by others.

Furthermore, the image analysis module included in the system of the present invention is used to analyze the driving image of the vehicle in the driving data data, and it is used to obtain some passive driving data through image recognition and object detection methods. The information is mainly as follows.

The image analysis module can perform traffic sign and signal analysis. It uses the object detection method to detect the continuous i-frames in the driving image of the vehicle in the driving data data, based on color, shape, and size. The feature detects that it may be a traffic sign or a sign. The image analysis module uses objects to identify whether the detected person is a traffic sign or a sign from the road network database or other databases. Then, the The image analysis module indicates the time point of the traffic signs and signs passed by the vehicle, and the meaning of the content represented by the traffic signs and signs.

In addition, the image analysis module can also perform lane offset and transformation analysis. The image analysis module passes the vehicle's driving image through the object detection method to detect the road surface based on characteristics such as color, shape and size. Possible traffic markings, and continue to calculate the changes in traffic marking positions over time to determine whether the offset value exceeds a predetermined lane deviation threshold. If it exceeds, it means that the vehicle has a lane deviation or Change, the image analysis module then marks the point in time when the vehicle lane shift or lane change occurs.

In addition, the image analysis module can also analyze the distance between the vehicles, which is mainly based on the image analysis module detecting the driving image through the object to detect the possible The traffic markings are adjusted according to the type of markings to generate a vehicle distance calculation correction parameter. Then, the image analysis module judges whether there is a vehicle in front of the vehicle through the object detection method and uses the vehicle distance calculation correction parameter as an aid to estimate One vehicle distance, and the image analysis module can continuously record each vehicle distance information over time to track the relative state of the vehicle and the preceding vehicle.

The vehicle data analysis module included in the system of the present invention is used to analyze the vehicle peripheral equipment data in the driving data data, especially the tire pressure value and the data of the driving computer (OBU, On Board Unit). The vehicle data analysis module extracts data related to driving behavior from the vehicle peripheral equipment data, including the direction signal of the vehicle, the driving gear used, the seat belt signals of the front and rear seats, and the signal of driving on the brakes. , Engine speed, cooling water temperature, throttle operating status, wheel steering angle, operating status of various active and passive auxiliary systems, etc., as auxiliary information, the positioning system analysis module and the gravity sensing analysis module described above can be used And the driving data detected by the image analysis module are verified or combined to infer the driving condition of the vehicle.

The driving behavior analysis module in the system of the present invention is an integrated module, which receives the rapid acceleration, sharp turning, or turning of the vehicle analyzed by the positioning system analysis module and the gravity sensing analysis module. The driving behavior of sudden braking, and various driving data detected by the image analysis module, and vehicle-related data extracted by the vehicle data analysis module, and the driving behavior analysis module will The received data and the real-time traffic information obtained from the traffic data information database are integrated and analyzed through pre-set weights or self-adjusted weights. Finally, a driving behavior score is calculated and calculated. The driving behavior score can be used as the basis. A driving behavior reference value output by the system.

However, the driving behavior analysis system for an accident of the present invention may further include an accident detection module and an accident report module, wherein the accident detection module is coupled with the positioning system analysis module and the gravity sensing analysis module. Team, the image analysis module and the vehicle data analysis module link and judge abnormal data to generate an accident alert, which can be provided to road police agencies, highway rescue operators, ambulance units or ordinary drivers, etc. To enable it to understand the accident in a short period of time and perform subsequent assistance actions; among them, the accident report module is linked with the driving behavior analysis module and the accident detection module, and the accident report module will be based on the driving The behavior score and the warning of the accident generate an accident report data, the content of which can be information obtained from each of the modules or the database of the system of the present invention, including the basic information of the accident (the time of the accident, the location of the accident) And weather information), driving behavior analysis information (speed information, gravity information, driving distance information, and collision direction judgment information) ), Accident images, sign information, and other information sources (CCTV) to help clarify the accident, etc., and the accident report data can be displayed through web pages, or through URL links, through applications, etc. Method to obtain, the accident report data can achieve the main purpose of the invention, quickly and easily provide a more fair and comprehensive consideration of third-party accident reference information.

Corresponding to the driving behavior analysis system for accidents of the present invention, the present invention also proposes a driving behavior analysis method for accidents, which mainly includes the following steps: 1. A vehicle data database receives and stores at least one vehicle record device from the outside One line of vehicle data transmitted; 2. One road network information database stores plural public road section data; 3. One traffic data information database receives and stores multiple real-time traffic information; 4. A positioning system analysis module stores the traffic data Vehicle positioning The driving trajectory and speed data formed by the data are compared with the relevant data of the vehicle's passage in the road network information database to analyze the driving behavior of the vehicle's rapid acceleration or braking; 5. A gravity sensing analysis model The group tracked the gravity change value of the vehicle in the driving data to analyze the road conditions of the vehicle and the driving behavior of the vehicle's rapid acceleration, braking or turning; 6. An image analysis module identified the Object detection method to detect traffic signs, traffic signs, lane changes in the driving image of the vehicle in the driving data and calculate the passive driving data of the vehicle and the distance between the vehicle and the vehicle in front; 7. Vehicle data The analysis module analyzes the vehicle peripheral equipment data in the driving data data, mainly through the tire pressure value and the driving computer data analysis to extract data related to driving behavior from the vehicle peripheral equipment; and 8. a driving behavior analysis module Receive the driving behavior of the vehicle's rapid acceleration, sharp cornering or sudden braking analyzed by the positioning system analysis module and the gravity sensing analysis module, and the image The passive driving data detected by the analysis module and the driving behavior-related data extracted by the vehicle data analysis module are used to synchronize the received data and the real-time data obtained from the traffic data database according to the time sequence. Traffic information is integrated and analyzed through weights. Analysis and calculation are performed through weights to generate a driving behavior score.

The above driving behavior analysis method for accidents may optionally include the following steps:

1. An accident detection module and the positioning system analysis module, the gravity sensing analysis module, the image analysis module, and the vehicle data analysis module Link, the accident detection module and determine abnormal data to generate an accident warning.

2. An accident report module is connected with the driving behavior analysis module and the accident detection module. The accident report module will generate an accident report data according to the driving behavior score and the accident occurrence warning.

To sum up, the driving behavior analysis system and method for accidents of the present invention is to comprehensively collect various objective auxiliary data and comprehensively collect various passive information generated during the travel of the vehicle to comprehensively judge the driving behavior. Accidents, the present invention can also provide the first-time warning or after-the-fact objective auxiliary report, which is an extremely effective invention in the aspect of traffic safety and accident-responsibility.

1‧‧‧ driving recorder

2‧‧‧Driving Behavior Analysis System

10‧‧‧ Positioning System Module

11‧‧‧Gravity Sensor Module

12‧‧‧Image capture module

13‧‧‧Storage Module

14‧‧‧Wireless Transmission Module

15‧‧‧trip computer

16‧‧‧ Tire Pressure Detector

20‧‧‧ Driving data database

21‧‧‧ Road Network Information Database

22‧‧‧Traffic Data Information Database

23‧‧‧ Positioning System Analysis Module

24‧‧‧ gravity sensing analysis module

25‧‧‧Image Analysis Module

26‧‧‧Vehicle data analysis module

27‧‧‧Driving Behavior Analysis Module

28‧‧‧ Incident Detection Module

29‧‧‧ Incident Report Module

S21 ~ S25‧‧‧Method steps

S221 ~ S224‧‧‧Method steps

FIG. 1 is a system architecture diagram of a driving behavior analysis system for an accident according to the present invention.

FIG. 2 is a method step diagram of the driving behavior analysis method according to the present invention.

FIG. 3 is a schematic diagram of related aspects of a driving behavior analysis method according to the present invention.

In the following, the present invention will be further described with examples and drawings. First, please refer to FIG. 1, which is a system architecture diagram of the driving behavior analysis system of the present invention for an accident. Driving recorder 1 and driving behavior analysis system of the present invention 2 to complete.

Among them, the driving recorder 1 outside the system includes a positioning system module 10, a gravity sensing module 11, an image capture module 12, a storage module 13, and a wireless transmission module 14, and the positioning system module 10 is a record The GPS positioning information of the vehicle installed with the driving recorder 1 can be composed of longitude, latitude, speed, direction angle, altitude, time stamp, etc .; and the gravity sensing module 11 is used to sense the load. The value of the change in the gravitational acceleration (G force) generated in the three-axis (X, Y, Z) space; the image capture module 12 is captured through a common driving recorder, blind spot detection system, and a reverse development system. Take the images recorded during the travel of the vehicle, for example: an image stream file composed of three-minute driving images, and encoded and compressed using an image format with both quality and compression ratio; and as described above GPS positioning information, three-axis gravity change values, and recorded images will be transmitted and stored to the storage module 13, and then periodically or continuously through the wireless transmission module 14 to wirelessly according to the settings of the system or the driving recorder 1. Sender The information is transmitted to the driving behavior analysis system 2. In addition, the driving computer 15 and the tire pressure detector 16 installed on the same vehicle will also send the driving data and tire pressure data retrieved by it to the wireless transmission. The module 14 transmits the driving data and the tire pressure data to the driving behavior analysis system 2.

The databases and modules included in the driving behavior analysis system 2 of the present invention will be described in detail in the following paragraphs.

First, the driving behavior analysis system 2 of the present invention includes a driving data database 20, which receives driving record information transmitted from the external driving recorder 1 when the vehicle is driven, and includes GPS reservation data and G-sensor gravity Data, image data, driving data, etc. These data will be stored in the system in the form of SQL or NoSQL.

The driving behavior analysis system 2 of the present invention includes a road network information database 21, which stores a complete and open section information, wherein each section data is the data obtained through actual road measurement by other units or government units, and is numbered by the section , Section positioning information collection, section directionality, section speed limit and section traffic regulations (one-way street, no left turn, ...) and other fields.

The driving behavior analysis system 2 of the present invention also includes a traffic data database 22, which is connected to the big data database of the Internet of Vehicles. It stores a plurality of real-time traffic information, including real-time road speed, real-time road conditions, and monitors. (CCTV) video materials and more.

The driving behavior analysis system 2 of the present invention includes a positioning system analysis module 23, which obtains the following analysis information related to GPS positioning data based on GPS positioning data:

1. Driving trajectory: The positioning system analysis module 23 will correct and filter the latitude and longitude information in the positioning data through GPS. After removing the GPS drift coordinate information, you can draw the driving path trajectory on the map or use the road network information database 21 , Specifically describe the information of the driving section.

2. Acceleration and deceleration: The positioning system analysis module 23 uses GPS speed information to calculate the acceleration and deceleration between two positioning points through the acceleration and deceleration formula. If the acceleration and deceleration value exceeds the threshold specified by the system, the driving state It can be regarded as a state of rapid acceleration and deceleration, and it can be displayed with time or a map to show the time and location of rapid acceleration and deceleration.

3. Speeding and Turtle Speed: The positioning system analysis module 23 uses GPS speed information and road network information database 21 to determine whether the speed of the driving section exceeds the road speed limit; and compares it with the real-time speed data of the traffic data database 22 To determine whether the speed of the road when driving is a turtle speed Driving, for example, driving on a highway section, if the instantaneous average speed is 90km / hr, driving at a speed of 65km / hr on this section can be regarded as turtle speed driving behavior.

4. Illegal driving: The positioning system analysis module 23 applies the GPS driving path trajectory and the road traffic regulations in the road network information database 21 to determine whether the driving is illegal on the road, such as driving on a one-way street, prohibiting left-turning or turning. Make illegal turns or left turns.

The driving behavior analysis system 2 of the present invention includes a gravity sensing analysis module 24, which obtains the following analysis information according to the gravity sensor (G-sensor) data:

1. Driving state: The gravity sensing analysis module 24 analyzes the changes in the numerical values of the three axes of gravity X, Y, and Z. If the instantaneous change exceeds the threshold specified by the system, it can determine and obtain that the vehicle has undergone rapid acceleration and emergency Deceleration, sharp turns and other driving conditions.

2. Pavement status: The gravity sensing analysis module 24 applies the analysis of the average change values of the X, Y, and Z axes in each section to determine the road surface status of the driving. For example, if the average change value is large, it may indicate that the vehicle is driving. On rough roads; on the contrary, it may mean that the vehicle is driving on a smooth road.

3. Judgment of impact direction: The gravity sensing analysis module 24 judges the change amount and maximum value of the three-axis gravity of X, Y, and Z, and analyzes the possible impact direction. For example, if the maximum G-force value of the Y-axis gravity appears and the G-force changes instantly A large amount means that the vehicle may have a frontal impact or a rear impact; if the maximum G force value of the X-axis gravity appears and the instantaneous G force changes a lot, it indicates that the vehicle may have a side impact; if the Z-axis gravity changes the instant G force A large amount indicates that the vehicle may roll or overturn.

In addition, the driving behavior analysis system 2 of the present invention includes an image analysis module 25, which mainly extracts the i-frame frame of the driving image for image recognition and object detection processing, and finds the road at each time point of the driving image. The information to generate supporting information for driving behavior judgment is as follows:

1.Traffic signs and signs analysis: The image analysis module 25 uses the features such as color, shape and size in the object detection method to detect possible traffic signs and signs, and then uses object recognition and object comparison to determine traffic signs. The significance of the number and the sign, and the description in the form of image annotation in the overall driving records that can be used to generate reports.

2. Lane change analysis: The image analysis module 25 determines whether the vehicle deviates from the lane or switches lanes by detecting various lane lines. If the vehicle deviates from the lane or changes lanes, it will be marked with an image. The form is noted in the overall driving record that can be used to generate the report.

3. Vehicle distance analysis: The image analysis module 25 will judge based on the detection of vehicle objects and lane markings in front. It calculates the relative distance from the preceding vehicle according to the relative distance conversion, and records it based on the image time stamp. Say: The image analysis module 25 can be judged based on the yellow or white dashed line of the traffic sign marking line. The line segment of the marking line is as long as four meters and the spacing is six meters. The line width is ten centimeters to set the rules. This method estimates the distance between the vehicle and the vehicle in front.

The driving behavior analysis system 2 of the present invention includes a vehicle data analysis module 26, which receives vehicle peripheral device data in the driving data, such as: transmitted by the tire pressure detector 16 and the driving computer 15 and stored. The information can also be used to obtain the software and hardware status information of the vehicle.

1. Tyre pressure data analysis: As insufficient tire pressure and high tire pressure may be the cause of tire burst, the vehicle data analysis module 26 determines Whether the tire pressure pressure detected by the broken tire pressure detector 16 is lower than or higher than the normal tire pressure set value and the tire pressure change value of the detected tire can determine whether the tire has been inserted by a foreign body and caused a leak or an explosive tire. And other conditions for the benefit of accident analysis and judgment.

2. Driving computer data analysis: The vehicle data analysis module 26 can obtain vehicle software and hardware information through the driving computer 15 such as: direction signal, driving range, seat belt signal, braking signal, engine speed and wheel steering angle, etc. These data can be used as a benchmark for judging driving behaviors. For example, integrating the direction signal and wheel angle with GPS positioning data can analyze the user's behavior of using the direction signal; detecting the seat belt signal can analyze whether the user It is a habit to fasten the seat belt before starting the vehicle, or to fasten the seat belt when driving. The analysis of the throttle data can determine whether the driver accidentally stepped on the brakes in the event of an accident.

The driving behavior analysis system 2 of the present invention finally includes a driving behavior analysis module 27. The driving behavior analysis module 27 is an integrated module, which integrates and analyzes the data analyzed by the above items according to the time sequence. The driving behavior analysis module 27 presents the relationship between the positioning data, gravity data, image data, and vehicle equipment data through the time axis, and can accurately determine (emergency) acceleration and (emergency) braking when driving a vehicle. Driving behaviors such as turning, speeding, and speeding can also determine whether the driving behavior is changing lanes, overtaking, and driving distances are not maintained. The driving behavior analysis module 27 can also be given by a preset weight score. Driving behavior calculates and generates a driving behavior score, which is used as a reference value for driving behavior analysis.

The driving behavior analysis system 2 of the present invention may additionally include an accident detection module 28, which may be mainly based on the detection behavior analysis module 27 Judging the changes in the values of the three axes of gravity including X, Y, and Z. When the value of gravity changes drastically, it means that the vehicle may produce severe vibrations. Generally, the vibration of vehicles is not on rugged roads. The vehicle is caused by an accident; therefore, when the accident detection module 28 learns that the gravity value is abnormal, it can set an event time point and begin to trace the overall driving record of a specific time (for example, 120 seconds). Subsequent driving trajectory results in a state of no movement. The accident detection module 28 will automatically consider the accident to occur, and can issue a warning of the accident through a text message or action push message to let the surrounding police units, rescuers or other drivers confirm whether it is true. An accident occurred.

The driving behavior analysis system 2 of the present invention may further include an accident report module 29, which is connected to the accident detection module 28. When an accident occurs, the accident report module 29 will be based on the accident. The driving behavior data for a period of time before and after the occurrence of the accident report data may include:

1. Basic information about the accident, such as the time of the accident, the location of the accident (GPS coordinates, GIS road names, GIS road types and maps), and weather information.

2. Analysis information of driving behavior, such as: sharp braking, sharp turns, etc.

3. Speed information, such as driving behavior analysis information using GPS analysis, including speed change graphs, estimated emergency braking values when an accident occurs, and estimated braking distance.

4. Gravity sensing information, such as: triaxial gravity analysis using a gravity sensing module can estimate the extent of emergency braking and the state of vehicle displacement after an accident.

5. Driving distance information, such as the application of driving distance judgment information, combined with the braking time of the accident, can estimate the distance between the vehicle and the collision object before the accident.

6. Collision orientation judgment information, such as: gravity-induced triaxial gravity Analysis of data changes at the time of the accident can infer the direction in which the vehicle was impacted.

7. Accident images, such as key images between two minutes before and 30 seconds after the accident.

8. No. information, such as: No. 30 no. Identification record information before the accident.

9. Real-time image information, such as real-time image equipment information of the monitor surrounding the accident.

10. Other information, such as information such as police stations or hospitals or garages, and emergency contact information such as insurance contact numbers.

The accident report data can be immediately displayed to the driver or appraisal personnel of the vehicle, so that the system of the present invention can achieve the effects of timely rescue and objective assistance information for the accident.

Next, please refer to FIG. 2, which is a method step diagram of the driving behavior analysis method for an accident according to the present invention. The steps are as follows:

1. Step S21 Collection of driving data: mainly collecting data for the driving data database in the system.

2. Step S22 driving data analysis: It also includes the four steps of step S221 positioning system analysis, step S222 gravity sensing analysis, step S223 image analysis, and step S224 vehicle data analysis. Among them, there is no sequence relationship between the four steps. Is a program that can operate in parallel; and the four steps are completed by analyzing the data in the driving data database of the corresponding positioning system analysis module, gravity sensing analysis module, image analysis module, and vehicle data analysis module.

3. Step S23 driving behavior analysis: The driving behavior analysis module of the present invention integrates the data analysis performed by each module in step S22 to complete.

4. Step S24 performs accident detection: the accident detection module of the present invention determines whether there is an accident based on abnormal data to generate an accident occurrence warning.

5. Step S25 generates an accident report: The accident report module generates an accident report data when an accident occurs.

Please refer to FIG. 3 in the figure, which is a schematic diagram of the related aspects of the driving behavior analysis method for accidents according to the present invention. Among them, the driving data database 20, road network information database 21, and traffic in the system of the present invention can be seen. The data information database 22 is located in the data layer. These databases only have the function of collecting data and storing them; and the positioning system analysis module 23, gravity sensing analysis module 24, image analysis module 25, vehicle data analysis module 26, and The driving behavior analysis module 27 belongs to the analysis layer of the present invention and is the core of the present invention. Finally, it can be seen that the accident detection module 28 and the accident report module 29 are located in the event layer of the present invention and are the last accidents. The part of the event report when it occurs.

To sum up, the present invention is actually an innovation in technical thought, in order to fully expose a system and method for comprehensively judging driving behaviors by comprehensively integrating various objective auxiliary data and comprehensively collecting various passive information generated during vehicle travel, It can be seen that the present invention has fully met the requirements for novel and progressive statutory invention patents, and submitted a patent application in accordance with the law. I urge your office to approve this invention patent application to encourage inventions.

1‧‧‧ driving recorder

2‧‧‧Driving Behavior Analysis System

10‧‧‧ Positioning System Module

11‧‧‧Gravity Sensor Module

12‧‧‧Image capture module

13‧‧‧Storage Module

14‧‧‧Wireless Transmission Module

15‧‧‧trip computer

16‧‧‧ Tire Pressure Detector

20‧‧‧ Driving data database

21‧‧‧ Road Network Information Database

22‧‧‧Traffic Data Information Database

23‧‧‧ Positioning System Analysis Module

24‧‧‧ gravity sensing analysis module

25‧‧‧Image Analysis Module

26‧‧‧Vehicle data analysis module

27‧‧‧Driving Behavior Analysis Module

28‧‧‧ Incident Detection Module

29‧‧‧ Incident Report Module

Claims (12)

A driving behavior analysis system for an accident includes a driving data database that receives and stores a driving data data transmitted from at least one external driving recorder, and the driving data data includes settings for each Positioning data, vehicle speed data, gravity data, driving image data, and vehicle peripheral equipment data of the driving recorder; a road network information database, which stores a plurality of public road section data, and each road section data contains Link number, link location collection data, link direction data, link speed limit data, and link traffic specification data; a traffic data information database, which is linked to the external vehicle networking big data database, and stores multiple real-time traffic Information, each of the real-time traffic information includes real-time speed data, real-time road condition data, surveillance image data, and electronic toll data; a positioning system analysis module, the positioning system analysis module is used to analyze the vehicles in the traffic data database Location information to add this driving data The driving trajectory and vehicle speed data formed by the positioning data of the vehicle are compared with the relevant data of the road section passed by the vehicle in the road network information database to analyze the driving behavior of the vehicle's rapid acceleration or braking; a gravity sensing analysis Module, the gravity sensing analysis module is used to analyze the gravity data of the vehicle. It is used to track the value of the gravity change of the vehicle in the traffic data to analyze the road conditions of the vehicle and the rapid acceleration and urgency of the vehicle. Driving behavior during braking or sharp cornering, or the direction and extent of collision of the vehicle; An image analysis module, which is used to analyze the driving image of the vehicle in the driving data. It uses image recognition and object detection to detect traffic signs, traffic signs, lane changes, and calculate and vehicle. These passive driving data are separated from the vehicle in front; a vehicle data analysis module, the vehicle data analysis module analyzes the vehicle peripheral equipment data in the driving data data, mainly the tire pressure value and driving computer data analysis, the The vehicle data analysis module extracts data related to driving behavior from the vehicle peripheral device data, including the direction signal, driving gear, seatbelt signal, braking signal, engine speed, cooling water temperature, throttle operation status, wheels Steering angle, operation status of various active and passive assist systems; and a driving behavior analysis module, which receives the rapid acceleration of the vehicle analyzed by the positioning system analysis module and the gravity sensing analysis module , Sharp cornering or braking behavior, as well as passive driving data detected by the image analysis module, and the vehicle The data analysis module extracts data related to driving behavior. The driving behavior analysis module integrates the received data and each of the real-time traffic information obtained from the traffic data information database according to the time sequence and analyzes them by weight. To calculate a driving behavior score. The driving behavior analysis system for an accident, as described in the first patent application scope, further includes an accident detection module, which is connected to the driving behavior analysis module and judges whether the abnormal data is There is an accident to generate an accident alert. According to the driving behavior analysis system for an accident described in item 2 of the scope of patent application, it further includes an accident report module, the accident report module is connected to the accident detection module, and the accident report module is in the accident When it happens Driving behavior scores and warnings of the accident generate an accident report. A driving behavior analysis method for an accident includes the following steps: a vehicle data database receives and stores a vehicle data transmitted from at least one external vehicle recording device, and the vehicle data includes a load of each vehicle recorder. Location information, vehicle speed data, gravity data, driving image data, and vehicle peripheral equipment data; a network information database stores multiple public road segment data, each of which contains road segment number, road segment collection data, road direction data, Road speed limit data and road traffic specification data; a traffic data information database is connected to an external car network big data database, and the traffic data information database receives and stores a plurality of real-time traffic information, each of which includes real-time speed data, real-time Road condition data, surveillance image data, and electronic toll data; a positioning system analysis module passes the driving trajectory and speed data formed by the vehicle's positioning data in the driving data data to the road sections passed by the vehicle in the road network information database Comparison of relevant data A vehicle ’s drastic acceleration or braking behavior; a gravity-sensing analysis module tracks the vehicle ’s gravity change value in the driving data to analyze the road conditions of the vehicle ’s passage and analyze the drastic acceleration, braking, or Driving behavior during sharp turns, or the direction and degree of collision of the vehicle; an image analysis module detects the traffic signs and traffic signs in the driving image of the vehicle in the driving data through image recognition and object detection methods 2. Lane change and calculation of these passive driving data from the distance between the vehicle and the vehicle in front; A vehicle data analysis module analyzes the vehicle peripheral equipment data in the driving data data. It mainly uses tire pressure values and driving computer data analysis to extract data related to driving behavior from the vehicle peripheral equipment data, including direction lights, Driving gear position, seatbelt signal, brake signal, engine speed, cooling water temperature, throttle operation status, wheel steering angle, operation status of various active and passive assist systems; and a driving behavior analysis module receives the positioning system analysis module And the vehicle ’s rapid acceleration, cornering, or braking behavior analyzed by the gravity sensing analysis module, the passive driving data detected by the image analysis module, and the data extracted by the vehicle data analysis module. The driving behavior-related data is used to integrate the received data and each of the real-time traffic information obtained from the traffic data information database in accordance with the chronological order and analyze and integrate the weights to generate a driving behavior score through the weights. The driving behavior analysis method for an accident as described in item 4 of the scope of patent application, which further includes the following steps: an accident detection module is connected with the driving behavior analysis module, and the accident detection module determines the abnormality. Data to generate an alert for an accident. For example, the driving behavior analysis system for an accident described in item 5 of the scope of patent application, which further includes the following steps: an accident report module is connected to the accident detection module, and the accident report module will be based on the The driving behavior score and the warning of the accident generate an accident report data. For example, the driving behavior analysis method for an accident described in item 4 of the scope of the patent application, wherein the positioning system analysis module can perform overspeed determination, and first determine the road section of the vehicle through the positioning data of the vehicle in the driving data. , And then through each of the real-time data in the driving data of the road network information database Obtain the speed limit data of the road section where the vehicle is located in the traffic information. The positioning system analysis module then uses the vehicle speed data in the traffic data to determine whether the vehicle is overspeed, and then tracks the vehicle's overspeed positioning data and calculates the Travel distance with speeding data. According to the driving behavior analysis method for accidents as described in item 4 of the scope of the patent application, the positioning system analysis module can perform rapid acceleration and rapid braking judgment, mainly based on the continuous positioning of the positioning system analysis module according to the driving data. The acceleration value calculated from the change of the positioning data of the vehicle within the time, the positioning system analysis module then determines whether the acceleration value of the vehicle exceeds the preset rapid acceleration or rapid braking threshold value to indicate that the vehicle has undergone rapid acceleration and emergency Positioning data of the brakes and the number of rapid acceleration and rapid braking occurrences are counted. According to the driving behavior analysis method for an accident described in item 4 of the scope of patent application, the gravity sensing analysis module can perform rapid acceleration, braking or turning judgment, which is mainly based on the driving data of the gravity sensing analysis module The gravity change value calculated from the triaxial changes of the vehicle's gravity data in continuous time in the data. The gravity sensing analysis module then judges whether the vehicle's gravity change value exceeds the preset rapid acceleration, braking or turning threshold Value to indicate the gravitational data of the vehicle undergoing rapid acceleration, sudden braking or sharp turning, and to count the number of occurrences of rapid acceleration and sudden braking. According to the driving behavior analysis method for accidents as described in the scope of the patent application No. 4, the image analysis module can perform traffic sign and sign analysis, which is mainly the image analysis module uses the vehicle data in the vehicle data The driving image uses object detection to detect possible traffic signs and signs based on characteristics such as color, shape and size. The image analysis module compares traffic signs and signs with object recognition. The image analysis module does Indicate the time and content of traffic signs and signs passed by the vehicle. According to the driving behavior analysis method for an accident described in item 4 of the scope of patent application, the image analysis module can perform lane shift and transformation analysis, which is mainly based on the image analysis module using the vehicle data in the vehicle data. The driving image uses object detection to detect possible traffic markings based on characteristics such as color, shape, and size. The image analysis module calculates over time whether the change in traffic marking position offset value exceeds the lane deviation threshold. Value, the image analysis module also indicates the point in time when the vehicle lane shift or change occurred. According to the driving behavior analysis method for an accident described in item 4 of the scope of the patent application, the image analysis module can perform vehicle distance analysis, mainly the image analysis module passes the driving image of the vehicle in the driving data data. Object detection to detect possible traffic markings based on characteristics such as color, shape, and size. The image analysis module adjusts a distance to calculate correction parameters based on the type of markings. The image analysis module uses object detection to determine Whether there is a vehicle in front of the vehicle and the vehicle distance calculation correction parameters are used to estimate a vehicle distance. The image analysis module records the distance information of each vehicle over time.