Driver training student class hour detection management method and system based on intelligent technology

Abstract

The invention relates to the technical field of driving training class hour management, in particular to a driving training student class hour detection management method and system based on an intelligent technology. The method comprises the following steps: installing intelligent class hour detection equipment on a coach vehicle, and collecting the position, action state and facial recognition image data of a student in real time; performing local processing on the acquired image data by using an edge calculation method, extracting key features, encrypting and transmitting the key features to a data center; and analyzing the data transmitted to the data center, judging whether the trainee is in the vehicle, whether the driver on the vehicle is the trainee himself, whether the trainee is immobile for a long time and whether the trainee is in the main driving position, and automatically calculating and recording the class hour according to the judgment result. According to the driving training student hour detection management method and system based on the intelligent technology, the effective training hour of a driving school student is obtained, the authenticity of student training hour statistics is ensured, driving school hour counterfeiting is effectively prevented, and meanwhile the requirements for network bandwidth and transmission efficiency are lowered.

Description

Technical Field

[0001] The present invention relates to the technical field of driving training hours management, and in particular to a driving training student hours detection management method and system based on intelligent technology. Background Art

[0002] In the traditional motor vehicle driver training industry, the detection and management of students' study hours mainly rely on manual records and supervision, which has many problems. First, manual records are prone to errors and it is difficult to ensure the accuracy and completeness of the data. Secondly, manual supervision is inefficient and cannot monitor students' learning progress and study hours in real time and comprehensively. In addition, traditional methods lack effective data analysis and feedback mechanisms, making it difficult to provide scientific and reasonable training suggestions for driving schools and students. In addition, during the driving training process, when students do not have enough study hours or do not practice on-site, they may resort to cheating and forge their identities to steal study hours when they are not present.

[0003] For example, in the patent with patent announcement number CN106557792A and name "Method and system for screening and verifying effective study hours of driving trainees", the method includes: obtaining the start time and end time of the current study hour; counting the training time of the trainees in the current study hour; calculating the average speed of the vehicle during the training time, and judging whether it is greater than the preset speed, if so, marking it as suspicious study hour; judging whether there is an intersection between the training time period and the trainee's other training time periods, if there is an intersection, marking the current study hour as suspicious study hour; if there is no intersection, matching the student's front face facial feature values ​​obtained in each study hour period within the training time with the basic model feature values, if the matches are the same, marking the study hour period as effective study hour, otherwise, it is suspicious study hour; counting the sum of the study hour periods with the same matches, which is the total effective study hour of the current study hour. The disadvantages are: it combines fingerprint recognition, face recognition, satellite positioning, speed monitoring and other technical means to verify the learning hours. It needs to collect and transmit a large amount of data such as students' fingerprints, faces, satellite positioning, and training duration in real time and upload them. The process is complicated and has high requirements for network bandwidth and transmission efficiency, which increases the risk of data leakage. It also cannot handle situations where students are in the car but are not operating the car themselves. Summary of the invention

[0004] In view of the problems in the prior art that the process of driving training students' learning hours detection and management is complicated and the requirements for network bandwidth and transmission efficiency are high, the present invention provides a driving training students' learning hours detection and management method and system based on intelligent technology, which can obtain the effective training hours of driving school students, ensure the authenticity of students' training hours statistics, effectively prevent driving school learning hours fraud, and reduce the requirements for network bandwidth and transmission efficiency.

[0005] In order to achieve the above technical purpose, a technical solution provided by the present invention is a method for managing the learning hours of driving trainees based on intelligent technology, comprising the following steps: S1, install intelligent learning time detection equipment on the training car to collect the trainee's location, movement status and facial recognition image data in real time; S2, uses edge computing methods to process the collected image data locally, extracts key features and encrypts and transmits them to the data center; S3, analyzes the data transmitted to the data center to determine whether the student is in the car, whether the person learning to drive in the car is the student himself, whether the student has been immobile for a long time, and whether the student is in the main driving position, and automatically calculates and records the learning hours based on the judgment results; S4 monitors the student’s driving operation data in real time, compares it with the preset standard actions, identifies abnormal actions, issues abnormal reminders, and sends back relevant data for analysis.

[0006] In this technical solution, by installing intelligent learning hours detection equipment on the training car, the location, action status and facial recognition image data of the trainees are collected in real time and comprehensively, ensuring the original authenticity and integrity of the learning hours data. The edge computing method is used to quickly process the image data locally, which not only extracts key features, but also effectively reduces the amount of data transmission, reduces the dependence on network bandwidth, and improves the transmission efficiency. It can accurately judge whether the trainee is in the car, whether the person learning to drive in the car is the trainee himself, whether the trainee is active, and whether the trainee is indeed in the main driving position, so as to automatically calculate and record the real and effective learning hours. This process effectively prevents the falsification of driving hours in driving schools and ensures the fairness and accuracy of learning hours statistics. At the same time, it also monitors the students' driving operation data in real time, compares it with the preset standard actions in detail, can promptly discover and identify abnormal actions, immediately issue abnormal reminders, and send back relevant data for further analysis. This not only helps to correct the students' wrong operations in a timely manner and improve the quality of training, but also further enhances the intelligence and refinement of learning hours management.

[0007] The present invention is further configured as follows: the automatic calculation and recording of study hours according to the judgment result includes: When the student passes identity verification and is confirmed to be in the main driving position in the car and performing normal driving operations, the time starts automatically and is recorded as valid learning hours; If the student leaves the main driving position, identity authentication fails, or abnormal behavior is detected, the learning hours will be suspended and marked as an abnormal period.

[0008] In this technical solution, when the student successfully passes the identity verification and is confirmed by the system to be sitting in the main driving position in the car and performing standard driving operations, the system will automatically start the timing function and record this period as the student's effective learning hours, ensuring that the learning hours will only be effectively calculated when the student actually practices driving in the car, thereby greatly improving the authenticity and reliability of the learning hour data. At the same time, once the student leaves the main driving position, the identity verification fails, or the system detects any abnormal driving behavior of the student, the system will immediately suspend the learning hour calculation and mark the time period as an abnormal period, which can not only effectively prevent the driving school from falsifying learning hours, but also promptly discover and correct the student's irregular operations, providing a powerful means of supervision for the driving school, while also ensuring the quality of student training and the fairness of learning hour statistics.

[0009] The present invention is further configured as follows: in step S3, the step of determining whether the trainee has been immobile for a long time comprises: Obtain the student's historical behavior data, pre-process it to distinguish valid behavior data from invalid behavior data, calculate the dynamic behavior anomaly coefficient threshold and the static behavior anomaly coefficient threshold based on the valid behavior data, and judge whether the student's behavior is abnormal based on the comparison result between the student's current behavior data and the threshold. Prompt when the number of abnormal behaviors reaches the preset threshold.

[0010] In this technical solution, the system will first collect the historical behavior data of the students, which covers the various actions and states of the students during the driving training process. Subsequently, the system will pre-process these historical data and accurately distinguish between valid behavior data and invalid behavior data through advanced algorithms and technical means. Next, the system will calculate the dynamic behavior abnormality coefficient threshold and the static behavior abnormality coefficient threshold based on the valid behavior data. These thresholds represent the reasonable range of dynamic and static behaviors that students should reach during normal driving training. During real-time monitoring, the system will compare the current behavior data of the students with these thresholds. If the student's behavior data exceeds the set threshold range, the system will judge its behavior as abnormal. When the number of abnormal behaviors accumulates to the preset threshold, the system will immediately issue a prompt to remind relevant personnel to pay attention to the abnormal state of the student.

[0011] The present invention is further configured such that when the student's historical behavior data is invalid behavior data, user habit data is output based on the invalid behavior data, and when the student's current behavior data does not match the user habit data, the student is judged to be abnormal and a prompt is given.

[0012] In this technical solution, when the historical behavior data of students collected by the system is judged as invalid behavior data, these seemingly useless data actually contain specific habits or patterns of students. The system will cleverly use these data to generate user habit data that reflects the unique behavior habits of students. In the subsequent real-time monitoring process, the system will carefully compare the current behavior data of students with the user habit data. Once it is found that the current behavior of the student is significantly inconsistent with his previous habit data, the system will immediately trigger the abnormal judgment mechanism, consider the current state of the student to be abnormal, and quickly send out a prompt signal. This not only improves the depth of the system's understanding of student behavior, but also significantly enhances the sensitivity and accuracy of abnormal detection, so that the system can more accurately capture subtle changes in student behavior, and promptly discover and remind relevant personnel to pay attention to possible problems or risks of students, thereby ensuring the safety and efficiency of the driving training process. At the same time, by making full use of invalid behavior data, the system realizes the comprehensive utilization and value maximization of data, and provides strong support for the intelligent management of the driving training industry.

[0013] The present invention is further configured as follows: the calculation of the dynamic behavior abnormal coefficient threshold and the static behavior abnormal coefficient threshold includes: dividing the historical dynamic behavior data into historical dynamic abnormal data and historical dynamic normal data, calculating the first abnormal coefficient of the historical dynamic abnormal data compared with the historical dynamic normal data, and using the minimum abnormal coefficient as the dynamic behavior abnormal coefficient threshold; dividing the historical static behavior data into historical static abnormal data and historical static normal data, calculating the first abnormal coefficient of the historical static abnormal data compared with the historical static normal data, and using the minimum abnormal coefficient as the static behavior abnormal coefficient threshold.

[0014] In this technical solution, first, the system will conduct an in-depth analysis of the historical dynamic behavior data and accurately divide it into two categories: historical dynamic abnormal data and historical dynamic normal data. Then, through advanced algorithms, the system will calculate the first abnormal coefficient of the historical dynamic abnormal data compared to the historical dynamic normal data. This coefficient reflects the degree of difference between abnormal and normal in dynamic behavior. In order to ensure the rationality and sensitivity of the threshold, the minimum value of these abnormal coefficients will be selected as the dynamic behavior abnormal coefficient threshold. Similarly, for historical static behavior data, the system will perform similar divisions and processing. After dividing the historical static behavior data into historical static abnormal data and historical static normal data, the system will calculate the first abnormal coefficient between the two and select the minimum value as the static behavior abnormal coefficient threshold.

[0015] The present invention is further configured as follows: the calculation of the first abnormal coefficient includes: Based on the matching degree of the student's trajectory and motion sequence, the matching value of the historical dynamic abnormal data and the historical dynamic normal data is output, and the abnormal coefficient is calculated.

[0016] In this technical solution, the system first extracts the student's driving trajectory and motion timing information, which includes key dynamic behavior features such as the student's position change, speed change, steering wheel rotation, etc. during driving. Subsequently, the system uses an efficient matching algorithm to compare these features with predefined standards or historical normal data to output a matching value. This matching value reflects the similarity or difference between the student's current dynamic behavior and normal behavior. Based on this matching value, the abnormality coefficient can be further calculated. The abnormality coefficient is a quantitative indicator used to measure the degree to which the student's dynamic behavior deviates from normal behavior. By comparing the matching values ​​of historical dynamic abnormal data and historical dynamic normal data and calculating the difference between them, an abnormality coefficient that accurately reflects the degree of abnormality of the student's behavior can be obtained.

[0017] The present invention is further configured as follows: in step S3, the step of judging whether the person learning to drive on the bus is the trainee himself comprises: Pre-enter the students’ facial feature data at the driving school to establish a student facial database; The facial features in the image data are extracted and compared with the student facial database to determine whether the person learning to drive in the car is the student himself.

[0018] In this technical solution, first, the driving school will pre-enter the students' facial feature data when they sign up, and establish a complete student facial database. This database contains the unique facial information of each student, providing an accurate and reliable comparison basis for subsequent identity verification. When the student starts to learn to drive, the system will extract the facial features from the image data in the car in real time. By using a high-precision facial recognition algorithm, the system will compare the extracted facial features with the information in the student's facial database one by one. In this process, the system will comprehensively consider multiple key feature points of the face, such as eyes, nose, mouth, etc., to ensure the accuracy and reliability of the comparison. If the comparison results show that the facial features of the person learning to drive in the car are highly matched with a record in the student's facial database, the system will determine that the person is the student himself, and continue to record his learning time data. On the contrary, if the comparison results do not match or there is doubt, the system will issue a prompt to remind relevant personnel to further verify the identity of the student to ensure the compliance and safety of the driving training process.

[0019] The present invention is further configured as follows: in step S3, the step of determining whether the trainee is in the main driving position includes: Install a seat position sensor under the main driver's seat of the training vehicle; When the seat position sensor detects that the main driver's seat is occupied, it determines whether the person occupying the seat is a trainee based on the recognition results of the image data, and determines whether the relative position relationship between the trainee's body position and the main driver's seat meets the preset conditions to confirm whether the trainee is in the main driver's position.

[0020] In this technical solution, first of all, a seat position sensor is carefully installed under the main driver's seat of the training car. This sensor can sensitively sense the occupancy status of the seat and provide key information for subsequent judgments. When the seat position sensor detects that the main driver's seat is occupied, the system will immediately start a further recognition process. It will combine the image data captured by the camera in the car and use advanced image recognition technology to confirm the identity of the person occupying the seat to determine whether it is the trainee himself. Not only that, the system will also carefully analyze the relative relationship between the trainee's body position and the main driver's seat. By comparing the preset conditions, such as the trainee's sitting posture, the distance between the body and the steering wheel, the position of the feet on the pedals, etc., the system can accurately determine whether the trainee is actually in the main driver's position and whether it meets the standard requirements for driving operations.

[0021] The present invention is further configured as follows: in step S3, the step of determining whether the trainee is in the vehicle comprises: Install GPS positioning module, infrared sensor and high-definition camera on the training vehicle; The vehicle location information is obtained in real time through the GPS positioning module, and the human body heat is detected through the infrared sensor to preliminarily determine whether the trainee is in the vehicle; When it is preliminarily determined that the student is in the car, the high-definition camera is activated to capture the image inside the car, and the deep learning algorithm is used to perform image recognition on the image to detect whether there is a student in the car, as well as the student's facial features or body features, to further confirm whether the student is in the car.

[0022] In this solution, first of all, the GPS positioning module, infrared sensor and high-definition camera are carefully installed on the training car, which together constitute the detection system. The GPS positioning module can provide the vehicle's location information in real time, providing basic data support for determining whether the trainee is in the car. Then, the infrared sensor plays its unique advantages. It can preliminarily determine whether there is someone in the car by detecting the heat emitted by the human body. This non-contact detection method is not only highly sensitive, but also can be carried out without affecting the trainee's driving. When it is preliminarily determined that the trainee may be in the car, the system will immediately start the high-definition camera to capture the real-time picture of the car. These pictures are then sent to the deep learning algorithm for image recognition. The algorithm will carefully analyze every detail in the picture to detect whether the trainee's figure appears, and will also compare the trainee's facial features or body features to ensure the accuracy of the judgment.

[0023] The driving training student learning hours detection management system based on intelligent technology is used to implement the driving training student learning hours detection management method, including: Intelligent learning time detection equipment: installed on the training vehicle, used to collect the trainee's location, movement status and facial recognition data in real time; Data center: used to receive, store and process data transmitted by intelligent learning time detection equipment; Study time detection module: using big data analysis technology to automatically detect and record students' study time, and identify abnormal actions or violations; Personalized training suggestion module: Through artificial intelligence technology, intelligent evaluation of students' driving operations is carried out to provide students with personalized training suggestions; Feedback and interaction module: used to provide real-time feedback of learning hours test results and personalized training suggestions to driving schools and students, supporting two-way interaction and communication; Abnormal reminder and data feedback module: In abnormal situations, abnormal reminders are sent to relevant personnel and relevant data are returned for further analysis.

[0024] In this technical solution, the driving training student learning hours detection and management system based on intelligent technology integrates intelligent learning hours detection equipment, data center, learning hours detection module, personalized training suggestion module, feedback and interaction module, and abnormal reminder and data return module. The system collects the student's position, movement and facial data in real time through the intelligent device on the training car. The data center receives and processes this data. The learning hours detection module uses big data analysis technology to automatically detect learning hours and identify abnormal behaviors. At the same time, the personalized training suggestion module uses artificial intelligence technology to provide students with customized training suggestions. The feedback and interaction module ensures real-time feedback of learning hours results and training suggestions, and supports two-way communication. In abnormal situations, the abnormal reminder and data return module will immediately send reminders and return data for further analysis, fully realizing efficient and accurate learning hours detection and management.

[0025] The beneficial effects of the present invention are as follows: (1) obtaining the effective training hours of driving school students, ensuring the authenticity of the students' training hours statistics, effectively preventing driving school hours fraud, and reducing the requirements for network bandwidth and transmission efficiency; (2) being able to accurately determine whether the student is in the car, whether the person learning to drive in the car is the student himself, whether the student remains active, and whether the student is actually in the main driving position, thereby automatically calculating and recording the real and effective training hours. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 The present invention is a flow chart of a method for managing driving training student learning hours based on intelligent technology. DETAILED DESCRIPTION

[0027] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific implementation method described herein is only an optimal embodiment of the present invention, which is only used to explain the present invention and does not limit the scope of protection of the present invention. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0028] like Figure 1 As shown, as the first embodiment of the present invention, the driving training student learning hours detection management method based on intelligent technology includes the following steps: S1, install intelligent learning time detection equipment on the training car to collect the trainee's location, movement status and facial recognition image data in real time; S2, uses edge computing methods to process the collected image data locally, extracts key features and encrypts and transmits them to the data center; S3, analyzes the data transmitted to the data center to determine whether the student is in the car, whether the person learning to drive in the car is the student himself, whether the student has been immobile for a long time, and whether the student is in the main driving position, and automatically calculates and records the learning hours based on the judgment results; S4 monitors the student’s driving operation data in real time, compares it with the preset standard actions, identifies abnormal actions, issues abnormal reminders, and sends back relevant data for analysis.

[0029] In this embodiment, by installing intelligent learning hours detection equipment on the coach car, the location, action status and facial recognition image data of the trainees are collected in real time and comprehensively, ensuring the original authenticity and integrity of the learning hours data. The edge computing method is used to quickly process the image data locally, which not only extracts key features, but also effectively reduces the amount of data transmission, reduces the dependence on network bandwidth, and improves the transmission efficiency. It can accurately judge whether the trainee is in the car, whether the person learning to drive in the car is the trainee himself, whether the trainee remains active, and whether the trainee is indeed in the main driving position, so as to automatically calculate and record the real and effective learning hours. This process effectively prevents the falsification of driving hours in driving schools and ensures the fairness and accuracy of learning hours statistics. At the same time, the driving operation data of the trainees is also monitored in real time, and compared with the preset standard actions in detail, abnormal actions can be discovered and identified in time, abnormal reminders can be given immediately, and relevant data can be sent back for further analysis. This not only helps to correct the trainees' erroneous operations in time and improve the quality of training, but also further enhances the intelligence and refinement of learning hours management.

[0030] It can be understood that the use of edge computing methods to locally process the collected image data, extract key features and encrypt and transmit them to the data center includes: preliminary preprocessing of the received image data, including denoising and image enhancement; using the computing resources on the edge computing device to extract features of the preprocessed image data; before transmitting the extracted key features to the data center, encrypting the feature data using an advanced encryption algorithm; compressing the encrypted feature data; and using the communication module on the edge computing device to transmit the encrypted and compressed feature data to the data center through a secure channel.

[0031] The automatic calculation and recording of study hours according to the judgment results includes: When the student passes identity verification and is confirmed to be in the main driving position in the car and performing normal driving operations, the time starts automatically and is recorded as valid learning hours; If the student leaves the main driving position, identity authentication fails, or abnormal behavior is detected, the learning hours will be suspended and marked as an abnormal period.

[0032] When the student successfully passes the identity verification and is confirmed by the system to be sitting in the main driving position in the car and performing standard driving operations, the system will automatically start the timing function and record this period as the student's effective learning hours, ensuring that the learning hours will only be effectively calculated when the student actually practices driving in the car, thereby greatly improving the authenticity and reliability of the learning hours data. At the same time, once the student leaves the main driving position, fails the identity verification, or the system detects any abnormal driving behavior of the student, the system will immediately suspend the learning hours calculation and mark the time period as an abnormal period, which can not only effectively prevent the driving school from falsifying learning hours, but also promptly discover and correct the students' irregular operations, providing a powerful means of supervision for the driving school, while also ensuring the quality of students' training and the fairness of learning hours statistics.

[0033] In step S3, the step of determining whether the trainee has been immobile for a long time includes: Obtain the student's historical behavior data, pre-process it to distinguish valid behavior data from invalid behavior data, calculate the dynamic behavior anomaly coefficient threshold and the static behavior anomaly coefficient threshold based on the valid behavior data, and judge whether the student's behavior is abnormal based on the comparison result between the student's current behavior data and the threshold. Prompt when the number of abnormal behaviors reaches the preset threshold.

[0034] The system will first collect the trainee's historical behavior data, which covers the trainee's various actions and states during the driving training process. Subsequently, the system will pre-process these historical data and accurately distinguish between valid and invalid behavior data through advanced algorithms and technical means. Next, the system will calculate the dynamic behavior abnormality coefficient threshold and the static behavior abnormality coefficient threshold based on the valid behavior data. These thresholds represent the reasonable range of dynamic and static behaviors that the trainee should reach during normal driving training. During real-time monitoring, the system will compare the trainee's current behavior data with these thresholds. If the trainee's behavior data exceeds the set threshold range, the system will determine that his behavior is abnormal. When the number of abnormal behaviors accumulates to the preset threshold, the system will immediately issue a prompt to remind relevant personnel to pay attention to the trainee's abnormal state.

[0035] When the student's historical behavior data is invalid behavior data, the user habit data is output based on the invalid behavior data. When the student's current behavior data does not match the user habit data, the student is judged to be abnormal and a prompt is given.

[0036] When the historical behavior data of students collected by the system is judged as invalid behavior data, these seemingly useless data actually contain specific habits or patterns of students. The system will cleverly use this data to generate user habit data that reflects the unique behavior habits of students. In the subsequent real-time monitoring process, the system will carefully compare the current behavior data of students with the user habit data. Once it is found that the current behavior of students is significantly inconsistent with their previous habit data, the system will immediately trigger the abnormal judgment mechanism, deeming the current state of students abnormal, and quickly send out a prompt signal. This not only improves the depth of the system's understanding of student behavior, but also significantly enhances the sensitivity and accuracy of abnormal detection, allowing the system to more accurately capture subtle changes in student behavior, and promptly discover and remind relevant personnel to pay attention to possible problems or risks of students, thereby ensuring the safety and efficiency of the driving training process. At the same time, by making full use of invalid behavior data, the system realizes the comprehensive utilization and value maximization of data, providing strong support for the intelligent management of the driving training industry.

[0037] The calculation of the dynamic behavior abnormal coefficient threshold and the static behavior abnormal coefficient threshold includes: dividing the historical dynamic behavior data into historical dynamic abnormal data and historical dynamic normal data, calculating the first abnormal coefficient of the historical dynamic abnormal data compared to the historical dynamic normal data, and taking the minimum abnormal coefficient as the dynamic behavior abnormal coefficient threshold; dividing the historical static behavior data into historical static abnormal data and historical static normal data, calculating the first abnormal coefficient of the historical static abnormal data compared to the historical static normal data, and taking the minimum abnormal coefficient as the static behavior abnormal coefficient threshold. First, the system will conduct an in-depth analysis of the historical dynamic behavior data and accurately divide it into two categories: historical dynamic abnormal data and historical dynamic normal data. Then, through advanced algorithms, the system will calculate the first abnormal coefficient of the historical dynamic abnormal data compared to the historical dynamic normal data. This coefficient reflects the degree of difference between abnormal and normal in dynamic behavior. In order to ensure the rationality and sensitivity of the threshold, the minimum value of these abnormal coefficients will be selected as the dynamic behavior abnormal coefficient threshold. Similarly, for historical static behavior data, the system will also perform similar divisions and processing. After dividing the historical static behavior data into historical static abnormal data and historical static normal data, the system calculates the first abnormal coefficient between the two and selects the minimum value therebetween as the static behavior abnormal coefficient threshold.

[0038] The calculation of the first abnormal coefficient includes: Based on the matching degree of the student's trajectory and motion sequence, the matching value of the historical dynamic abnormal data and the historical dynamic normal data is output, and the abnormal coefficient is calculated. The system first extracts the student's driving trajectory and motion sequence information, which includes key dynamic behavior characteristics such as the student's position change, speed change, steering wheel rotation, etc. during driving. Subsequently, the system uses an efficient matching algorithm to compare these characteristics with predefined standards or historical normal data to output a matching value. This matching value reflects the similarity or difference between the student's current dynamic behavior and normal behavior. Based on this matching value, the abnormal coefficient can be further calculated. The abnormal coefficient is a quantitative indicator used to measure the degree to which the student's dynamic behavior deviates from normal behavior. By comparing the matching values ​​of the historical dynamic abnormal data and the historical dynamic normal data and calculating the difference between them, an abnormal coefficient that accurately reflects the degree of abnormality of the student's behavior can be obtained.

[0039] In step S3, the step of determining whether the person learning to drive on the bus is the trainee himself includes: Pre-enter the students’ facial feature data at the driving school to establish a student facial database; The facial features in the image data are extracted and compared with the student's facial database to determine whether the person learning to drive in the car is the student himself. First, the driving school will pre-enter their facial feature data when the students register and establish a complete student facial database. This database contains the unique facial information of each student, providing an accurate and reliable comparison basis for subsequent identity verification. When the student starts to learn to drive, the system will extract the facial features in the image data in the car in real time. By using a high-precision facial recognition algorithm, the system will compare the extracted facial features with the information in the student's facial database one by one. In this process, the system will comprehensively consider multiple key feature points of the face, such as eyes, nose, mouth, etc., to ensure the accuracy and reliability of the comparison. If the comparison results show that the facial features of the person learning to drive in the car are highly matched with a record in the student's facial database, the system will determine that the person is the student himself and continue to record his learning time data. On the contrary, if the comparison results do not match or there is doubt, the system will issue a prompt to remind relevant personnel to further verify the identity of the student to ensure the compliance and safety of the driving training process.

[0040] In step S3, the step of determining whether the trainee is in the main driving position includes: Install a seat position sensor under the main driver's seat of the training vehicle; When the seat position sensor detects that the main driver's seat is occupied, combined with the recognition results of the image data, it is determined whether the person occupying the seat is a trainee, and whether the relative position of the trainee's body position and the main driver's seat meets the preset conditions to confirm whether the trainee is in the main driver's position. First, a seat position sensor is carefully installed under the main driver's seat of the training car. This sensor can sensitively sense the seat's occupancy status and provide key information for subsequent judgments. When the seat position sensor detects that the main driver's seat is occupied, the system will immediately start a further recognition process. It will combine the image data captured by the in-car camera and use advanced image recognition technology to confirm the identity of the person occupying the seat and determine whether it is the trainee himself. Not only that, the system will also carefully analyze the relative relationship between the trainee's body position and the main driver's seat. By comparing the preset conditions, such as the trainee's sitting posture, the distance between the body and the steering wheel, the position of the feet on the pedals, etc., the system can accurately determine whether the trainee is really in the main driver's position and whether it meets the standard requirements for driving operations.

[0041] In step S3, the step of determining whether the trainee is in the vehicle includes: Install GPS positioning module, infrared sensor and high-definition camera on the training vehicle; The vehicle location information is obtained in real time through the GPS positioning module, and the human body heat is detected through the infrared sensor to preliminarily determine whether the trainee is in the vehicle; When it is initially determined that the trainee is in the car, the high-definition camera is started to capture the image inside the car, and the deep learning algorithm is used to perform image recognition on the image to detect whether there is a trainee in the car, as well as the trainee's facial features or body features, to further confirm whether the trainee is in the car. First, the GPS positioning module, infrared sensor and high-definition camera are carefully installed on the training car, which together constitute the detection system. The GPS positioning module can provide real-time vehicle location information, providing basic data support for determining whether the trainee is in the car. Then, the infrared sensor plays its unique advantages. It can preliminarily determine whether there is someone in the car by detecting the heat emitted by the human body. This non-contact detection method is not only highly sensitive, but also can be performed without affecting the trainee's driving. When it is initially determined that the trainee may be in the car, the system will immediately start the high-definition camera to capture the real-time image inside the car. These images are then sent to the deep learning algorithm for image recognition. The algorithm will carefully analyze every detail in the picture to detect whether the trainee appears, and will also compare the trainee's facial features or body features to ensure the accuracy of the judgment.

[0042] As a second embodiment of the present invention, a driving training student learning hours detection management system based on intelligent technology is used to implement a driving training student learning hours detection management method, including: Intelligent learning time detection equipment: installed on the training vehicle, used to collect the trainee's location, movement status and facial recognition data in real time; Data center: used to receive, store and process data transmitted by intelligent learning time detection equipment; Study time detection module: using big data analysis technology to automatically detect and record students' study time, and identify abnormal actions or violations; Personalized training suggestion module: Through artificial intelligence technology, intelligent evaluation of students' driving operations is carried out to provide students with personalized training suggestions; Feedback and interaction module: used to provide real-time feedback of learning hours test results and personalized training suggestions to driving schools and students, supporting two-way interaction and communication; Abnormal reminder and data feedback module: In abnormal situations, abnormal reminders are sent to relevant personnel and relevant data are returned for further analysis.

[0043] In this embodiment, the driving training student learning hours detection and management system based on intelligent technology integrates intelligent learning hours detection equipment, data center, learning hours detection module, personalized training suggestion module, feedback and interaction module, and abnormal reminder and data return module. The system collects the student's position, movement and facial data in real time through the intelligent device on the training car. The data center receives and processes these data. The learning hours detection module uses big data analysis technology to automatically detect learning hours and identify abnormal behaviors. At the same time, the personalized training suggestion module uses artificial intelligence technology to provide customized training suggestions for students. The feedback and interaction module ensures real-time feedback of learning hours results and training suggestions, and supports two-way communication. In abnormal situations, the abnormal reminder and data return module will immediately send reminders and return data for further analysis, fully realizing efficient and accurate learning hours detection and management.

[0044] The specific description of the present invention in the above embodiments is only used to further illustrate the present invention and cannot be understood as limiting the scope of protection of the present invention. Technical engineers in this field may make some non-essential improvements and adjustments to the present invention based on the contents of the above invention, which fall within the scope of protection of the present invention.

Claims

1. A method for managing driving training students' learning hours based on intelligent technology, characterized in that: The following steps are involved: S1, install intelligent learning time detection equipment on the training car to collect the trainee's location, movement status and facial recognition image data in real time; S2, uses edge computing methods to locally process the collected image data, extract key features and encrypt and transmit them to the data center; S3, analyzes the data transmitted to the data center to determine whether the student is in the car, whether the person learning to drive in the car is the student himself, whether the student has been immobile for a long time, and whether the student is in the main driving position, and automatically calculates and records the learning hours based on the judgment results; S4 monitors the student’s driving operation data in real time, compares it with the preset standard actions, identifies abnormal actions, issues abnormal reminders, and sends back relevant data for analysis.

2. The method for managing driving training student learning hours based on intelligent technology according to claim 1 is characterized in that: The automatic calculation and recording of study hours according to the judgment results includes: When the student passes identity verification and is confirmed to be driving normally in the main driving position in the car, the time starts automatically and is recorded as valid learning hours; If the student leaves the main driving position, identity authentication fails, or abnormal behavior is detected, the learning hours will be suspended and marked as an abnormal period.

3. The method for managing driving training students' learning hours based on intelligent technology according to claim 1 is characterized in that: In step S3, the step of determining whether the trainee has been immobile for a long time includes: Obtain the student's historical behavior data, pre-process it to distinguish valid behavior data from invalid behavior data, calculate the dynamic behavior anomaly coefficient threshold and the static behavior anomaly coefficient threshold based on the valid behavior data, and judge whether the student's behavior is abnormal based on the comparison result between the student's current behavior data and the threshold. Prompt when the number of abnormal behaviors reaches the preset threshold.

4. The method for managing driving training student learning hours based on intelligent technology according to claim 3 is characterized in that: When the student's historical behavior data is invalid behavior data, the user habit data is output based on the invalid behavior data. When the student's current behavior data does not match the user habit data, the student is judged to be abnormal and a prompt is given.

5. The method for managing driving training student learning hours based on intelligent technology according to claim 3 or 4, characterized in that: The calculation of the dynamic behavior abnormal coefficient threshold and the static behavior abnormal coefficient threshold includes: dividing the historical dynamic behavior data into historical dynamic abnormal data and historical dynamic normal data, calculating the first abnormal coefficient of the historical dynamic abnormal data compared with the historical dynamic normal data, and using the minimum abnormal coefficient as the dynamic behavior abnormal coefficient threshold; dividing the historical static behavior data into historical static abnormal data and historical static normal data, calculating the first abnormal coefficient of the historical static abnormal data compared with the historical static normal data, and using the minimum abnormal coefficient as the static behavior abnormal coefficient threshold.

6. The method for managing driving training students' learning hours based on intelligent technology according to claim 5 is characterized in that: The calculation of the first abnormal coefficient includes: Based on the matching degree of the student's trajectory and motion sequence, the matching value of the historical dynamic abnormal data and the historical dynamic normal data is output, and the abnormal coefficient is calculated.

7. The method for managing driving training students' learning hours based on intelligent technology according to claim 1 is characterized in that: In step S3, the step of determining whether the person learning to drive on the bus is the trainee himself includes: Pre-enter the students’ facial feature data at the driving school to establish a student facial database; The facial features in the image data are extracted and compared with the student facial database to determine whether the person learning to drive in the car is the student himself.

8. The method for managing driving training students' learning hours based on intelligent technology according to claim 1 is characterized in that: In step S3, the step of determining whether the trainee is in the main driving position includes: Install a seat position sensor under the main driver's seat of the training vehicle; When the seat position sensor detects that the main driver's seat is occupied, it determines whether the person occupying the seat is a trainee based on the recognition results of the image data, and determines whether the relative position relationship between the trainee's body position and the main driver's seat meets the preset conditions to confirm whether the trainee is in the main driver's position.

9. The method for managing driving training students' learning hours based on intelligent technology according to claim 1 is characterized in that: In step S3, the step of determining whether the trainee is in the vehicle includes: Install GPS positioning module, infrared sensor and high-definition camera on the training vehicle; The vehicle location information is obtained in real time through the GPS positioning module, and the human body heat is detected through the infrared sensor to preliminarily determine whether the trainee is in the vehicle; When it is preliminarily determined that the student is in the car, the high-definition camera is activated to capture the image inside the car, and the deep learning algorithm is used to perform image recognition on the image to detect whether there is a student in the car, as well as the student's facial features or body features, to further confirm whether the student is in the car.

10. A driving training student learning hours detection and management system based on intelligent technology, used to implement the driving training student learning hours detection and management method as claimed in any one of claims 1 to 9, characterized in that: include: Intelligent learning time detection equipment: installed on the training vehicle, used to collect the trainee's location, movement status and facial recognition data in real time; Data center: used to receive, store and process data transmitted by intelligent learning time detection equipment; Study time detection module: using big data analysis technology to automatically detect and record students' study time, and identify abnormal actions or violations; Personalized training suggestion module: Through artificial intelligence technology, intelligent evaluation of students' driving operations is carried out to provide students with personalized training suggestions; Feedback and interaction module: used to provide real-time feedback of learning hours test results and personalized training suggestions to driving schools and students, supporting two-way interaction and communication; Abnormal reminder and data feedback module: In abnormal situations, abnormal reminders are sent to relevant personnel and relevant data are returned for further analysis.

Images