CN115527289A

CN115527289A - Intelligent targeted training method based on driving behaviors and storage medium

Info

Publication number: CN115527289A
Application number: CN202211327564.7A
Authority: CN
Inventors: 周业俊
Original assignee: Chongqing Tibi Network Technology Co ltd
Current assignee: Chongqing Tibi Network Technology Co ltd
Priority date: 2022-10-27
Filing date: 2022-10-27
Publication date: 2022-12-27

Abstract

The invention relates to the field of safe driving behavior training, in particular to an intelligent targeted training method based on driving behaviors and a storage medium. The method comprises the following steps: s1: collecting driving behavior data; s2: analyzing the driving behavior data, wherein the analyzed content comprises illegal driving behaviors and the identity of a driver; s3: informing a driver according to the analysis result of the S2; s4: generating safety training content according to the analysis result of the S2 in a matching mode; s5: and (5) checking the driver. The technical scheme can be used for training in a targeted manner, the training effectiveness is improved, and the training effect is guaranteed.

Description

Intelligent targeted training method based on driving behaviors and storage medium

Technical Field

The invention relates to the field of safe driving behavior training, in particular to an intelligent targeted training method based on driving behaviors and a storage medium.

Background

With the continuous development of economy, the demand of society on road transportation increases day by day, more and more road transportation vehicles participate in operation, the problem of road transportation safety is not of great variety, and according to the requirements of relevant laws and regulations, road transportation enterprises should establish safety education training and assessment systems of enterprise employees.

The popularization of the internet is benefited, the online safe driving training platform is gradually popularized, and the road transportation enterprise formulates a training plan through the training platform every month to inform all practitioners to learn; when dangerous driving behaviors of the vehicle are acquired, the training plan is added through the training platform, then the added training plan is informed to a driver registered correspondingly to the vehicle to learn, the driver can automatically select time and place to complete learning according to actual conditions, and the problems that the traditional off-line training mode is not high in efficiency, inconvenient to participate and the like are solved.

However, the existing training platform still has the following problems:

1. the safety training content only includes general contents such as laws and regulations, major traffic safety incidents and the like, the whole safety training content has training significance, but the pertinence is not strong, and the training effectiveness needs to be enhanced;

2. generally, each transport vehicle can register a plurality of drivers, the conventional collection of dangerous driving behaviors of the vehicles can only be accurate to the vehicles, which driver implements the dangerous driving behaviors cannot be determined, learning tasks can be distributed to other registered drivers which do not implement the dangerous driving behaviors, and the drivers can generate adverse psychology, thus conflict is generated on learning, and the effectiveness of training is greatly reduced;

3. the operations of obtaining dangerous driving behaviors of the vehicle, making a training plan, notifying and the like generally depend on manual work, and perfect automatic adaptation is not carried out, so that the timeliness and the accuracy of training are insufficient, and the working benefit is reduced.

4. The driver finishes browsing and learning all safety training contents on the training platform, namely training is finished, and the platform does not adopt a data analysis method to check and evaluate the learning effectiveness subsequently so as to continuously improve the effectiveness of the whole training system.

Disclosure of Invention

Aiming at the defects in the prior art, the invention aims to: the technical scheme can carry out targeted training, improve training effectiveness and ensure training effect.

In order to solve the problems, the invention provides a basic scheme that: an intelligent target training method based on driving behaviors and a storage medium comprise the following steps: s1: collecting driving behavior data; s2: analyzing the driving behavior data, wherein the analyzed content comprises illegal driving behaviors and the identity of a driver; s3: informing a driver according to the analysis result of the S2; s4: generating safety training content according to the analysis result of the S2 in a matching mode; s5: and (5) checking the driver.

The beneficial effects of the basic scheme are as follows: at present, technical inertia and prejudice exist in the industry, illegal driving behaviors are managed by taking vehicles as units, and management is not performed by taking human as a core, and according to the technical scheme, drivers who implement the behaviors are accurately identified through S2, so that the illegal driving behaviors of the drivers collected and analyzed through S1 and S2 can correspond to the drivers who actually implement the illegal driving behaviors, the drivers can be accurately managed in the subsequent process, and meanwhile, the corresponding drivers can be accurately notified based on the notification of the result S3 of accurate identification of the identities of the drivers.

In addition, the prior art adopts general courseware without targeted training, and the scheme adopts a scheme of managing by human core, so that S4 can generate targeted safe training contents according to the conditions of different drivers, and the learning effectiveness is improved; therefore, S5 can also carry out targeted assessment, thereby supervising and urging careful learning and ensuring the learning effect.

Further, the S2 specifically includes:

s21: identifying specific dangerous driving behaviors according to the driving behavior data;

s22: identifying a driver implementing dangerous driving behaviors according to the driving behavior data;

s23: evaluating whether the dangerous driving behaviors identified in the S21 need reminding or timely training;

s24: and carrying out spot check on the identification results of the S21, the S22 and the S23.

According to the technical scheme, the specific dangerous driving behaviors are identified through S21 and used as the basis of subsequent judgment, and the driver implementing the dangerous driving behaviors is accurately locked through S22, so that misjudgment is avoided, and management is facilitated by taking a human as a core; s23, evaluating whether the behavior needs reminding or timely training or not so as to trigger the corresponding step in S3; s24, performing spot check on the identification result to ensure the accuracy of the identification algorithm and the evaluation algorithm.

Further, the sources of the driving behavior data collected in the step S1 include a monitoring device in the vehicle and a third-party monitoring platform outside the vehicle, the driving behavior data include driving image recording data, vehicle positioning data and vehicle condition information data, and the step S21 calls the driving behavior to comprehensively identify the driving behavior data, so that dangerous driving behaviors are finally obtained.

According to the technical scheme, the driving behavior data in the vehicle is obtained through the monitoring device in the vehicle, the driving behavior data outside the vehicle is obtained through the third-party monitoring platform, the data source is widened, and the scheme is more accurate and objective when dangerous driving behaviors are judged.

According to the technical scheme, dangerous driving behaviors are judged in various different modes, and the sensitivity of judging the dangerous driving behaviors is enhanced.

Further, the S22 includes:

s22-1: judging the identity of the driver by carrying out face recognition on the driving behavior data acquired in the step S1;

s22-2: acquiring the identity of a driver by calling information of a transportation task list;

s22-3: and comparing and analyzing the results obtained by the S22-1 and the S22-2, and finally confirming the identity of the driver.

According to the technical scheme, the identity of the driver is judged in two modes respectively, and comparison and analysis are performed to check the recognition results of the two previous steps, so that the accuracy of the identity judgment of the driver is improved.

Further, the S23 specifically includes:

s23-1: grading and scoring the dangerous driving behaviors according to the recognition result of the S21;

s23-2: judging whether reminding is needed or not according to the rating result of the S23-1, and matching a reminding strategy if the reminding is needed;

s23-3: judging whether training and learning are needed to be carried out in time or not according to the grading result of the S23-1, if so, skipping S23-4 to continue execution, and if not, continuing execution according to the flow;

s23-4: and accumulating the scores of the S23-1, judging whether the accumulated scores are larger than a preset value, if so, continuing to execute according to the flow, otherwise, executing to S3 and ending.

According to the technical scheme, whether reminding needs to be carried out or not is judged through a rating mechanism, different reminding modes are matched for different behaviors, and differential management is carried out, so that the reminding mode is more reasonable; judging whether training needs to be carried out in time or not through a grading mechanism; and a grading accumulation mechanism is set, the behaviors which do not need to be trained are graded and accumulated, the behaviors are implemented for a plurality of times, so that the training can still be triggered when the grading reaches a set threshold value, and different differential management is adopted for dangerous driving behaviors with different severity degrees, so that the reminding and the training are more reasonable.

Further, the step S23-2 matches different reminding modes according to the severity of the rating result, matches the dangerous driving behavior reminding strategy with the severe rating result as immediate reminding, and matches the dangerous driving behavior reminding strategy with the slight rating result as post reminding.

According to the technical scheme, for dangerous driving behaviors with serious rating results, the driver is immediately reminded of implementing the dangerous driving behaviors, so that accidents are avoided as much as possible; for dangerous driving behaviors with slight rating results, the driver can be distracted by reminding immediately because accidents cannot be caused, and therefore the driver can be reminded to pay follow-up attention by reminding afterwards.

Further, the safety training content generated in S4 includes supporting electronic data of the implemented dangerous driving behavior and multimedia data of a third person perspective acquired from a third-party supervision platform, where the multimedia data includes images recorded of the dangerous driving behavior of the driver.

In the technical scheme, the safety training content is matched with the implemented dangerous driving behaviors, so that the data is more targeted; the data source is increased, the data form is expanded, the data content is richer, the learning effect is enhanced, and the training effectiveness is improved.

Further, the content of the S5 examination is associated with the safety training content generated by the S4, and the difficulty and the learning duration of the S5 examination refer to the rating and scoring results obtained by the S23.

In the technical scheme, the assessment content is related to the safety training content, and a driver can be supervised to carefully learn the safety training content; the rating and scoring results obtained in the S23 influence the difficulty of the S5 assessment, so that the difficulty setting is more reasonable, the technical scheme can enhance the learning effect and improve the training effectiveness.

Further, the method also comprises the step S6: driver behavior is continuously tracked, and courseware is reversely scored based on learning effectiveness.

According to the technical scheme, S6 is managed by a human core based on the steps, the subsequent driving behaviors of the driver can be continuously tracked, the effectiveness of safety training contents is checked, and the training effect is ensured.

Further, the storage medium has stored thereon a computer program which, when being executed by a processor, carries out the steps of a method for intelligent targeted driving behavior-based training as claimed in any one of claims 1 to 9.

In the technical scheme, the computer program on the storage medium can execute any one of the intelligent target training methods based on the driving behaviors, and the method can be guaranteed to run and implement on different computer equipment normally.

Drawings

Fig. 1 is a logic diagram of an intelligent target training method based on driving behaviors and a storage medium according to the invention.

Detailed Description

The technical scheme of the application is further explained in detail through the following specific implementation modes:

example one

Fig. 1 shows an intelligent target training method based on driving behavior, which includes the following steps:

s1: the driving behavior data are collected, and the data sources for collecting the driving behaviors in the S1 comprise an in-vehicle supervision device and a third-party supervision platform outside the vehicle. In this embodiment, all the acquisition behaviors are automatically acquired through configured interfaces, and the in-vehicle monitoring device includes, but is not limited to, a camera, a vehicle data recorder, a vehicle-mounted positioning terminal of a GPS or beidou satellite, an instrument panel, a vehicle gyroscope, an acceleration sensor, an ultrasonic sensor, and a vehicle-mounted bio-detector.

The camera is used for collecting video pictures and dynamic and static images in the vehicle, including facial images of a driver, driving behavior images of the driver and the like; the automobile data recorder is used for shooting video pictures and dynamic and static images near the vehicle, including traffic signs, traffic light information and the like; the vehicle-mounted positioning terminal of the GPS or Beidou satellite is used for positioning the vehicle, and can acquire information such as a vehicle running track and a vehicle speed; the instrument panel is used for collecting information such as vehicle speed, vehicle fault alarm, residual energy and the like; the driving gyroscope is used for sensitively sensing the vehicle driving direction and the acceleration sensor is used for acquiring and providing real-time acceleration information when the vehicle travels; the ultrasonic sensor is used for detecting the distance between the ultrasonic sensor and surrounding pedestrians, vehicles and obstacles; the vehicle-mounted biological detector is used for sensing the physiological and psychological conditions of a driver.

The data provided by the third-party supervision platform outside the vehicle comprises but is not limited to video pictures and dynamic and static images of the vehicle in running, which are shot by a third-party camera, and violation information of the vehicle. The data serve as the basis for S2 analysis of dangerous driving behaviors and serve as the content of S4 safety training for the driver to review. The driving behaviors are collected through various channels, and the dangerous driving behaviors can be evaluated in multiple dimensions, so that more kinds of dangerous driving behaviors can be judged by the technical scheme, and the accuracy of a judgment result is also ensured.

S2: analyzing the data acquired by the S1, specifically comprising:

s21: automatically identifying specific dangerous driving behaviors according to the driving behavior data, and specifically comprising the following steps:

s21-1: and (2) carrying out image recognition on video pictures and dynamic and static images acquired by the camera, the automobile data recorder and the third-party supervision platform in the S1 vehicle, wherein the image recognition is carried out through an image recognition algorithm, and the dangerous driving behaviors of the driver are comprehensively judged by automatically calling a corresponding algorithm mainly through subjective information such as actions and emotions of the driver and objective information such as the external environment condition of the vehicle and the driving state of the vehicle. Dangerous driving behaviors which can be judged by the scheme include but are not limited to unbelted safety belts, fatigue driving, smoking in driving, call receiving and making, fatigue driving, distraction driving and reverse driving.

S21-2: the driving track of the vehicle can be automatically obtained through the data of the vehicle-mounted positioning system, and whether the vehicle runs on a specified route or not can be easily judged according to the driving track; meanwhile, the vehicle speed and other information can be acquired, whether the vehicle is driven at a specified speed can be judged, and comprehensive study and judgment should be carried out by combining information collected by an instrument panel and an acceleration sensor when the vehicle speed is judged, so that the judgment accuracy is ensured. The types of dangerous driving behaviors that can be judged by the above scheme include, but are not limited to, the above two.

S21-3: the driving state of the vehicle can be comprehensively reflected by automatically acquiring the vehicle condition information recorded by various sensors, instrument panels and detectors, and the information can verify the dangerous driving behaviors judged by S21-1 and S21-2 from another angle. For example, the situation that the driver wears the safety belt is judged through the image recognition of S21-1, but the safety belt alarm is displayed through the instrument panel information acquired by the scheme, so that the situation that the driver does not actually wear the safety belt or the safety belt detection device is in failure can be further judged, the follow-up processing is facilitated, and the condition that the dangerous driving behavior that the driver does not wear the safety belt is not judged is avoided. In addition, this scheme also can independently discern dangerous driving action, and the type that can discern includes but not limited to acceleration in violation of rules and regulations, vehicle trouble not in time maintain, "fill air vehicle".

S22: automatically identifying a driver implementing dangerous driving behaviors according to the driving behavior data, specifically comprising:

s22-1: s1, an in-vehicle camera collects a video picture and a dynamic and static image, wherein the picture comprises a driver face image, and the driver face information in the driver face image is automatically compared with the driver face information recorded in advance in a background server database by using a face recognition algorithm to judge the identity of a driver; it is worth noting that when the driver inputs the face information, the face information without wearing glasses, sunglasses and a mask and the face information when wearing glasses, sunglasses and a mask should be simultaneously input, so that the driver identity can be accurately identified when the face of the driver is shielded, and the driver identity can be more accurately identified.

S22-2: the technical scheme also comprises the step of acquiring the identity of a driver driving the vehicle at the current time period in a mode of automatically calling information of the background transportation task list, wherein the transportation task list should be recorded into a background server in advance.

S22-3: comparing the driver identity judged by the S22-1 through face recognition with the driver identity called in the S22-2, if the two results are the same, determining that the driver identity is correctly locked, and if the two results are different, performing manual processing, and finally confirming the driver identity by manual according to the image information acquired by the camera in the vehicle in the S1, the called transportation task list and other information; meanwhile, for the situation of difference caused by S22-1 face recognition error, marking is carried out, and reasons are analyzed to optimize a face recognition algorithm; in addition, according to the technical scheme, whether the driver has a designated driving condition or whether the driver finishes a scheduling task can be checked if necessary.

S23: and evaluating whether the dangerous driving behaviors identified in the S21 need to be reminded or not and whether timely training is needed or not, wherein the specific steps comprise:

s23-1: according to the identification result of the S21, the dangerous driving behaviors are rated and graded, in the embodiment, automatic grading and grading are carried out through a preset model, the model can be changed and maintained according to follow-up requirements, the dangerous driving behaviors are divided into a plurality of grades in advance, the dangerous driving behaviors are divided into slight, general and serious according to different danger degrees, the corresponding grade is automatically matched for the behaviors after the dangerous driving behaviors are identified in the S21 so as to represent the danger degree of the dangerous driving behaviors, for example, behaviors such as rear-end collision, too close distance to a pedestrian, reverse driving and the like are set to be serious, behaviors such as small-amplitude overspeed, solid line pressing and the like are set to be general, and behaviors such as mobile phone use, conversation and the like are set to be slight; meanwhile, according to the technical scheme, the dangerous driving behavior is scored according to the identification result of the S21, and the scoring is comprehensively judged according to the preset basic score of the behavior, the amplitude of the behavior exceeding a specified standard value, the duration of the behavior, the occurrence frequency of the behavior and the like. The rating result will be bound and associated with the driver identity recognized at S22.

S23-2: whether reminding is needed or not is automatically judged according to the rating result of 23-1, in the embodiment, the serious and slight ratings are judged to be required to be reminded, and a suitable reminding strategy is matched; the reminding strategies comprise immediate reminding and after-event reminding, the dangerous driving behavior reminding strategy with serious grade is matched with the immediate reminding, the dangerous driving behavior reminding strategy with slight grade is matched with the after-event reminding, and the after-event reminding is to remind after the driving task is finished; the purpose of immediately reminding the driver of the serious behavior is to remind the driver to take measures immediately as much as possible to avoid accidents; the purpose of alerting minor actions after the end of a driving task is that these actions, although very minor and without immediate training, violate regulations, and that the driver is brought to follow-up attention by a post-alert because immediate alerts may also be distracting to the driver. The general behaviors do not need to be reminded immediately because the drivers are in a stage of needing more energy to control the general behaviors such as small-amplitude overspeed, and the reminding of the drivers can divert the attention of the drivers to distract the drivers and cause dangers to happen instead, so the behaviors can trigger a follow-up training mechanism generally, but except for other behaviors happening at the moment, the reminding is not carried out.

S23-3: according to the scheme, the trigger rule is set according to scores of different risk degrees of driving behaviors, training can be triggered immediately when the score of the dangerous driving behaviors reaches a set threshold value, the step S23-4 is skipped, subsequent steps are directly executed, the dangerous driving behaviors with high scores such as general grades and severe grades of reverse driving, fatigue driving and the like are set, a driver is added into a list needing training and triggers the training immediately, other slight-grade behaviors such as mobile phone use, conversation and the like cannot trigger the training immediately, and the steps are only continuously executed until the step S3 is finished. In addition, the score of the step is correspondingly deducted after the step S5 passes the assessment, so as to avoid repeated learning.

S23-4: in the above step S23-3, the dangerous driving behaviors whose scores do not meet the set trigger rule criteria accumulate the scores obtained in the step S23-1, and the accumulation situations include single behavior accumulation and composite behavior accumulation. The single behavior accumulation means that an accumulated score threshold value is set for a single dangerous driving behavior, and when the accumulated score of the single dangerous driving behavior reaches the set threshold value, training and learning need to be carried out in time; the composite behavior accumulation means that a total accumulated score threshold value is set, and when the sum of accumulated scores of a plurality of different dangerous driving behaviors reaches the threshold value, training and learning need to be carried out in time, and subsequent steps are continuously executed. It should be noted that, the scores in this step are accumulated by taking the driver as a unit, and the accumulated scores are also deducted after the assessment is passed in S5.

S24: because the S21, the S22 and the S23 are identified by the algorithm by using the artificial intelligence technology and the results are automatically obtained, the identification results need to be continuously checked, and the accuracy of the results is ensured; the identification error problem discovered by spot check should be marked, and the algorithm is optimized in time, so that the accuracy of algorithm identification is improved; in particular, since S22-4 in S22 has already been processed correspondingly to the recognition error, S24 only checks the recognition result with the same comparison result of S22-4; in addition, the problem of recognition error found in the driver complaint process should be marked and dealt with in time.

S3: the vehicle-mounted management and control equipment further comprises an alarm device for reminding and informing a driver according to the analysis result of the S2; the S3 specifically includes:

s31: for the dangerous driving behaviors matched as immediate reminding in the S23-2 step, the driver is immediately and automatically reminded through an alarm device of the vehicle-mounted management and control equipment, wherein the alarm device comprises a loudspeaker and can inform the driver of the specific dangerous driving behaviors in a voice broadcasting mode; in addition, to the built-in vehicle that is equipped with the display screen of car, can also link with this display screen during the alarm, utilize characters, sign etc. of striking colour to indicate, avoid the driver to play tone volume too big cover voice broadcast's sound, guarantee as far as possible that the driver in time receives the alarm, guarantee to drive safety.

S32: and for dangerous driving behaviors with a reminding strategy of reminding afterwards, reminding is carried out after the driving task is finished, wherein the reminding mode comprises but is not limited to short messages and emails, and the driver is required to reply after reminding to be regarded as successful reminding.

S33: the technical scheme can automatically send out a notice to the driver needing to learn through the analysis of S23-3 or S23-4, wherein the notice is in a form including but not limited to short messages, e-mails and message pushing in a safe driving training platform station so as to remind the driver to participate in learning in time. The content of the notification includes, but is not limited to, a list of conducted dangerous driving behaviors, a rating of conducted dangerous driving behaviors, and a score of conducted dangerous driving behaviors.

The dangerous driving behavior list comprises a certain driving behavior or a certain driving behavior set implemented by the driver, and the driver can clearly know the dangerous driving behavior implemented by the driver through the list; the dangerous driving behavior rating can inform the driver of the degree of danger of the dangerous driving behavior implemented; the dangerous driving behavior scoring prompts reference data used for calculating the scoring, such as the amplitude of the behavior exceeding a specified standard value, the duration of the behavior, the occurrence frequency of the behavior and the like, so that the aim of warning a driver before training is fulfilled, and meanwhile, the driver can take the dangerous driving behavior scoring oral liquid, thereby avoiding the generation of adverse psychology and reducing the learning effect.

S4: the safe training content carries out automatic matching generation according to dangerous driving behaviors analyzed and recognized by the S2, and specifically comprises the following steps:

s41: the generated safety training content comprises matched electronic data of the implemented dangerous driving behaviors and multimedia data of a third person perspective acquired from a third-party supervision platform. The matched electronic data is compiled in advance aiming at different dangerous driving behaviors, the content of the electronic data comprises but is not limited to relevant laws and regulations, and major accidents caused by the relevant dangerous driving behaviors, and the form of the electronic data comprises but is not limited to courseware, PDF and matched explanation videos. The third-party supervision platform acquires multimedia data of a third person perspective, wherein the multimedia data comprises a monitoring video, the video records dangerous driving behaviors implemented by a driver at the third person perspective, so that the driver can see the dangerous driving behaviors implemented when driving by himself at the third person perspective, and the driver can see the danger of the behaviors by changing the third person perspective; compared with the prior art, the technical scheme enriches the training forms, widens the training content and improves the training effectiveness.

S42: the safety training content is automatically uploaded to a safety driving training platform, so that a driver can conveniently learn on line.

S43: before the safety training content is sent to the driver for learning, the safety training content which is learned before the driver and is examined through the S5 step is intelligently filtered.

S5: the assessment and the grade deduction are carried out after the driver finishes learning, and the method specifically comprises the following steps:

s51: the driver is examined after learning, the examination is in an online or offline test mode, the examination is mainly performed on a safe driving training platform in an online mode for convenience, but the examination can be performed strictly in an offline mode for behaviors with serious ratings or other behaviors with strict grazing performance; the examination forms include but are not limited to single-choice questions, multiple-choice questions, judgment questions, short-answer questions, case analysis questions and law statement analysis questions, the examination questions form a question bank in advance with reference to the safety training content generated by S4, the questions are intelligently selected from the question bank according to the set examination question extraction rules during examination, and meanwhile, the question bank is continuously maintained in the follow-up process, and the maintenance mode includes but is not limited to adding new questions and modifying conflict questions.

As a preferred embodiment, when forming the question bank, the examined questions can also mark difficulty in advance, and when examining, the driver can extract difficult problems and simple questions in different proportions according to the rating and scoring results of dangerous driving behaviors obtained through S23; when the rating is more serious or the score is higher, the proportion of the difficult problems is obviously improved, the proportion of the simple problems is correspondingly reduced, and when the rating is more slight or the score is general, the proportion of the simple problems is correspondingly improved, and the proportion of the difficult problems is obviously reduced; this technical scheme can supervise the comparatively serious driver of dangerous driving action to study carefully, and the dangerous driving action of the work of correcting oneself improves the validity of training.

In addition, the dangerous driving behavior rating and grading result obtained in the S23 also determines the learning duration of the driver, and the grading factor relates to the times of implementing the dangerous driving behavior, the amplitude of the behavior exceeding a specified standard value and the duration of the behavior, so that the learning duration and the grading hook can match the training method with the behavior of the driver, the training planning can be performed on the driver in a more targeted manner, and the training effectiveness can be improved.

S52: for the driver passing the assessment, the corresponding score of the training assessment of the driver is automatically deducted, and it needs to be noted that the corresponding deduction is only carried out on the score of the dangerous driving behavior which has triggered the training and passes the assessment, and the scores are continuously accumulated and not deducted for other dangerous driving behaviors which are still in the accumulation stage. The action accumulated score deduction mechanism mainly avoids repeated learning of a driver and avoids negative emotion of the driver caused by repeated learning, and the operation mechanism of the system is guaranteed to be healthy and beneficial. After deduction of the cumulative score, the driver is removed from the list of drivers requiring training.

S6: after training is finished, marking the driver, continuously tracking the behavior of the driver, automatically calling behavior analysis data of the driver, and detecting whether the same dangerous driving behavior is implemented again to evaluate the learning effectiveness; according to the technical scheme, a threshold value is set for each dangerous driving behavior, if the number of drivers still implementing the dangerous driving behavior after safety training reaches the threshold value, the safety training content aiming at the dangerous driving behavior is judged to be invalid, maintenance and adjustment are needed, a training system is continuously optimized, and the effectiveness of the whole training system is continuously improved.

A storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of a driving behavior based intelligent targeted training method as described above.

It will be understood by those skilled in the art that all or part of the processes for implementing the intelligent driving behavior based targeted training method can be implemented by a computer program instructing associated hardware, where the computer program can be stored in a non-volatile computer readable storage medium, and when executed, the computer program can include processes of the various embodiments of the intelligent driving behavior based targeted training method. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), rambus (Rambus) direct RAM (RDRAM), direct Rambus Dynamic RAM (DRDRAM), and Rambus Dynamic RAM (RDRAM), among others.

The foregoing are embodiments of the present invention and are not intended to limit the scope of the invention to the particular forms or details of the embodiments shown, and are not intended to limit the scope of the claims, which follow as broadly described herein, but may be defined by the appended claims, along with the full scope of equivalents to which such claims are entitled. It should be noted that, for those skilled in the art, without departing from the structure of the present invention, several variations and modifications can be made, which should also be considered as the protection scope of the present invention, and these will not affect the effect of the implementation of the present invention and the utility of the patent. The scope of the claims of the present application shall be determined by the contents of the claims, and the description of the embodiments and the like in the specification shall be used to explain the contents of the claims.

Claims

1. An intelligent target training method based on driving behaviors is characterized in that: the method comprises the following steps:

s1: collecting driving behavior data;

s2: analyzing the driving behavior data, wherein the analyzed content comprises illegal driving behaviors and the identity of a driver;

s3: informing a driver according to the analysis result of the S2;

s4: generating safety training content according to the analysis result of the S2 in a matching mode;

s5: and (5) checking the driver.

2. The intelligent targeted training method based on driving behaviors as claimed in claim 1, wherein the method comprises the following steps: the S2 specifically comprises the following steps:

s24: and performing spot check on the identification results of the S21, the S22 and the S23.

3. The intelligent targeted training method based on driving behaviors as claimed in claim 2, wherein the method comprises the following steps: the sources of the driving behavior data collected in the S1 comprise an in-vehicle monitoring device and an out-vehicle third-party monitoring platform, the driving behavior data comprise driving image recording data, vehicle positioning data and vehicle condition information data, and the driving behavior is called in the S21 to comprehensively identify the driving behavior data, so that dangerous driving behaviors are finally obtained.

4. The intelligent targeted training method based on driving behaviors as claimed in claim 3, wherein the training method comprises the following steps: the S22 includes:

5. The intelligent targeted training method based on driving behaviors as claimed in claim 4, wherein the training method comprises the following steps: the S23 specifically includes:

s23-1: grading and grading dangerous driving behaviors according to the recognition result of the S21;

6. The intelligent targeted training method based on driving behaviors as claimed in claim 5, wherein: and the S23-2 matches different reminding modes according to the severity of the rating result, matches the dangerous driving behavior reminding strategy with the severe rating result into immediate reminding, and matches the dangerous driving behavior reminding strategy with the slight rating result into post-event reminding.

7. The intelligent targeted training method based on driving behaviors as claimed in claim 6, wherein: and the safety training content generated by the S4 comprises matched electronic data of the implemented dangerous driving behaviors and multimedia data of a third person weighing view angle acquired from a third-party supervision platform, and the multimedia data comprises images recorded on the dangerous driving behaviors of the driver.

8. The intelligent targeted training method based on driving behaviors as claimed in claim 7, wherein the method comprises the following steps: the content of the S5 examination is associated with the safety training content generated in the S4, and the difficulty and the learning duration of the S5 examination refer to the rating and scoring results obtained in the S23.

9. The intelligent targeted training method based on driving behaviors of claim 8, wherein: further comprising step S6: driver behavior is continuously tracked, and courseware is reversely scored based on learning effectiveness.

10. A storage medium, characterized by: the storage medium has stored thereon a computer program which, when executed by a processor, carries out the steps of a method of intelligent targeted driving behavior-based training as claimed in any one of claims 1 to 9.