Driving behavior analysis and displaying method

Abstract

A driving behavior analysis and display method, applicable to vehicles equipped with image sensors and used for fleet dispatch management, can transmit vehicle movement and driver behavior data to a server for driving behavior analysis. The method includes: obtaining actual driving data for a specific road segment, including driver behavior; obtaining driving rules corresponding to the specific road segment, which are variable; comparing the differences between the actual driving data and the driving rules; and presenting the data graphically according to the degree of difference.

Description

Driving behavior analysis and display methods The present invention relates to a method for analyzing driving behavior patterns, and more particularly to a method for displaying driving behavior analysis results. To improve driving safety, a great deal of research and implementation has been conducted on analyzing abnormal driving behavior. This is particularly true for fleet management platforms. Because so many vehicles need to be managed, designing an appropriate display interface is a challenge. The conventional design determines whether a driver is speeding by displaying the difference between the speed limit and the driving section of each road, and then creating a statistical chart based on the degree of the difference. Different color codes are set for specific speeding situations to achieve a graphical display of the speeding situation. Conventional design targets speeding based on the "fixed" speed limit of a specific road section. However, whether driving behavior is abnormal and potentially dangerous depends on the environmental conditions in which the vehicle is traveling. For example, if the speed limit on a specific road section is 50 km / h, it may be necessary to reduce speed to increase response time when encountering rain or unruly vehicles nearby. Therefore, conventional design does not meet the requirements. Furthermore, conventional design focuses on vehicle behavior as the management object, or the vehicle behavior exhibited by the driver when operating the vehicle. For example, if we say a driver likes to speed, we are observing the vehicle's speeding situation, not the driver's behavior. In other words, the danger of driving behavior is ultimately exhibited by the vehicle. The management focus is limited to the vehicle itself, ignoring the driver (human) factor. Therefore, the need to improve driving behavior through management is not met. The present invention is a method for analyzing and displaying driving behavior, which can change driving standards according to the driving (person) and driving (environment) conditions to accurately reflect whether the driving behavior is safe, which is one of the purposes of the present invention. A driving behavior analysis and display method, applicable to vehicles equipped with image sensors, can transmit vehicle movement and driver behavior data to a server for driving behavior analysis. The method includes: obtaining actual driving data of the vehicle corresponding to a specific road section, wherein the actual driving data includes driver (person) behavior; obtaining driving specifications corresponding to the specific road section, wherein the driving specifications are variable; comparing the actual driving data with the driving specifications; and presenting the differences in a graphical manner based on the degree (size) of the differences. The purpose of the present invention is to provide a design that is close to the goal of driving safety, thereby proposing the idea that driving standards can be changed. The change of driving standards varies with some factors, and how to measure these factors is a technical problem. This invention is a driving behavior analysis and display method that can be implemented within a computer software program. This software program can be installed on electronic devices such as servers, laptops, or mobile phones. It must have remote data transmission capabilities, such as LTE, Wi-Fi, or BT. (Product reference: https: / / www.mio.com / tw / products / car-camera / all-series / cdr / misentry-12t) and be connected to a device that collects image data, such as a dashcam. This dashcam has front and rear cameras that can capture driving images from the front and rear of the vehicle. (Product reference: https: / / www.mio.com / tw / products / car-camera / all-series / dual / mivue-783dual) The MiSentry™ 12T is a 4G LTE connected dashcam with dual front and interior cameras. It also features an internal camera that can capture the driver's upper body posture and movements. See: https: / / www.mio.com / tw / products / car-camera / all-series / cdr / misentry-12t Referring to Figure 1, the system architecture of this embodiment includes a dashcam 1, a rear camera 2, and onboard sensors 3. The dashcam 1 can be connected to the rear camera 2 via a wired or wireless connection. Wireless connection can be achieved through Ethernet, simplifying in-vehicle wiring. While cable connections must consider the in-vehicle environment, communication stability is high. Onboard sensors 3 include various sensors, such as temperature, pressure, and position, to determine the current vehicle status. They can be connected to the dashcam 1 via a controller area network 31. The dashcam 1 is mounted on the vehicle's windshield, near the driver's seat, typically next to the rearview mirror. It includes a front camera 11, an interior camera 12, a positioning chip 13, a wireless communication chip 14, and a processor 15. The front camera 11 and interior camera 12 each include an image sensor 111 and 121. Image sensors 111 and 121 receive image data captured by the front and interior cameras 11 and 12. The front camera 11 points toward the front of the vehicle, while the interior camera 12 points toward the interior of the vehicle, capturing images of the driver's face. The rear camera 2 is mounted inside the vehicle and located on the rear windshield, facing the rear of the vehicle. It includes an image sensor 21 that captures traffic images behind the vehicle, such as images of the following vehicle (rear vehicle), particularly the license plate of the following vehicle. The rear camera 2 is connected to the dashcam 1 via a wired or wireless connection. Traffic images captured by the rear camera 2 are transmitted to the dashcam 1's processor 15 via the image sensor 21. The front camera 11 , the interior camera 12 , the rear camera 2 , the vehicle sensor 3 , and the positioning chip 13 are all connected to the processor 15 . The processor 15 transmits processed data to the server 4 via the wireless communication chip 14 . Server 4 includes a wireless communication chip 41, a processor 42, and a database 43. The wireless communication chip 41 is connected to the wireless communication chip 14 of the dashcam 1. This data, including information about the vehicle's condition, traffic conditions in front of and behind the vehicle, and the driver's behavior, such as facial expressions, is integrated into the database 43 of server 4. Database 43 is connected to processor 42, and data from database 43 is processed by processor 42 and displayed on display 5. Database 43 includes map data 431, vehicle operation data 432, driver behavior data 433, and driving regulations data 434. Map data 431 may include geographic coordinates such as latitude and longitude, road speed limits, speed limit periods, road widths, bridge height limits, vehicle type restrictions, and road types. Map data 431 is compiled using pre-collected data and can be subsequently updated based on driver feedback. Vehicle operation data 432 is compiled using data acquired by the onboard sensors 3. Driver behavior data 433 primarily captures the driver's facial expressions through the interior camera 12, such as eye narrowing, reduced blinking frequency, and changes in head angle. Physiological sensors (not shown) may also be added to understand the physiological state of driving and make a comprehensive assessment of the driver's behavior. The vehicle operation status acquired through the onboard sensors 3 is combined with time, latitude and longitude coordinates, and road sections (map data) to create basic data for analyzing vehicle behavior. This is conventional technology and will not be described in detail here. Driving standard data 434 includes a road section (e.g., between No. 3 and No. 21 on Section 1 of Wenhua 3rd Road, or a road section separated by two intersections), a 90 km / h speed limit on that section, restrictions on large trucks and lane changes, and general driver requirements, such as keeping both hands on the steering wheel and not allowing the driver's view to deviate more than 30 degrees from the lane ahead. While the driving standard data 434 is stored on server 4, it can also be stored on dashcam 1. The driving standard adjustment of the present invention is executed by processor 15 of dashcam 1 or processor 42 of server 4. The dashcam 1 includes a positioning chip 13 (i.e., GPS) that can obtain the vehicle's coordinates. Based on the dashcam 1's journey (starting point to end point), playback software can be installed in the server 4's memory (not shown) to integrate Google Maps with the video of the corresponding location for simultaneous playback. The map includes speed limits for each road section, so the actual driving situation can be displayed on Google Maps to determine whether the vehicle is below the speed limit (marked in green). If the speed exceeds the speed limit by 10%, it is marked yellow, and if it exceeds the speed limit by 20%, it is marked red. Therefore, compliance with the driving speed and the degree of compliance can be simply displayed through color coding. The driver can also be categorized and counted by time duration, such as yellow for 30 minutes, red for 15 minutes, and so on. "Driving Standards" uses speed limits as an example. In this embodiment, the degree of compliance is divided into three levels, but this is not limited to this. If a noteworthy event occurs during the trip, it will be marked. For example, there is a front-vehicle collision risk event on the left side of the trip, a lane departure risk event on the right side, etc. Each risk event category can be set as a hyperlink, and users can click to view it. For example, if there are two front-vehicle collision risk events during the trip, the video of each event can also be clicked to view. In addition to following the speed limit, other behaviors of the driver may also change the level of compliance, for example: (1) The driver follows the speed limit but is distracted for a short time. Image recognition can be used to identify changes in the driver's line of sight or facial direction, and the level will not change at this time; (2) The driver follows the speed limit but is seriously distracted. The driver lowers his head, making it impossible to identify the driver's face, and the level will change from green to yellow; (3) The driver exceeds the speed limit by 10%. The sign is yellow, but because the driver is seriously distracted, the sign changes from yellow to red. This is an example to illustrate. The driver score can be set out of 100 and begins calculating after a driver exceeds, for example, 1,000 kilometers. The score is based on sensor data. Points are deducted for sudden acceleration or deceleration. Points are deducted based on the overall driving average, with the greater the deviation from the average, the more points are deducted. This includes, but is not limited to, various types of driving compliance (ADAS / LDWS / Stop & Go), following too closely, swerving, sudden braking, sharp turns, and other vehicle behaviors. Changes in the driver's score can be reported to the driver. Furthermore, the score changes can be highlighted, or notifications can be provided only for significant changes. It is particularly mentioned that driving regulations are centered on the perspective of driving safety. For example: (1) When the speed is high, the driver's reaction time is shortened, and the driving regulations need to adjust the distance between vehicles according to the speed. (2) When changing lanes, different driving regulations are required because of the distance between the front and rear vehicles and the speed. To confirm this, it is necessary to record the distance between the front and rear vehicles and the speed of each vehicle before changing lanes, and the reactions of the front and rear vehicles during the change process, such as whether a certain safe distance is gradually given to the vehicle changing lanes, and use this to define a "friendly traffic environment"; (3) When driving on the highway, the focus is on maintaining a safe distance. If the speed is lower than the average speed of the surrounding vehicles, unnecessary danger will arise due to the overtaking behavior of surrounding vehicles, and the highway's Driving standards should be to drive within the speed limit and try to stay within the average speed of the surrounding vehicles. Therefore, it is necessary to detect the speed of the surrounding traffic and calculate the average speed to define the driving standards. For example, the speed limit on the highway is 60-90 km / h, but the average speed of the surrounding vehicles is 85 km / h. The driving standard should be close to 85 km / h, rather than 60 km / h, which will not result in a ticket. The average speed of the surrounding vehicles can be calculated on the server through the speed data sent back by the mobile phone. If the speed of the vehicle is lower than the average speed, resulting in frequent rear-end collisions or overtaking by the rear vehicles, the driving standards should be adjusted; (4) The weave behavior of the adjacent vehicles ( in), the number of times a single vehicle encounters the above-mentioned snaking behavior on different sections or journeys, the frequency, the distance between the vehicle and the vehicle, and the speed relationship between the vehicles at that time can be used to judge the driving risk, and the occurrence of snaking behavior on a specific section or journey relative to the overall journey can be compared with the same definition. There are many statistical methods that can be used. In short, the degree of difference between the specific and the average is judged, and the driving standards are changed accordingly; (5) Driving standards are adjusted according to the vehicle type. For example, a large vehicle weighs more than a small vehicle. In the event of a collision, the damage to other vehicles may be greater because the momentum of the large vehicle is greater than that of the small vehicle. Therefore, for certain sections of road, large vehicles may be restricted from entering or at least required to reduce the speed limit. The vehicle type is input by the vehicle driver, and the driving standards corresponding to the vehicle type can be replaced by external statistical data. The embodiment (3) of the driving standard adjustment is supplemented with the following explanation: the vehicle detects whether the rear vehicle is tailgating through the rear camera 2, and then determines whether the front vehicle is moving away from the vehicle through the front camera 11, that is, the front vehicle is moving away from the vehicle, but the rear vehicle is approaching the vehicle. The judgment method is to obtain the license plate of the front vehicle or the rear vehicle. Because the license plate has a standard size and has a specific shape, size and text content, the error rate of image recognition is low. The change in the size of the license plate within a certain period of time is used to determine the speed difference between the two vehicles. For example, if the distance between the front vehicle and the vehicle is 20 meters, the license plate occupies 1 pixel of the screen. The number of pixels is 20 in length and 12 in width. The distance between the front vehicle and the vehicle in front is 10 meters. The number of pixels occupied by the license plate on the screen is 40 in length and 20 in width. Based on the relationship between the number of pixels on the screen and the distance in the real world, the distance and speed changes between the vehicle in front and the vehicle in front are calculated. If the speed of the vehicle obtained from on-board sensor 3 is 80 km / h, but the license plate of the rear vehicle becomes larger and the license plate of the front vehicle becomes smaller, and the speed limit on this road section is 90 km / h, then the driving regulations require the vehicle to increase its speed to 90 km / h. When driving regulations change, such as increasing the speed from 80 km / h to 90 km / h, the changes in the license plates of the leading and trailing vehicles are again obtained. For example, if the leading vehicle's license plate becomes 5% smaller and then 2% smaller, it can be considered that the distance and speed between the leading and trailing vehicles have decreased. If the trailing vehicle's license plate becomes 8% larger and then 2% larger, it can be considered that the distance and speed between the leading and trailing vehicles have decreased. Alternatively, the appropriate distance between vehicles can be determined based on vehicle speed. This can be used to infer the appropriate size of the license plate on the screen under these conditions and whether this size should be maintained, thereby determining the effectiveness of the driving regulation change. If the vehicle does not fully comply with the driving regulation adjustment, such as only increasing the speed to 85 km / h, this can be recorded as a failure to comply with the driving regulation adjustment, or recorded as a 50% increase based on the speed increase for subsequent compliance statistics. Vehicle speed can be detected by the speedometer of the onboard sensor 3 or estimated by the positioning chip 13 of the driving recorder 1 and transmitted to the database 43 of the server 4 via the wireless communication chip 14 for recording. The embodiment (4) of the driving standard adjustment supplements the following explanation: regarding the judgment of weave in, please refer to the patent certificate No. I757964 proposed by the applicant. It is mainly based on the frequency of the front or rear vehicle entering the control area of ​​the vehicle. This case further records the situation in which the vehicle encounters the weave in behavior of the front and rear vehicles. In combination with time and road section, the total number of weaves in a certain road section can be divided by the average value of time, and then the total number of weaves in a certain period is accumulated. If it is found that the accumulated total exceeds the average, the running road section is defined as a dangerous road section. It is judged whether the speed of the vehicle is lower than the speed limit of the road section and the proportion of the speed limit. If it is lower than the speed limit by more than 10%, and the situation of the rear vehicle approaching and moving away from the front vehicle does not exceed the threshold, the driving standard is recommended to be adjusted to shorten the distance between vehicles and increase the degree of driver involvement in driving, such as canceling the fixed speed following function. This is calculated by the processor 41 of the server 4 in combination with the database 43. The calculation result and the recommendation are transmitted through the wireless communication chip 41. The data is transmitted to the driving recorder 1, and the onboard sensor 3 records the vehicle's behavior and uses it to observe whether the driver complies with driving regulations. For example, if the distance between vehicles is reduced, the distance between vehicles can be inferred through the image data obtained by the front camera 11. Each of the above suggestions can be considered as a driving regulation, and the number of violations is counted to determine the driver's compliance level. Driving regulations may change due to environmental changes. For example, the normal speed limit on a specific road section is 80 km / h, but the special speed limit is reduced to 50 km / h during heavy rain or snow. Or, if there is a car accident or fallen rocks on the road, the TMC accident data obtained by the dashcam or through the central control center may cause the driving regulations on a specific road section to change. For such temporary environmental changes, the driver's compliance and cooperation measures have an impact on driving safety and are therefore also within the scope of driving behavior analysis. However, it does not deal with immediate risk prevention, so there is no need to consider the data calculation period or even the processing burden. It is particularly mentioned that the interior camera is used to record the driver's movements. Currently, the common interior and exterior cameras are both installed on the dash cam, which is installed on the vehicle's front windshield or interior rearview mirror. Therefore, the driving shooting range is mainly around the driver's face. The judgment of whether the driver has at least one hand off the steering wheel is based on whether the interior camera can see the hand around the driver's face. However, the interior camera can also be replaced with a 360-degree panoramic lens, which can be installed in the center of the car to also shoot the steering wheel. The use of driver ratings can be combined with incentive mechanisms, such as assigning more tasks to drivers with high scores and giving them bonuses, etc.; for routes or tasks where driving standards are expected to change more frequently, drivers with a "high degree of compliance with driving standard changes" can be assigned to handle them. The present invention considers both the driver's qualities and external environmental factors, combining these two factors to define driving standards. The driving conditions are analyzed based on these driving standards, and the analysis results are then displayed. The driving standards in this case can be adapted to specific circumstances, thus enabling a reasonable and appropriate evaluation of good or bad driving behavior. Furthermore, the driving rules of the present invention include quantitative data definitions, such as: the type of road with a sharp turn, the distance before the curvature of the road, and the driver's deceleration amount. If the weather changes, the above driving rules will change accordingly, such as requiring further deceleration at the same location, or the distance before the location before deceleration. The above quantitative data can be obtained through various software and hardware such as GPS, G sensor, Gyro, and map data, and the obtained quantitative data can be used as a reference to judge good and bad drivers.

Claims

1. A driving behavior analysis and display method, applicable to vehicles including at least an image sensor, for transmitting vehicle data back to a server for driving behavior analysis, comprising: (a) obtaining actual driving data of the vehicle for a specific road segment, wherein the actual driving data includes driver (person) behavior; (b) obtaining driving rules corresponding to the specific road segment, wherein the driving rules are variable, and the method for determining the driving rules includes: (b1) Obtain current driving environment condition data, including but not limited to real-time weather conditions, road slippage, real-time traffic flow density, or traffic accident data; (b2) Based on the current driving environment condition data obtained in step (b1), dynamically calculate and determine at least one safety parameter benchmark value, including but not limited to the minimum safe following distance or recommended driving range for vehicle speed; (b3) Based on the dynamically determined safety parameter benchmark value, form and update the driving rules corresponding to the specific road segment, which are defined by the safety parameter benchmark value and serve as the evaluation benchmark for subsequent driving behavior analysis; (c) Compare the differences between actual driving data and the driving rules; (d) Present the analysis results graphically according to the severity of the differences in step (c).

2. A driving behavior analysis and display method as described in claim 1, wherein the driving rules mentioned in step (b) are adjusted based on data obtained from an "image sensor", and the data includes the situation inside and outside the vehicle.

3. A driving behavior analysis and display method as described in claim 2, wherein both inside and outside the vehicle are violations of the driving rules, will increase the severity of step (d) and change the presentation result.

4. A driving behavior analysis and display method as described in claim 2, wherein the image sensor obtains the frequency of the vehicle being weave-in to define the nearby "traffic" situation and adjusts the driving rules accordingly.

5. A driving behavior analysis and display method as described in claim 1, wherein the driving rules are changed based on real-time traffic information obtained by the central control center.

6. A driving behavior analysis and display method as described in claim 2 or 5, wherein step (d) presents the analysis results before and after the change of the driving regulations in a two-stage manner.

7. A driving behavior analysis and display method as described in claim 1, wherein the driving rules are adjusted based on road curvature data from map data.

8. A driving behavior analysis and display method as described in claim 1, wherein the driving specifications are adjusted according to vehicle model information.

9. A driving behavior analysis and display method as described in claim 6, further comprising step (e) feeding back the analysis results of step (d) and the "next task dispatch decision factors" to the individual driver.

Images