CN111209796B

CN111209796B - Driving behavior detection method and device, electronic equipment and medium

Info

Publication number: CN111209796B
Application number: CN201911319425.8A
Authority: CN
Inventors: 许世勋
Original assignee: Ping An Property and Casualty Insurance Company of China Ltd
Current assignee: Ping An Property and Casualty Insurance Company of China Ltd
Priority date: 2019-12-19
Filing date: 2019-12-19
Publication date: 2023-08-29
Anticipated expiration: 2039-12-19
Also published as: CN111209796A

Abstract

The invention provides a driving behavior detection method, a driving behavior detection device, electronic equipment and a medium. According to the method, gesture data and pressure data on a driver terminal device to be detected in a first preset time can be obtained and preprocessed, first data after gesture data preprocessing and second data after pressure data preprocessing are obtained, an arcsine function is adopted to calculate a pitch angle of each second of the terminal device, behavior characteristics of each second of the driver to be detected are determined according to the pitch angle of each second and the second data, when the behavior characteristics of the current second are detected to be a second result and the behavior characteristics of the last second of the current second are detected to be a first result, target data of the current second are extracted and input into a pre-built target model, a recognition result is obtained, and when the recognition result is determined to comprise walking characteristics or static characteristics and when the walking characteristics or static characteristics last for the second preset time from the current second, the driver to be detected is determined to finish using the terminal device from the current second and enter a walking or static state through intelligent decision.

Description

Driving behavior detection method and device, electronic equipment and medium

Technical Field

The present invention relates to the field of intelligent decision making technologies, and in particular, to a driving behavior detection method, apparatus, electronic device, and medium.

Background

With the popularization of terminal devices such as mobile phones, low-head use of terminal devices such as mobile phones has become a common phenomenon, and the probability of occurrence of accidents with terminal devices such as mobile phones is about 23 times higher than that of normal driving during driving, so that the use of terminal devices such as mobile phones by drivers is very dangerous during driving of vehicles.

In order to reduce the occurrence probability of traffic accidents, a detection method of the behavior of the driver playing the mobile phone based on deep learning is also generated, however, in the prior art, an additional data acquisition device is required to be installed or detection data is acquired by means of a vehicle data recorder of other vehicles, and secondly, the duration of the driver continuously playing the mobile phone when driving cannot be determined.

Disclosure of Invention

In view of the foregoing, it is necessary to provide a driving behavior detection method, apparatus, electronic device, and medium, which are capable of directly solving the problems of difficult data acquisition and data quality without using other data acquisition devices, and determining the time when the driver to be tested finishes using the terminal device when driving.

A driving behavior detection method, the method comprising:

when a driving behavior detection instruction is received, acquiring gesture data of an acceleration sensor and pressure data of a touch screen sensor on a driver terminal device to be detected within a first preset time;

preprocessing the gesture data and the pressure data to obtain first data preprocessed by the gesture data and second data preprocessed by the pressure data;

calculating the first data by adopting an arcsine function to obtain a pitch angle of the terminal equipment per second;

determining the behavior characteristics of the driver to be tested every second according to the pitch angle every second and the corresponding second data;

when detecting that the behavior characteristic of the current second is a second result and the behavior characteristic of the second previous to the current second is a first result, extracting target data of the current second from the first data;

inputting the target data into a pre-constructed target model to obtain a recognition result;

determining whether the identification result comprises walking characteristics or static characteristics;

when the walking characteristic or the static characteristic is included in the identification result, determining whether the walking characteristic or the static characteristic lasts for a second preset time from the current second;

And when the walking characteristic or the static characteristic is determined to last for the second preset time from the current second, determining that the driver to be tested finishes using the terminal device from the current second and enters a walking or static state.

According to a preferred embodiment of the present invention, the preprocessing the gesture data and the pressure data to obtain first data preprocessed by the gesture data and second data preprocessed by the pressure data includes:

carrying out normal distribution processing on the gesture data and the pressure data to obtain a first normal distribution curve corresponding to the gesture data and a second normal distribution curve corresponding to the pressure data;

acquiring data which does not satisfy 99.7 rule from the first normal distribution curve as a first abnormal point of the attitude data, and acquiring data which does not satisfy 99.7 rule from the second normal distribution curve as a second abnormal point of the pressure data;

data cleaning is carried out on the first abnormal point and the second abnormal point, and third data corresponding to the gesture data and fourth data corresponding to the pressure data are obtained;

and filtering the third data and the fourth data by adopting a band-pass filtering method to obtain first data preprocessed by the attitude data and second data preprocessed by the pressure data.

According to a preferred embodiment of the present invention, determining the behavior feature of the driver under test per second according to the pitch angle per second and the second data includes:

when the pitch angle per second is larger than or equal to the configuration angle and the second data per second is larger than or equal to a first threshold value, determining that the behavior characteristic of the driver to be tested corresponding to seconds is the first result;

and when the pitch angle per second is smaller than the configuration angle or the second data per second is smaller than the first threshold value, determining the behavior characteristic of the driver to be tested corresponding to seconds as the second result.

According to a preferred embodiment of the present invention, after determining that the behavior feature of the driver under test is the first result, the method further includes:

acquiring a first time corresponding to the first result;

judging whether the driver to be tested is in a driving state or not at the first time;

when the driver to be tested is determined to be in a driving state at the first time, acquiring road information of a driving road where the driver to be tested is located;

generating prompt information according to the road information;

sending the prompt information to the terminal equipment;

when the prompt information is detected to be unprocessed, acquiring the face information of the driver to be detected;

And storing the face information into a configuration library.

According to a preferred embodiment of the present invention, before inputting the target data into a pre-constructed target model to obtain a recognition result, the method further includes:

acquiring first historical data on all terminal equipment of a driver to be tested;

dividing the first historical data to obtain a training data set and a verification data set;

training the training data set to obtain at least one primary learner;

adjusting the at least one primary learner according to the verification data set to obtain at least one secondary learner;

acquiring test data on the terminal equipment and the total quantity of the test data;

testing the at least one secondary learner by adopting the test data to obtain the target quantity of the target test data passing the test in each secondary learner;

dividing the target number by the total number to obtain at least one pass rate;

and determining the secondary learner with the highest passing rate as the target model.

According to a preferred embodiment of the invention, the method further comprises:

and when the recognition result is detected to not comprise the walking characteristic and the static characteristic, determining that the driver to be detected uses the terminal device at the end of the current second.

determining at least one third time when the driver to be tested starts to use the terminal device when driving in the first preset time;

acquiring at least one fourth time when the driver to be tested finishes using the terminal equipment when driving;

subtracting the at least one fourth time from the at least one third time which is adjacent to the at least one fourth time to obtain at least one target duration of the driver to be tested using the terminal device when driving;

determining the target time of the driver to be tested for using the terminal device when driving according to the at least one target time length;

calculating the number of the at least one target duration to obtain the target times of using the terminal device by the driver to be tested when driving in the first preset time;

dividing the target times by the first preset time to obtain the frequency of the driver to be tested using the terminal equipment;

and generating a behavior report of the driver to be tested according to the frequency and the target time.

A driving behavior detection apparatus, the apparatus comprising:

The acquisition unit is used for acquiring the gesture data of the acceleration sensor and the pressure data of the touch screen sensor on the driver terminal equipment to be tested within a first preset time when a driving behavior detection instruction is received;

the preprocessing unit is used for preprocessing the gesture data and the pressure data to obtain first data preprocessed by the gesture data and second data preprocessed by the pressure data;

the calculating unit is used for calculating the first data by adopting an arcsine function to obtain a pitch angle of the terminal equipment per second;

the determining unit is used for determining the behavior characteristics of the driver to be tested every second according to the pitch angle every second and the corresponding second data;

an extracting unit, configured to extract target data of a current second from the first data when it is detected that a behavior feature of the current second is a second result and a behavior feature of a second previous to the current second is a first result;

the input unit is used for inputting the target data into a pre-constructed target model to obtain a recognition result;

the determining unit is further used for determining whether the identification result comprises walking characteristics or static characteristics;

The determining unit is further configured to determine, when the walking feature or the rest feature is included in the identification result, whether the walking feature or the rest feature lasts for a second preset time from the current second;

the determining unit is further configured to determine that the driver to be tested uses the terminal device from the current second to enter a walking or resting state when it is determined that the walking feature or the resting feature continues for the second preset time from the current second.

According to a preferred embodiment of the invention, the preprocessing unit is specifically configured to:

According to a preferred embodiment of the present invention, the determining unit determines the behavior feature of the driver to be tested per second according to the pitch angle per second and the second data, including:

According to a preferred embodiment of the present invention, the obtaining unit is further configured to obtain a first time corresponding to the first result after determining that the behavior feature of the driver to be tested is the first result;

the apparatus further comprises:

the judging unit is used for judging whether the driver to be tested is in a driving state or not at the first time;

the obtaining unit is further configured to obtain road information of a driving road where the driver to be tested is located when it is determined that the driver to be tested is in a driving state at the first time;

The generating unit is used for generating prompt information according to the road information;

the sending unit is used for sending the prompt information to the terminal equipment;

the acquisition unit is further used for acquiring face information of the driver to be detected when the prompt information is detected to be unprocessed;

and the storage unit is used for storing the face information into a configuration library.

According to a preferred embodiment of the present invention, the obtaining unit is further configured to obtain first historical data on all terminal devices of the driver to be tested before inputting the target data into a pre-constructed target model to obtain a recognition result;

the apparatus further comprises:

the dividing unit is used for dividing the first historical data to obtain a training data set and a verification data set;

the training unit is used for training the training data set to obtain at least one primary learner;

the adjustment unit is used for adjusting the at least one primary learner according to the verification data set to obtain at least one secondary learner;

the acquisition unit is further used for acquiring the test data on the terminal equipment and the total quantity of the test data;

the test unit is used for testing the at least one secondary learner by adopting the test data to obtain the target quantity of the target test data passing the test in each secondary learner;

The calculating unit is further used for dividing the target quantity by the total quantity to obtain at least one passing rate;

the determining unit is further configured to determine, as the target model, a secondary learner with a highest passing rate.

According to a preferred embodiment of the present invention, the determining unit is further configured to determine that the driver under test is using the terminal device at the end of the current second when it is detected that the recognition result does not include the walking feature and the stationary feature.

According to a preferred embodiment of the present invention, the determining unit is further configured to determine, in the first preset time, at least one third time when the driver to be tested starts to use the terminal device while driving;

the obtaining unit is further configured to obtain at least one fourth time when the driver to be tested finishes using the terminal device during driving;

the computing unit is further configured to perform a subtraction operation on the at least one fourth time and the at least one third time that is adjacent to the fourth time, so as to obtain at least one target duration of the driver to be tested using the terminal device when driving;

the determining unit is further configured to determine, according to the at least one target duration, a target time when the driver to be tested uses the terminal device during driving;

The calculating unit is further configured to calculate the number of the at least one target duration, so as to obtain a target number of times that the driver to be tested uses the terminal device when driving in the first preset time;

the calculating unit is further configured to divide the target frequency by the first preset time to obtain a frequency of the driver to be tested using the terminal device;

the generating unit is further configured to generate a behavior report of the driver to be tested according to the frequency and the target time.

An electronic device, the electronic device comprising:

a memory storing at least one instruction; a kind of electronic device with high-pressure air-conditioning system

And the processor executes the instructions stored in the memory to realize the driving behavior detection method.

A computer-readable storage medium having stored therein at least one instruction that is executed by a processor in an electronic device to implement the driving behavior detection method.

According to the technical scheme, the data processing is directly performed by utilizing the data on the terminal equipment without other data acquisition equipment, so that the problem that the data is difficult to acquire caused by the fact that additional data acquisition equipment is required to be installed in the prior art is directly solved, the problem of data quality caused by detecting by means of driving records of other vehicles is solved, the time for the driver to be detected to finish using the terminal equipment when driving can be determined, and data support is provided for determining the duration of continuous use of the terminal equipment when the driver to be detected is driven.

Drawings

FIG. 1 is a flow chart of a driving behavior detection method according to a preferred embodiment of the present invention.

Fig. 2 is a functional block diagram of a preferred embodiment of the driving behavior detection apparatus of the present invention.

Fig. 3 is a schematic structural diagram of an electronic device according to a preferred embodiment of the present invention for implementing a driving behavior detection method.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in detail with reference to the accompanying drawings and specific embodiments.

Fig. 1 is a flowchart of a driving behavior detection method according to a preferred embodiment of the present invention. The order of the steps in the flowchart may be changed and some steps may be omitted according to various needs.

The driving behavior detection method is applied to one or more electronic devices, wherein the electronic devices are devices capable of automatically performing numerical calculation and/or information processing according to preset or stored instructions, and the hardware of the electronic devices comprises, but is not limited to, microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASICs), programmable gate arrays (Field-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSPs), embedded devices and the like.

The electronic device may be any electronic product that can interact with a user in a human-computer manner, such as a personal computer, tablet computer, smart phone, personal digital assistant (Personal Digital Assistant, PDA), game console, interactive internet protocol television (Internet Protocol Television, IPTV), smart wearable device, etc.

The electronic device may also include a network device and/or a user device. Wherein the network device includes, but is not limited to, a single network server, a server group composed of a plurality of network servers, or a Cloud based Cloud Computing (Cloud Computing) composed of a large number of hosts or network servers.

The network in which the electronic device is located includes, but is not limited to, the internet, a wide area network, a metropolitan area network, a local area network, a virtual private network (Virtual Private Network, VPN), and the like.

And S10, acquiring the gesture data of an acceleration sensor and the pressure data of a touch screen sensor on the terminal equipment of the driver to be tested within a first preset time when a driving behavior detection instruction is received.

In at least one embodiment of the present invention, the driving behavior detection command may be triggered by a user, or may be triggered automatically when a certain condition is met, which is not limited by the present invention.

Wherein, the meeting of certain conditions includes, but is not limited to: the configuration time is met, the electronic equipment detects that the driver drives, and the like.

The configuration time may include a determined point in time (e.g., the configuration time may be seven points in the morning each day), or a period of time, etc.

In at least one embodiment of the present invention, the gesture data is obtained from an acceleration sensor on the driver terminal device under test, the pressure data is obtained from a touch screen sensor on the driver terminal device under test, and the gesture data and the pressure data are both data generated within the first preset time. The gesture data and the pressure data can be directly obtained from the terminal equipment, so that the detection data is easy to obtain.

The first preset time may be a time period, which is not limited by the present invention.

S11, preprocessing the gesture data and the pressure data to obtain first data preprocessed by the gesture data and second data preprocessed by the pressure data.

In at least one embodiment of the present invention, the electronic device preprocessing the gesture data and the pressure data to obtain first data preprocessed by the gesture data and second data preprocessed by the pressure data includes:

The electronic equipment performs normal distribution processing on the gesture data and the pressure data to obtain a first normal distribution curve corresponding to the gesture data and a second normal distribution curve corresponding to the pressure data, obtains data which does not meet 99.7 rule from the first normal distribution curve, serves as a first abnormal point of the gesture data, obtains data which does not meet 99.7 rule from the second normal distribution curve, serves as a second abnormal point of the pressure data, performs data cleaning on the first abnormal point and the second abnormal point to obtain third data corresponding to the gesture data and fourth data corresponding to the pressure data, and further performs filtering processing on the third data and the fourth data by adopting a band-pass filtering method to obtain the first data after the gesture data preprocessing and the second data after the pressure data preprocessing.

Wherein, the data satisfying 99.7 rule refers to data within three positive and negative standard deviation ranges of the average value of the data on the normal distribution curve.

Firstly, since the output data of the acceleration sensor and the touch screen sensor generally include abnormal points, the abnormal points existing in the attitude data and the pressure data can be removed by performing data processing based on the 99.7 rule, and secondly, since a certain vibration exists in the vehicle during the running process (the driver to be tested is on a vehicle which has started but is not running on an engine, and the acceleration sensor still generates the attitude data although the driver to be tested is actually in a stationary state due to the vibration of the engine), the data related to the vibration can be reduced or eliminated by the filtering processing.

And S12, calculating the first data by adopting an arcsine function to obtain the pitch angle of the terminal equipment per second.

In at least one embodiment of the present invention, the pitch angle refers to an angle formed by the terminal device and an X-axis direction of a rectangular space coordinate system.

In at least one embodiment of the present invention, the electronic device calculating the first data using an arcsine function, and obtaining the pitch angle of the terminal device per second includes:

the electronic equipment extracts a first acceleration in the X-axis direction and a second acceleration in the Z-axis direction from the first data, divides the first acceleration in each second by the second acceleration in the corresponding second to obtain a target ratio in each second, and performs arcsine operation on the opposite number of the target ratio to obtain a pitch angle in each second.

For example: the electronic equipment extracts a first acceleration of 9 o 'clock from the first data to be-1, extracts a second acceleration of 9 o' clock to be 2, divides the first acceleration by the second acceleration to obtain a target ratio of 9 o 'clock to be-0.5, and performs arcsine operation on the opposite number 0.5 of the target ratio to obtain a pitch angle of 9 o' clock to be 30 degrees.

Through the implementation mode, the pitch angle of the terminal equipment per second can be accurately determined, and a data basis is provided for subsequent determination of the behavior characteristics of the driver to be tested.

In other embodiments, the electronic device may further calculate an inclination angle of the terminal device according to the first data, and process the inclination angle by integrating all angles, so that the determined behavior characteristics of the driver to be tested are more accurate due to more comprehensive data.

S13, determining the behavior characteristics of the driver to be tested every second according to the pitch angle every second and the corresponding second data.

In at least one embodiment of the present invention, the behavior feature refers to a behavior of the driver to be tested on the terminal device, where the behavior feature includes a first result and a second result.

Specifically, the first result means that the driver under test is using the terminal device, and further, the second result means that the driver under test is not currently using the terminal device.

In at least one embodiment of the present invention, the determining, by the electronic device, the behavior feature of the driver under test per second according to the pitch angle per second and the second data includes:

and when the pitch angle per second is smaller than the configuration angle or the second data per second is smaller than the first threshold, the electronic equipment determines that the behavior characteristic of the driver to be tested corresponding to seconds is the second result.

Wherein the configuration angle is determined according to a pitch angle of at least one driver when using the terminal device, and the present invention is not limited.

Further, the first threshold is determined according to the pressure when at least one driver touches the terminal device, which is not limited by the present invention.

And when the pitch angle is larger than or equal to the configuration angle and the second data (preprocessed pressure data) is larger than or equal to a first threshold value, determining the behavior characteristics of the driver to be detected as the first result, and otherwise, determining the behavior characteristics as the second result.

Through the implementation manner, the behavior characteristics of the driver to be tested can be rapidly determined according to the pitch angle and the second data.

In at least one embodiment of the present invention, after determining that the behavior feature of the driver under test is the first result, the method further includes:

the electronic equipment acquires first time corresponding to the first result, judges whether the driver to be tested is in a driving state or not when the first time is, acquires road information of a driving road where the driver to be tested is when the driver to be tested is determined to be in the driving state when the first time is, generates prompt information according to the road information, further, the electronic equipment sends the prompt information to the terminal equipment, acquires face information of the driver to be tested when the prompt information is detected to be unprocessed, and further, the electronic equipment stores the face information into a configuration library.

Wherein, the prompt information includes, but is not limited to: road information, hazard information, driving time, etc.

Further, the configuration library stores face information of a driver playing the mobile phone when driving.

Through the above embodiment, not only can prompt information be timely sent when the driver to be tested drives and uses the terminal equipment is detected, the occurrence of traffic accidents is avoided, the effect of reminding the driver to be tested is achieved, but also the face information of the driver to be tested can be recorded when the prompt information is not processed, and the driver to be tested is convenient to be subjected to subsequent punishment and abstinence.

And S14, when the behavior characteristic of the current second is detected to be a second result and the behavior characteristic of the second previous to the current second is detected to be a first result, extracting target data of the current second from the first data.

In at least one embodiment of the present invention, the target data refers to data detected by the acceleration sensor at the current second.

In at least one embodiment of the present invention, the electronic device extracting the target data of the current second from the first data includes:

and the electronic equipment judges whether the second time corresponding to each first data is the current second or not, and further, the electronic equipment determines that the second time is the data of the current second as the target data.

Because the electronic equipment detects that the behavior characteristic of the current second is inconsistent with the behavior characteristic of the second previous to the current second, the target data of the current second is extracted, and basic data can be provided for judging whether interference data exist in the target data.

S15, inputting the target data into a pre-constructed target model to obtain a recognition result.

In at least one embodiment of the invention, the target model is a model constructed using first historical data on the terminal device, the target model also having adaptive capabilities.

The recognition result refers to the state of the driver under test in the current second, and the recognition result may include, but is not limited to: driving state, walking state, stationary state, etc.

In at least one embodiment of the present invention, before inputting the target data into a pre-constructed target model to obtain a recognition result, the method further includes:

the electronic equipment acquires first historical data on all driver terminal equipment to be tested, divides the first historical data to obtain a training data set and a verification data set, further trains the training data set by the electronic equipment to obtain at least one primary learner, adjusts the at least one primary learner according to the verification data set to obtain at least one secondary learner, acquires test data on the terminal equipment and the total number of the test data, tests the at least one secondary learner by adopting the test data to obtain the target number of the target test data passing through the test in each secondary learner, divides the target number by the total number to obtain at least one passing rate, and determines the secondary learner with the highest passing rate as the target model.

Because the terminal equipment records the gesture data and the pressure data of the driver to be tested when using the terminal equipment, the electronic equipment acquires the gesture data and the pressure data as the test data on the terminal equipment.

Through the implementation mode, the accurate target model can be obtained through training, so that intelligent decision can be made on the state of the driver to be tested based on the target model.

Specifically, the dividing the first historical data by the electronic device to obtain a training data set and a verification data set includes:

the electronic device randomly divides the first historical data into at least one data packet according to a preset proportion, any one data packet in the at least one data packet is determined to be the verification data set, the rest data packets are determined to be the training data set, and the steps are repeated until all the data packets are sequentially used as the verification data set.

The preset proportion can be set in a self-defined mode, and the invention is not limited.

For example: the electronic device divides the first historical data into 3 data packets, namely a data packet E, a data packet F and a data packet G, and determines the data packet E as the verification data set and the data packet F and the data packet G as the training data set. Next, the data packet F is determined as the verification data set, and the data packet E and the data packet G are determined as the training data set. Finally, the data packet G is determined as the verification data set, and the data packet E and the data packet F are determined as the training data set.

Through the embodiment, the first historical data are divided, and each data in the first historical data participates in training and verification, so that the fitting degree of training the target model is improved.

Further, the electronic device adjusting the at least one primary learner according to the verification data set to obtain at least one secondary learner includes:

the electronic equipment adopts a super-parameter grid searching method to determine optimal super-parameter points from the verification data set, and further, the electronic equipment adjusts the at least one primary learner through the optimal super-parameter points to obtain the at least one secondary learner.

Specifically, the electronic device splits the verification data set according to a fixed step length to obtain a target subset, traverses parameters of two end points on the target subset, verifies the at least one primary learner through the parameters of the two end points to obtain a learning rate of each parameter, determines the parameter with the best learning rate as a first super-parameter point, reduces the step length in a neighborhood of the first super-parameter point, continues traversing until the step length is a preset step length, and the obtained super-parameter point is the optimal super-parameter point, and further, the electronic device adjusts the at least one primary learner according to the optimal super-parameter point to obtain the at least one secondary learner.

The preset step length is not limited.

In at least one embodiment of the present invention, after obtaining the target model, the method further comprises:

and acquiring second historical data of the acceleration sensor on the driver terminal equipment to be detected by the electronic equipment every second preset time, determining the second historical data as a target verification data set, and further, adaptively adjusting the target model by the electronic equipment according to the target verification data set to obtain a target model with adaptive capacity.

The second history data may be included in the first history data, or may be data sent by an acceleration sensor on any driver terminal device to be tested.

The second preset time may be set in a self-defined manner, which is not limited by the present invention.

Through the implementation mode, the target model can carry out corresponding negative feedback adjustment according to the behavior habit of each driver to be tested, and the obtained recognition result is more accurate.

S16, determining whether the identification result comprises walking characteristics or static characteristics.

In at least one embodiment of the invention, the method further comprises:

And when the recognition result is detected to not comprise the walking characteristic and the static characteristic, the electronic equipment determines that the driver to be detected uses the terminal equipment at the end of the current second.

Specifically, since the driver to be tested is always in a driving state, and the recognition result does not include the walking feature and the stationary feature in the current second, the interference data does not exist in the target data, and it is further determined that the driver to be tested is finished using the terminal device in the current second.

The interference data refers to data which interfere with the electronic equipment to judge the behavior characteristics of the driver to be tested.

And determining that the driver to be tested finishes using the terminal equipment in the current second, so as to obtain the time for the driver to be tested to finish using the terminal equipment when driving.

And S17, when the walking characteristic or the static characteristic is included in the identification result, determining whether the walking characteristic or the static characteristic lasts for a second preset time from the current second.

In at least one embodiment of the present invention, the second preset time refers to a period of time from the current second.

In at least one embodiment of the invention, the electronic device determining whether the walk feature or the rest feature last a second preset time from the current second comprises:

and the electronic equipment extracts fifth data from the first data lasting the second preset time from the current second, and further inputs the fifth data into the target model to obtain a third result per second.

And S18, when the walking characteristic or the static characteristic is determined to last for the second preset time from the current second, determining that the driver to be tested finishes using the terminal device from the current second and enters a walking or static state.

In at least one embodiment of the present invention, the electronic device determines that the walking feature or the parking feature last for the second preset time from the current second when the at least one third result is detected as the walking feature or the stationary feature.

In at least one embodiment of the invention, the method further comprises:

in the first preset time, the electronic device determines at least one third time when the driver to be tested starts to use the terminal device when driving, further, the electronic device obtains at least one fourth time when the driver to be tested finishes using the terminal device when driving, performs subtraction operation on the at least one fourth time and the at least one third time adjacent to the fourth time, obtains at least one target time when the driver to be tested uses the terminal device when driving, determines the target time when the driver to be tested uses the terminal device when driving according to the at least one target time, further, the electronic device calculates the number of the at least one target time, obtains the frequency of using the terminal device when the driver to be tested when driving in the first preset time, divides the target frequency by the first preset time, and generates a report according to the frequency of using the terminal device by the driver to be tested and the target time.

Through the implementation mode, the target duration of the terminal equipment used by the driver to be tested in driving can be determined, the behavior report can be generated, and the user can make an intelligent decision based on the behavior report conveniently.

The target time is a sum of times that the driver to be tested uses the terminal device when driving in the first preset time, for example: the first preset time is 9:00 to 11:00, and in the range of 9:00 to 11:00, the target times of using the terminal equipment by the driver to be tested when driving are 2 times, the target time length of using the terminal equipment for the first time is 2 minutes, and the target time length of using the terminal equipment for the second time is 5 minutes, so that the target time is 7 minutes.

Specifically, the behavior report may be used as basic information of a user decision, and the behavior report may include UBI (Usage Based Insurance, utility risk) car risk report and the like, so that by generating the behavior report, a user can be assisted in making a suitable decision, which is beneficial to the experience of the user.

For example: when the behavior report is a UBI car insurance report, the information in the UBI car insurance report may include, but is not limited to: the risk factor, the target time and the face information of the driver to be tested are beneficial to determining the amount of UBI car insurance of the driver to be tested by generating the UBI car insurance report of the driver to be tested.

Specifically, the determining, by the electronic device, at least one third time when the driver to be tested starts to use the terminal device while driving includes:

and when the behavior characteristic of the driver to be detected in the target second is detected to be the first result and the behavior characteristic of the driver to be detected in the last second in the target second is detected to be the second result, the electronic equipment determines the target second as the third time.

According to the technical scheme, when a driving behavior detection instruction is received, gesture data of an acceleration sensor and pressure data of a touch screen sensor on a driver terminal device to be detected in a first preset time are obtained, the gesture data and the pressure data are preprocessed to obtain first data preprocessed by the gesture data and second data preprocessed by the pressure data, an arcsine function is adopted to calculate the first data to obtain a pitch angle of the terminal device per second, the behavior characteristics of the driver to be detected are determined according to the pitch angle per second and the corresponding second data, when the second result is detected as the second result and the behavior characteristics of the last second are detected as the first result, target data of the current second are extracted from the first data, the target data are input into a pre-built target model to obtain a recognition result, whether walking characteristics or static characteristics are included in the recognition result is determined, when the recognition result includes the walking characteristics or the static characteristics, whether the pre-determined characteristics are required to be acquired from the second data or the second data are not required to be acquired by the terminal device, the second data are not required to be acquired from the terminal device to be detected, the second data are not required to be acquired, the current characteristics are not required to be acquired, and the current characteristics are not required to be acquired by the terminal device is required to be directly acquired from the terminal device to be directly acquired by the second data, and the data quality problem caused by detection by means of the driving records of other vehicles can also be determined, and the time for the driver to be tested to finish using the terminal equipment when driving can be also determined, so that data support is provided for the problems of determining the duration of continuous use of the terminal equipment by the driver to be tested when driving, and the like.

Fig. 2 is a functional block diagram of a driving behavior detection device according to a preferred embodiment of the present invention. The driving behavior detection device 11 includes an acquisition unit 110, a preprocessing unit 111, a calculation unit 112, a determination unit 113, an extraction unit 114, an input unit 115, a judgment unit 116, a generation unit 117, a transmission unit 118, a storage unit 119, a division unit 120, a training unit 121, an adjustment unit 122, and a test unit 123. The module/unit referred to in the present invention refers to a series of computer program segments capable of being executed by the processor 13 and of performing a fixed function, which are stored in the memory 12. In the present embodiment, the functions of the respective modules/units will be described in detail in the following embodiments.

When receiving the driving behavior detection instruction, the obtaining unit 110 obtains gesture data of an acceleration sensor and pressure data of a touch screen sensor on the driver terminal device to be detected within a first preset time.

Wherein, the meeting of certain conditions includes, but is not limited to: meeting the configuration time, detecting the driving of the driver, etc.

The preprocessing unit 111 performs preprocessing on the gesture data and the pressure data to obtain first data preprocessed by the gesture data and second data preprocessed by the pressure data.

In at least one embodiment of the present invention, the preprocessing unit 111 performs preprocessing on the gesture data and the pressure data to obtain first data preprocessed by the gesture data and second data preprocessed by the pressure data, where the preprocessing includes:

The preprocessing unit 111 performs normal distribution processing on the gesture data and the pressure data to obtain a first normal distribution curve corresponding to the gesture data and a second normal distribution curve corresponding to the pressure data, obtains data that does not satisfy 99.7 rule from the first normal distribution curve, uses the data as a first abnormal point of the gesture data, obtains data that does not satisfy 99.7 rule from the second normal distribution curve, uses the data as a second abnormal point of the pressure data, performs data cleaning on the first abnormal point and the second abnormal point to obtain third data corresponding to the gesture data and fourth data corresponding to the pressure data, and further, the preprocessing unit 111 performs filtering processing on the third data and the fourth data by using a band-pass filtering method to obtain first data after the gesture data preprocessing and second data after the pressure data preprocessing.

The calculation unit 112 calculates the first data by using an arcsine function, so as to obtain a pitch angle of the terminal equipment per second.

In at least one embodiment of the present invention, the calculating unit 112 calculates the first data using an arcsine function, and obtaining the pitch angle of the terminal device per second includes:

the calculating unit 112 extracts a first acceleration in the X-axis direction and a second acceleration in the Z-axis direction from the first data, divides the first acceleration per second by the second acceleration corresponding to the second to obtain a target ratio per second, and performs an arcsine operation on the opposite number of the target ratio to obtain a pitch angle per second.

For example: the calculating unit 112 extracts a first acceleration of-1 to 9 o 'clock from the first data, extracts a second acceleration of 2 to 9 o' clock, divides the first acceleration by the second acceleration to obtain a target ratio of-0.5 to 9 o 'clock, and performs an arcsine operation on the opposite number 0.5 of the target ratio to obtain a pitch angle of 30 degrees to 9 o' clock.

In other embodiments, the calculating unit 112 may further calculate the inclination angle of the terminal device according to the first data, and integrate all angles to process, so that the determined behavior characteristics of the driver to be tested are more accurate due to more comprehensive data.

The determining unit 113 determines the behavior characteristics of the driver to be tested per second according to the pitch angle per second and the corresponding second data.

In at least one embodiment of the present invention, the determining unit 113 determines the behavior feature of the driver to be tested per second according to the pitch angle per second and the second data, including:

The determining unit 113 determines that the behavior feature of the driver under test corresponding to seconds is the first result when the pitch angle per second is greater than or equal to the configuration angle and the second data per second is greater than or equal to the first threshold, and determines that the behavior feature of the driver under test corresponding to seconds is the second result when the pitch angle per second is less than the configuration angle or the second data per second is less than the first threshold.

In at least one embodiment of the present invention, after determining that the behavior feature of the driver to be tested is the first result, the obtaining unit 110 obtains a first time corresponding to the first result, the judging unit 116 judges whether the driver to be tested is in a driving state at the first time, when determining that the driver to be tested is in the driving state at the first time, the obtaining unit 110 obtains road information of a driving road on which the driver to be tested is located, the generating unit 117 generates prompt information according to the road information, further, the sending unit 118 sends the prompt information to the terminal device, when detecting that the prompt information is not processed, the obtaining unit 110 obtains face information of the driver to be tested, and further, the storing unit 119 stores the face information in a configuration library.

When detecting that the behavior feature of the current second is the second result and the behavior feature of the second preceding the current second is the first result, the extraction unit 114 extracts the target data of the current second from the first data.

In at least one embodiment of the present invention, the extracting unit 114 extracts the target data of the current second from the first data includes:

the extracting unit 114 determines whether the second time corresponding to each first data is the current second, and further, the extracting unit 114 determines that the second time is the data of the current second as the target data.

Since the behavior characteristics of the current second are detected to be inconsistent with the behavior characteristics of the second previous to the current second, the target data of the current second are extracted, and basic data can be provided for judging whether interference data exist in the target data.

The input unit 115 inputs the target data into a target model constructed in advance, and obtains a recognition result.

In at least one embodiment of the present invention, before the target data is input into a pre-built target model to obtain a recognition result, the obtaining unit 110 obtains first historical data on all terminal devices of drivers to be tested, the dividing unit 120 divides the first historical data to obtain a training data set and a verification data set, further, the training unit 121 trains the training data set to obtain at least one primary learner, the adjusting unit 122 adjusts the at least one primary learner according to the verification data set to obtain at least one secondary learner, the obtaining unit 110 obtains test data on the terminal devices and the total number of the test data, the test unit 123 tests the at least one secondary learner by using the test data to obtain a target number of the target test data passing through the test in each secondary learner, the calculating unit 112 divides the target number by the total number to obtain at least one passing rate, and the determining unit 113 determines the secondary learner with the highest passing rate as the target model.

Specifically, the dividing unit 120 divides the first historical data to obtain a training data set and a verification data set, which includes:

the dividing unit 120 randomly divides the first historical data into at least one data packet according to a preset proportion, determines any one data packet of the at least one data packet as the verification data set, determines the rest data packets as the training data set, and repeats the above steps until all data packets are sequentially used as the verification data set.

For example: the dividing unit 120 divides the first history data into 3 data packets, namely, a data packet E, a data packet F, and a data packet G, and determines the data packet E as the verification data set and the data packet F and the data packet G as the training data set. Next, the data packet F is determined as the verification data set, and the data packet E and the data packet G are determined as the training data set. Finally, the data packet G is determined as the verification data set, and the data packet E and the data packet F are determined as the training data set.

Further, the adjusting unit 122 adjusts the at least one primary learner according to the verification data set, so as to obtain at least one secondary learner, which includes:

the adjusting unit 122 determines an optimal superparameter point from the verification data set by adopting a superparameter grid search method, and further, the adjusting unit 122 adjusts the at least one primary learner through the optimal superparameter point to obtain the at least one secondary learner.

Specifically, the adjusting unit 122 splits the verification data set according to a fixed step length to obtain a target subset, traverses parameters of two end points on the target subset, verifies the at least one primary learner through the parameters of the two end points to obtain a learning rate of each parameter, determines a parameter with the best learning rate as a first super-parameter point, reduces the step length in a neighborhood of the first super-parameter point, continues traversing until the step length is a preset step length, and the obtained super-parameter point is the optimal super-parameter point, and further, the adjusting unit 122 adjusts the at least one primary learner according to the optimal super-parameter point to obtain the at least one secondary learner.

The preset step length is not limited.

In at least one embodiment of the present invention, after the target model is obtained, the obtaining unit 110 obtains, at every second preset time, second historical data of the acceleration sensor on the driver terminal device to be tested, the determining unit 113 determines the second historical data as a target verification data set, and further, the adjusting unit 122 performs adaptive adjustment on the target model according to the target verification data set, so as to obtain a target model with adaptive capability.

The determination unit 113 determines whether a walking feature or a stationary feature is included in the recognition result.

In at least one embodiment of the present invention, when it is detected that the recognition result does not include the walking feature and the stationary feature, the determination unit 113 determines that the driver under test is finished using the terminal device at the current second.

The interference data refers to data interfering with the behavior characteristics of the driver to be tested.

When the walking feature or the stationary feature is included in the recognition result, the determination unit 113 determines whether the walking feature or the stationary feature continues for a second preset time from the current second.

In at least one embodiment of the present invention, the determining unit 113 determines whether the walking feature or the stationary feature last for a second preset time from the current second includes:

The determining unit 113 extracts fifth data from the first data lasting for the second preset time from the current second, and further, the determining unit 113 inputs the fifth data into the target model, resulting in a third result per second.

When it is determined that the walk characteristic or the rest characteristic continues for the second preset time from the current second, the determination unit 113 determines that the driver under test has ended using the terminal device from the current second and enters a walk or rest state.

In at least one embodiment of the present invention, when it is detected that the at least one third result is the walking feature or the stationary feature, the determination unit 113 determines that the walking feature or the parking feature continues for the second preset time from the current second.

In at least one embodiment of the present invention, in the first preset time, the determining unit 113 determines at least one third time when the driver to be tested starts to use the terminal device when driving, further, the obtaining unit 110 obtains at least one fourth time when the driver to be tested ends to use the terminal device when driving, the calculating unit 112 performs subtraction operation on the at least one fourth time and the at least one third time adjacent to the fourth time, so as to obtain at least one target time when the driver to be tested uses the terminal device when driving, the determining unit 113 determines the target time when the driver to be tested uses the terminal device when driving according to the at least one target time, further, the calculating unit 112 calculates the number of the at least one target time when the driver to be tested ends to use the terminal device when driving, in the first preset time, the calculating unit 112 divides the target time by the first preset time, the frequency when the driver to obtain the terminal device to be tested uses the terminal device to generate the report 117, and the time when the driver to be tested uses the terminal device to be tested generates the report according to the frequency.

Specifically, the determining unit 113 determines at least one third time when the driver to be tested starts to use the terminal device at the time of driving includes:

when it is detected that the behavior feature of the driver under test at the target second is the first result and the behavior feature at the last second of the target second is the second result, the determination unit 113 determines the target second as the third time.

Fig. 3 is a schematic structural diagram of an electronic device according to a preferred embodiment of the present invention for implementing the driving behavior detection method.

In one embodiment of the invention, the electronic device 1 includes, but is not limited to, a memory 12, a processor 13, and a computer program, such as a driving behavior detection program, stored in the memory 12 and executable on the processor 13.

It will be appreciated by those skilled in the art that the schematic diagram is merely an example of the electronic device 1 and does not constitute a limitation of the electronic device 1, and may include more or less components than illustrated, or may combine certain components, or different components, e.g. the electronic device 1 may further include input-output devices, network access devices, buses, etc.

The processor 13 may be a central processing unit (Central Processing Unit, CPU), but may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. The general purpose processor may be a microprocessor or the processor may be any conventional processor, etc., and the processor 13 is an operation core and a control center of the electronic device 1, connects various parts of the entire electronic device 1 using various interfaces and lines, and executes an operating system of the electronic device 1 and various installed applications, program codes, etc.

The processor 13 executes the operating system of the electronic device 1 and various types of applications installed. The processor 13 executes the application program to implement the steps in the respective driving behavior detection method embodiments described above, such as the steps shown in fig. 1.

Illustratively, the computer program may be partitioned into one or more modules/units that are stored in the memory 12 and executed by the processor 13 to complete the present invention. The one or more modules/units may be a series of instruction segments of a computer program capable of performing a specific function for describing the execution of the computer program in the electronic device 1.

The memory 12 may be used to store the computer program and/or module, and the processor 13 may implement various functions of the electronic device 1 by running or executing the computer program and/or module stored in the memory 12 and invoking data stored in the memory 12. The memory 12 may mainly include a memory program area and a memory data area. In addition, the memory 12 may include non-volatile memory, such as a hard disk, memory, plug-in hard disk, smart Media Card (SMC), secure Digital (SD) Card, flash Card (Flash Card), at least one disk storage device, flash memory device, or other non-volatile solid state storage device.

The integrated modules/units of the electronic device 1 may be stored in a computer readable storage medium if implemented in the form of software functional units and sold or used as separate products. Based on such understanding, the present invention may implement all or part of the flow of the method of the above embodiment, or may be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, and when the computer program is executed by a processor, the computer program may implement the steps of each of the method embodiments described above.

In connection with fig. 1, the memory 12 in the electronic device 1 stores a plurality of instructions to implement a driving behavior detection method, the processor 13 being executable to implement: when a driving behavior detection instruction is received, acquiring gesture data of an acceleration sensor and pressure data of a touch screen sensor on a driver terminal device to be detected within a first preset time; preprocessing the gesture data and the pressure data to obtain first data preprocessed by the gesture data and second data preprocessed by the pressure data; calculating the first data by adopting an arcsine function to obtain a pitch angle of the terminal equipment per second; determining the behavior characteristics of the driver to be tested every second according to the pitch angle every second and the corresponding second data; when detecting that the behavior characteristic of the current second is a second result and the behavior characteristic of the second previous to the current second is a first result, extracting target data of the current second from the first data; inputting the target data into a pre-constructed target model to obtain a recognition result; determining whether the identification result comprises walking characteristics or static characteristics; when the walking characteristic or the static characteristic is included in the identification result, determining whether the walking characteristic or the static characteristic lasts for a second preset time from the current second; and when the walking characteristic or the static characteristic is determined to last for the second preset time from the current second, determining that the driver to be tested finishes using the terminal device from the current second and enters a walking or static state.

Specifically, the specific implementation method of the above instructions by the processor 13 may refer to the description of the relevant steps in the corresponding embodiment of fig. 1, which is not repeated herein.

In the several embodiments provided in the present invention, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be other manners of division when actually implemented.

Furthermore, it is evident that the word "comprising" does not exclude other elements or steps, and that the singular does not exclude a plurality. A plurality of units or means recited in the system claims can also be implemented by means of software or hardware by means of one unit or means. The terms second, etc. are used to denote a name, but not any particular order.

Finally, it should be noted that the above-mentioned embodiments are merely for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A driving behavior detection method, characterized in that the method comprises:

2. The driving behavior detection method according to claim 1, wherein the preprocessing the posture data and the pressure data to obtain the first data preprocessed by the posture data and the second data preprocessed by the pressure data includes:

3. The driving behavior detection method according to claim 1, wherein determining the behavior feature of the driver to be measured per second based on the pitch angle per second and the second data comprises:

4. The driving behavior detection method according to claim 1, characterized in that after determining that the behavior characteristic of the driver under test is the first result, the method further comprises:

acquiring a first time corresponding to the first result;

generating prompt information according to the road information;

sending the prompt information to the terminal equipment;

and storing the face information into a configuration library.

5. The driving behavior detection method according to claim 1, characterized in that before inputting the target data into a pre-built target model to obtain a recognition result, the method further comprises:

training the training data set to obtain at least one primary learner;

6. The driving behavior detection method according to claim 1, characterized in that the method further comprises:

7. The driving behavior detection method according to claim 1, characterized in that the method further comprises:

8. A driving behavior detection apparatus, characterized in that the apparatus comprises:

9. An electronic device, the electronic device comprising:

A processor executing instructions stored in the memory to implement the driving behavior detection method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized by: the computer-readable storage medium has stored therein at least one instruction that is executed by a processor in an electronic device to implement the driving behavior detection method according to any one of claims 1 to 7.