A method, device, equipment and medium for identifying bad driving behaviors

Abstract

An embodiment of the present invention discloses a method, device, equipment and medium for identifying bad driving behaviors. The method includes: obtaining current driving-related data and determining the current driving behavior category of the current driving behavior; determining the driving environment group to which the current driving behavior belongs and the corresponding bad driving behavior threshold according to the current driving behavior category and a pre-trained bad driving behavior recognition model; determining the current driving behavior characteristics corresponding to the current driving behavior according to the bad driving behavior recognition model, and judging whether the current driving behavior is a bad driving behavior according to the current driving behavior characteristics and the bad driving behavior threshold. By running the technical solution provided by the embodiment of the present invention, it is possible to solve the problem that changes in the environment will cause changes in people and vehicles, while in the prior art, only the driving images of drivers are often recognized, reducing the accuracy of bad driving behavior recognition, and achieve the effect of improving the accuracy and efficiency of bad driving behavior recognition.

Description

Technical Field

[0001] Embodiments of the present invention relate to behavior recognition technology, and in particular, to a method, device, equipment and medium for recognizing bad driving behaviors. Background Art

[0002] With the development and progress of technology, the cost of purchasing a car is getting lower and lower. At the same time, with the overall improvement of people's living standards, more and more people choose to buy private cars for more convenient life. In the case of more and more vehicles on the road, bad driving behaviors can, on the one hand, easily lead to traffic jams, and on the other hand, increase the probability of traffic accidents. Therefore, it is necessary to recognize bad driving behaviors.

[0003] Since the human-vehicle-environment belongs to a closed loop, changes in the environment will cause changes in people and vehicles. However, in the prior art, only the driving images of the driver are often recognized, which reduces the accuracy of bad driving behavior recognition. Summary of the Invention

[0004] Embodiments of the present invention provide a method, device, equipment and medium for recognizing bad driving behaviors to improve the accuracy and efficiency of bad driving behavior recognition.

[0005] In a first aspect, embodiments of the present invention provide a method for recognizing bad driving behaviors, the method comprising:

[0006] Obtaining current driving-related data, and determining a current driving behavior category of the current driving behavior based on the current driving-related data through a preset driving behavior recognition algorithm;

[0007] Determining a driving environment group to which the current driving behavior belongs and a bad driving behavior threshold corresponding to the driving environment group according to the current driving behavior category and a pre-trained bad driving behavior recognition model;

[0008] Obtaining a current driving behavior feature corresponding to the current driving behavior, and determining whether the current driving behavior is a bad driving behavior according to the current driving behavior feature and the bad driving behavior threshold.

[0009] In a second aspect, embodiments of the present invention further provide a device for recognizing bad driving behaviors, the device comprising:

[0010] A current behavior category determination module, configured to obtain current driving-related data, and determine a current driving behavior category of the current driving behavior based on the current driving-related data through a preset driving behavior recognition algorithm;

[0011] A grouping threshold determination module, configured to determine the driving environment group to which the current driving behavior belongs and the bad driving behavior threshold corresponding to the driving environment group according to the current driving behavior category and a pre-trained bad driving behavior recognition model;

[0012] A bad driving behavior judgment module, configured to obtain the current driving behavior characteristics corresponding to the current driving behavior, and judge whether the current driving behavior is a bad driving behavior according to the current driving behavior characteristics and the bad driving behavior threshold.

[0013] In a third aspect, an embodiment of the present invention further provides an electronic device, which includes:

[0014] One or more processors;

[0015] A storage device, configured to store one or more programs,

[0016] When the one or more programs are executed by the one or more processors, the one or more processors implement the bad driving behavior recognition method as described above.

[0017] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the bad driving behavior recognition method as described above is implemented.

[0018] In the embodiment of the present invention, by obtaining current driving-related data and through a preset driving behavior recognition algorithm, the current driving behavior category of the current driving behavior is determined based on the current driving-related data; according to the current driving behavior category and a pre-trained bad driving behavior recognition model, the driving environment group to which the current driving behavior belongs and the bad driving behavior threshold corresponding to the driving environment group are determined; the current driving behavior characteristics corresponding to the current driving behavior are determined according to the bad driving behavior recognition model, and it is judged whether the current driving behavior is a bad driving behavior according to the current driving behavior characteristics and the bad driving behavior threshold. The problem that changes in the environment will cause changes in people and vehicles, and in the prior art, only the driving images of the driver are often recognized, reducing the accuracy of bad driving behavior recognition, is solved, and the effect of improving the accuracy and efficiency of bad driving behavior recognition is achieved. Description of the Drawings

[0019] Figure 1 It is a flowchart of a bad driving behavior recognition method provided by Embodiment 1 of the present invention;

[0020] Figure 2 It is a flowchart of a bad driving behavior recognition model training process provided by Embodiment 2 of the present invention;

[0021] Figure 3Schematic structural diagram of a bad driving behavior recognition device provided in Embodiment 3 of the present invention;

[0022] Figure 4 Schematic structural diagram of an electronic device provided in Embodiment 4 of the present invention. Detailed implementation manners

[0023] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the present invention, rather than limiting the present invention. In addition, it should be noted that for the convenience of description, only parts related to the present invention are shown in the drawings, rather than all the structures.

[0024] Embodiment 1

[0025] Figure 1 Flowchart of a bad driving behavior recognition method provided in Embodiment 1 of the present invention. This embodiment is applicable to the situation of determining whether the current driving behavior is a bad driving behavior. This method can be executed by the bad driving behavior recognition device provided in the embodiments of the present invention, and the device can be implemented in software and / or hardware. Refer to Figure 1 , the bad driving behavior recognition method provided in this embodiment includes:

[0026] Step 110, obtain current driving-related data, and determine the current driving behavior category of the current driving behavior based on the current driving-related data through a preset driving behavior recognition algorithm.

[0027] Among them, the current driving-related data is multi-source data related to driving, such as current weather data, current traffic data, current driving behavior data, etc.

[0028] The preset driving behavior recognition algorithm can be a behavior recognition algorithm based on big data, and this embodiment does not limit this. Calculate the current driving-related data through the preset driving behavior recognition algorithm, and determine the current driving behavior category of the current driving behavior, where the current driving behavior category can be at least one of acceleration, deceleration, automatic emergency braking, steering, braking in a curve, forward collision warning, lane departure warning, electronic stability control of the vehicle body, driving anticipation, idling, etc.

[0029] Step 120, determine the driving environment group to which the current driving behavior belongs and the bad driving behavior threshold corresponding to the driving environment group according to the current driving behavior category and a pre-trained bad driving behavior recognition model.

[0030] Group the current driving behavior according to the current driving behavior category. For example, if the current driving behavior category can be deceleration, then group the current driving behavior into the group related to deceleration.

[0031] Among them, the deceleration-related groups can be multiple groups, and the grouping can be performed according to specific data in the current driving-related data. Exemplarily, deceleration behavior group 1 is sunny weather, congested traffic conditions, urban road for driving, and driving speed of 0 - 20 km / h; deceleration behavior group 2 is sunny weather, congested traffic conditions, rural road for driving, and driving speed of 0 - 20 km / h. The current driving-related data is sunny weather, congested traffic conditions, rural road for driving, and driving speed of 11 km / h, then the grouping of the current driving behavior is group 2.

[0032] Among them, each driving environment group corresponds to a corresponding pre-determined threshold for bad driving behavior. Therefore, when determining the driving environment group to which the current driving behavior belongs, the corresponding threshold for bad driving behavior is determined.

[0033] Step 130: Obtain the current driving behavior characteristics corresponding to the current driving behavior, and determine whether the current driving behavior is a bad driving behavior according to the current driving behavior characteristics and the threshold for bad driving behavior.

[0034] Among them, the current driving behavior characteristics are representative characteristics in the current driving behavior. For example, if the current driving behavior is acceleration, the current driving behavior characteristics can be the maximum acceleration during the acceleration process; if the current driving behavior is steering, the current driving behavior characteristics can be the maximum steering wheel rotation speed during the steering process.

[0035] Compare the current driving behavior characteristics with the threshold for bad driving behavior. If the threshold comparison requirement is met, for example, greater than the threshold for bad driving behavior, then determine that the current driving behavior is a bad driving behavior.

[0036] The technical solution provided in this embodiment determines the driving environment group to which the current driving behavior belongs through the current driving behavior category and the pre-trained bad driving behavior recognition model, and determines whether the current driving behavior is a bad driving behavior according to the current driving behavior characteristics and the threshold for bad driving behavior corresponding to the driving environment group. Different thresholds for bad driving behavior are determined in different driving environments such as different road traffic conditions, avoiding a unified behavior judgment standard in all driving environments, reducing the influence of the driving environment on the judgment of bad driving behavior, and improving the accuracy of the judgment of bad driving behavior. And the bad driving behavior recognition model can be established and trained in advance, improving the efficiency of the recognition of bad driving behavior while improving the accuracy of the recognition of bad driving behavior.

[0037] Embodiment 2

[0038] Figure 2The flowchart of the training process of a bad driving behavior recognition model provided in the second embodiment of the present invention. This technical solution is a supplementary description of the training process of the bad driving behavior recognition model. Compared with the above solution, this solution is specifically optimized as follows: The training process of the bad driving behavior recognition model includes:

[0039] Obtain historical driving-related data, and based on the preset driving behavior recognition algorithm, determine the historical driving behavior categories of each historical driving behavior based on the historical driving-related data;

[0040] Based on the historical driving behavior categories, determine the first driving environment groups to which each of the historical driving behaviors belongs, and obtain the grouping features corresponding to each of the first driving environment groups;

[0041] Determine the candidate influencing factors of the historical driving behavior, and determine the target influencing factors from the candidate influencing factors according to the grouping features;

[0042] Based on the historical driving behavior and the target influencing factors, determine the second driving environment groups to which each of the historical driving behaviors belongs, and determine the bad driving behavior thresholds corresponding to the second driving environment groups according to the members in each of the second driving environment groups. Specifically, the flowchart of the training process of the bad driving behavior recognition model is as Figure 2 shown:

[0043] Step 210: Obtain historical driving-related data, and based on the preset driving behavior recognition algorithm, determine the historical driving behavior categories of each historical driving behavior based on the historical driving-related data.

[0044] Among them, the historical driving-related data is multi-source data related to driving, such as historical weather data, historical traffic data, historical driving behavior data, etc.

[0045] The preset driving behavior recognition algorithm can be a behavior recognition algorithm based on big data, and this embodiment does not limit this. Calculate the historical driving-related data through the preset driving behavior recognition algorithm to determine the historical driving behavior categories of the historical driving behaviors, where the historical driving behavior categories can be at least one of acceleration, deceleration, automatic emergency braking, steering, curve braking, forward collision warning, lane departure warning, electronic stability control of the vehicle body, driving anticipation, idling, etc.

[0046] In this embodiment, optionally, before determining the first driving environment groups to which each of the historical driving behaviors belongs based on the historical driving-related data, it further includes:

[0047] Determine the preset screening conditions according to the historical driving behavior categories, and screen the historical driving behaviors according to the preset screening conditions.

[0048] The historical driving behaviors for which the historical driving behavior categories have been determined are screened according to preset screening conditions. The preset screening conditions corresponding to different historical driving behavior categories may be different. Exemplarily, if the historical driving behavior category is acceleration, the preset screening condition may be to retain the historical driving behaviors in which the maximum acceleration during the acceleration process is greater than 0, the duration of the acceleration process is greater than 1 s, and the vehicle speed change during the acceleration process is greater than 5 km / h. Thereby enhancing the pertinence of the training data and improving the accuracy of subsequent model training.

[0049] Step 220: Determine the first driving environment group to which each of the historical driving behaviors belongs based on the historical driving behavior category, and obtain the grouping features corresponding to each of the first driving environment groups.

[0050] According to the historical driving behavior category, the historical driving behaviors are grouped. For example, if the historical driving behavior category is acceleration, the historical driving behaviors are grouped into the first driving environment group related to acceleration.

[0051] Among them, there can be multiple first driving environment groups related to acceleration, and they can be grouped according to the historical driving related data in the historical driving related data. Exemplarily, the historical driving related data of the acceleration behavior can be the road traffic environment information during the acceleration process, the maximum acceleration during the acceleration process and the corresponding vehicle speed and vehicle information, etc. Determine the first driving environment group to which each historical driving behavior belongs based on one or more of the historical driving related data.

[0052] Exemplarily, acceleration behavior group 1 is sunny weather, congested traffic conditions, urban road for driving, and driving speed of 0 - 20 km / h, and acceleration behavior group 2 is sunny weather, smooth traffic conditions, urban road for driving, and driving speed of 0 - 20 km / h. The historical driving related data is sunny weather, smooth traffic conditions, urban road for driving, and driving speed of 11 km / h, then the grouping of the historical driving behavior is group 2.

[0053] After grouping all the historical driving behaviors, determine the grouping features corresponding to each of the first driving environment groups. Among them, the grouping feature is the representative feature of the first driving environment group. For example, the grouping feature can be the median of the maximum acceleration during the acceleration process of each acceleration behavior in this first driving environment group.

[0054] In this embodiment, optionally, obtaining the grouping features corresponding to each of the first driving environment groups includes:

[0055] Determine the corresponding historical driving behavior features and historical statistical features according to the historical driving behavior category corresponding to the first driving environment group;

[0056] Obtain the grouping feature according to the historical driving behavior feature and the historical statistical feature.

[0057] The grouping feature is determined according to the historical driving behavior feature and the historical statistical feature corresponding to the historical driving behavior category, where the historical driving behavior feature is a representative feature in the historical driving behavior, and the historical statistical feature is a statistical feature of all driving behavior features in each driving environment grouping.

[0058] Exemplarily, when the historical driving behavior category is acceleration, the historical driving behavior feature is the maximum acceleration during the acceleration process, and the historical statistical feature is the median, then the grouping feature can be the median of the maximum acceleration during the acceleration process of each historical driving behavior in the first driving environment grouping.

[0059] Among them, the driving behavior features and statistical features corresponding to each driving behavior category are shown in Table 1 below:

[0060]

[0061]

[0062] According to the content in Table 1, the grouping features corresponding to each first driving environment grouping can be statistically determined, thereby improving the pertinence of the determination of the grouping features and the representativeness of the grouping features for the first driving environment grouping.

[0063] Step 230, determine the candidate influencing factors of the historical driving behavior, and determine the target influencing factor from the candidate influencing factors according to the grouping feature.

[0064] Among them, the candidate influencing factors of the historical driving behavior can be all possible influencing factors, or the influencing factors preliminarily screened according to the historical driving behavior category, and this embodiment does not limit this. Exemplarily, the candidate influencing factors are vehicle speed range, road type, traffic status, and weather status.

[0065] By analyzing the correlation relationship between the grouping feature and the candidate influencing factors, the candidate influencing factors with a greater degree of correlation are determined as the target influencing factors.

[0066] In this embodiment, optionally, determining the candidate influencing factors of the historical driving behavior and determining the target influencing factor from the candidate influencing factors according to the grouping feature includes:

[0067] Determine the candidate influencing factors of the historical driving behavior corresponding to the driving behavior category of the first driving environment grouping;

[0068] Analyze the correlation relationship between each candidate influencing factor and the grouping feature according to the partial correlation analysis method;

[0069] Determine the target influencing factor from the candidate influencing factors according to the above-mentioned correlation relationship.

[0070] Among them, the candidate influencing factors corresponding to different driving behavior categories may be the same or different. If the candidate influencing factors are different, the corresponding relationship can be determined by preliminary statistics in advance. Exemplarily, the candidate influencing factors affecting the deceleration behavior may be road type and traffic status, and the candidate influencing factors affecting the acceleration behavior may be vehicle speed range, road type, traffic status, and weather status.

[0071] After determining the candidate influencing factors according to the driving behavior category, analyze the correlation relationship between each candidate influencing factor and the grouping characteristics according to the partial correlation analysis method.

[0072] Partial correlation analysis uses the method of controlling variables to eliminate the influence and interference of other variables, and studies the correlation relationship between two variables. First, check whether the significance level shows significance, then analyze the direction of the correlation relationship through the positive or negative of the correlation coefficient, and at the same time, the tightness of the relationship can also be explained by the magnitude of the correlation coefficient.

[0073] If the correlation relationship between the candidate influencing factor and the grouping characteristics is greater than the preset threshold, then determine the candidate influencing factor as the target influencing factor.

[0074] Exemplarily, when the candidate influencing factors are vehicle speed range, road type, traffic status, and weather status, and the driving behavior characteristic corresponding to the grouping characteristic is the maximum acceleration during the acceleration process, the partial correlation analysis process is as follows: during the analysis of the influence of a certain variable, fix the values of the other three variables. The partial correlation analysis results include the average value, standard deviation, as well as the partial correlation coefficient and significance level of the two. Among them, the significance level of the vehicle speed range on the acceleration behavior characteristic is p < 0.01, and the partial correlation coefficient is -0.903. That is to say, the vehicle speed range has a significant influence on the acceleration behavior characteristic, and as the vehicle speed increases, the acceleration behavior characteristic becomes smaller.

[0075] The partial correlation analysis results between the road type and the acceleration behavior characteristic show that there is no significant correlation relationship between the two, that is, based on historical data, different road types have no significant correlation relationship with the acceleration behavior.

[0076] The partial correlation analysis results between the traffic status and the acceleration behavior characteristic show that there is no significant correlation relationship between the two, that is, based on historical data, different traffic statuses have no significant correlation relationship with the acceleration behavior.

[0077] The partial correlation analysis results between the weather status and the acceleration behavior characteristic show that there is no significant correlation relationship between the two, that is, based on historical data, different weather statuses have no significant correlation relationship with the acceleration behavior.

[0078] In summary, when other factors are controlled to remain unchanged, there is an obvious correlation between the vehicle speed range and the maximum acceleration during the acceleration process. The partial correlation analysis results of road type, traffic status, and weather status are not significant, and it can be determined that the target influencing factor is the vehicle speed range.

[0079] Taking the candidate influencing factors as independent variables and the grouping characteristics as dependent variables, the partial correlation coefficient is used to describe the influence of a certain independent variable on historical driving behavior when other independent variables are determined, so as to solve the problem that when there are multiple candidate influencing factors, if ordinary correlation analysis is used, the influence of other factors cannot be eliminated, and the accuracy of determining the target influencing factor is improved.

[0080] Step 240: Based on the historical driving behavior and the target influencing factor, determine the second driving environment group to which each historical driving behavior belongs, and determine the threshold of bad driving behavior corresponding to the second driving environment group according to the members in each second driving environment group.

[0081] According to the historical driving behavior category, group the historical driving behaviors. For example, if the historical driving behavior category is acceleration, then group the historical driving behaviors into the second driving environment group related to acceleration.

[0082] Among them, the classification factor in the second driving environment group is the target influencing factor. There can be multiple second driving environment groups. Exemplarily, acceleration behavior group 1 has a driving speed range of 0 - 20 km / h, and acceleration behavior group 2 has a driving speed of 20 - 40 km / h. If the historical driving related data shows a driving speed of 11 km / h, then the grouping of the historical driving behavior is group 1.

[0083] The driving behavior characteristics of the second driving environment group can be determined according to the members in each second driving environment group, and the threshold of bad driving behavior corresponding to the second driving environment group can be determined according to the driving behavior characteristics. For example, the driving behavior characteristics can be the maximum acceleration during the acceleration process of each acceleration behavior in this second driving environment group. The driving behavior characteristics can be obtained in the manner of the driving behavior characteristics described above in this embodiment, which will not be elaborated here.

[0084] The determination method of the bad driving behavior threshold can be based on multiple batches of data grouped under the same group, and obtain the corresponding multiple driving behavior characteristics, so as to statistically obtain the bad driving behavior threshold corresponding to this second driving environment group. This embodiment does not limit this.

[0085] In this embodiment, optionally, determining the threshold of bad driving behavior corresponding to each second driving environment group according to the members in each second driving environment group includes:

[0086] Determine the bad driving behavior thresholds corresponding to each of the second driving environment groups according to the objective analysis method and the members in each of the second driving environment groups.

[0087] The objective evaluation method for determining the behavior evaluation threshold can be through the analysis of the historical driving behavior characteristics of a large number of samples. Exemplarily, select the historical driving behavior characteristic values corresponding to 98% or other quantiles as the bad driving behavior thresholds. When the historical driving behavior characteristic value of an acceleration behavior exceeds the historical driving behavior characteristic values of 98% of the members in its corresponding second driving environment group, determine that the acceleration behavior is a bad driving behavior. When the number of samples is large, the distribution form of the samples tends to be stable, and the result credibility is high.

[0088] Solve the problem that the subjective evaluation method requires professional evaluators to conduct a large number of on-road vehicle tests, it is difficult to distinguish the acceleration difference of 0.1 m / s 2 and the subjective feelings of people are affected by various factors, making it difficult to achieve a unified and stable evaluation standard.

[0089] In the embodiments of the present invention, the second driving environment group to which each historical driving behavior belongs is determined based on the historical driving behavior and the target influencing factors, and the bad driving behavior recognition thresholds under different second driving environment groups are determined to train the bad driving behavior recognition model, so as to more precisely identify the bad driving behaviors of different driving behavior categories through the bad driving behavior recognition model on the premise of considering the driving environment.

[0090] Embodiment III

[0091] Figure 3 FIG. is a schematic structural diagram of a bad driving behavior recognition device provided in Embodiment III of the present invention. This device can be implemented in a hardware and / or software manner, and can execute a bad driving behavior recognition method provided in any embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method. As Figure 3 shown, the device includes:

[0092] A current behavior category determination module 310, configured to obtain current driving-related data, and determine the current driving behavior category of the current driving behavior based on the current driving-related data through a preset driving behavior recognition algorithm;

[0093] A grouping threshold determination module 320, configured to determine the driving environment group to which the current driving behavior belongs and the bad driving behavior threshold corresponding to the driving environment group according to the current driving behavior category and a pre-trained bad driving behavior recognition model;

[0094] The bad driving behavior judgment module 330 is configured to obtain the current driving behavior characteristics corresponding to the current driving behavior, and determine whether the current driving behavior is a bad driving behavior according to the current driving behavior characteristics and the bad driving behavior threshold.

[0095] The technical solution provided in this embodiment determines the driving environment group to which the current driving behavior belongs through the current driving behavior category and the pre-trained bad driving behavior recognition model, and determines whether the current driving behavior is a bad driving behavior according to the current driving behavior characteristics and the bad driving behavior threshold corresponding to the driving environment group. Different bad driving behavior thresholds are determined in different driving environments such as road traffic, avoiding a unified behavior judgment standard in all driving environments, reducing the influence of the driving environment on the judgment of bad driving behavior, and improving the accuracy of the judgment of bad driving behavior. And the bad driving behavior recognition model can be established and trained in advance, improving the efficiency of bad driving behavior recognition while improving the accuracy of bad driving behavior recognition.

[0096] Based on the above technical solutions, optionally, the behavior recognition model training module includes:

[0097] The historical behavior category determination module is configured to obtain historical driving-related data, and determine the historical driving behavior categories of each historical driving behavior based on the historical driving-related data through the preset driving behavior recognition algorithm;

[0098] The grouping feature acquisition module is configured to determine the first driving environment group to which each historical driving behavior belongs based on the historical driving behavior category, and acquire the grouping features corresponding to each first driving environment group;

[0099] The target influencing factor determination module is configured to determine the candidate influencing factors of the historical driving behavior, and determine the target influencing factors from the candidate influencing factors according to the grouping features;

[0100] The behavior threshold determination module determines the second driving environment group to which each historical driving behavior belongs based on the historical driving behavior and the target influencing factor, and determines the bad driving behavior threshold corresponding to each second driving environment group according to the members in each second driving environment group.

[0101] Based on the above technical solutions, optionally, the device further includes:

[0102] The behavior screening module is configured to, before the grouping feature acquisition module, determine a preset screening condition according to the historical driving behavior category, and screen the historical driving behavior according to the preset screening condition.

[0103] Based on the above technical solutions, optionally, the grouping feature acquisition module includes:

[0104] A feature determination unit, configured to determine corresponding historical driving behavior features and historical statistical features according to the historical driving behavior categories corresponding to the first driving environment grouping;

[0105] A feature acquisition unit, configured to acquire the grouping features according to the historical driving behavior features and the historical statistical features.

[0106] Based on the above technical solutions, optionally, the target influencing factor determination module includes:

[0107] A candidate influencing factor determination unit, configured to determine candidate influencing factors of the historical driving behavior corresponding to the driving behavior category of the first driving environment grouping;

[0108] A correlation analysis unit, configured to analyze the correlation between each candidate influencing factor and the grouping features according to the partial correlation analysis method;

[0109] A target influencing factor determination unit, configured to determine the target influencing factor from the candidate influencing factors according to the correlation.

[0110] Based on the above technical solutions, optionally, the behavior threshold determination module includes:

[0111] A behavior threshold determination unit, configured to determine the bad driving behavior thresholds corresponding to each second driving environment grouping according to an objective analysis method and the members in each second driving environment grouping.

[0112] Embodiment 4

[0113] Figure 4 FIG. is a schematic structural diagram of an electronic device provided in Embodiment 4 of the present invention. As Figure 4 shown, the electronic device includes a processor 40, a memory 41, an input device 42, and an output device 43; the number of processors 40 in the electronic device may be one or more. Figure 4 Taking one processor 40 as an example; the processor 40, the memory 41, the input device 42, and the output device 43 in the electronic device may be connected through a bus or other means. Figure 4 Taking the connection through a bus as an example.

[0114] The memory 41 serves as a computer-readable storage medium and can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the method for identifying bad driving behaviors in the embodiments of the present invention. The processor 40 executes various functional applications and data processing of the electronic device by running the software programs, instructions, and modules stored in the memory 41, that is, implements the above-mentioned method for identifying bad driving behaviors.

[0115] The memory 41 may mainly include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created according to the use of the terminal, etc. In addition, the memory 41 may include high-speed random access memory and may also include non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage devices. In some instances, the memory 41 may further include a memory remotely set relative to the processor 40, and these remote memories can be connected to the electronic device through a network. Examples of the above network include but are not limited to the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0116] Embodiment Five

[0117] Embodiment Five of the present invention also provides a storage medium containing computer-executable instructions, and the computer-executable instructions are used to execute a method for identifying bad driving behaviors when executed by a computer processor. The method includes:

[0118] Obtain current driving-related data, and determine the current driving behavior category of the current driving behavior based on the current driving-related data through a preset driving behavior recognition algorithm;

[0119] Determine the driving environment group to which the current driving behavior belongs and the bad driving behavior threshold corresponding to the driving environment group according to the current driving behavior category and a pre-trained bad driving behavior recognition model;

[0120] Obtain the current driving behavior characteristics corresponding to the current driving behavior, and determine whether the current driving behavior is a bad driving behavior according to the current driving behavior characteristics and the bad driving behavior threshold.

[0121] Of course, for a storage medium containing computer-executable instructions provided by the embodiments of the present invention, the computer-executable instructions are not limited to the method operations as described above, and can also execute related operations in the method for identifying bad driving behaviors provided by any embodiment of the present invention.

[0122] From the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software and necessary general hardware. Of course, it can also be implemented by hardware, but in many cases, the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as a floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disc of a computer, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0123] It should be noted that in the embodiments of the above-mentioned bad driving behavior recognition device, the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of the functional units are only for the convenience of mutual distinction and do not limit the protection scope of the present invention.

[0124] Note that the above is only the preferred embodiment of the present invention and the technical principles applied. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein. Various obvious changes, re-adjustments and substitutions can be made by those skilled in the art without departing from the protection scope of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments. Without departing from the concept of the present invention, more other equivalent embodiments can be included, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for identifying bad driving behaviors, characterized in that, it includes: Obtain current driving-related data, and determine the current driving behavior category of the current driving behavior based on the current driving-related data through a preset driving behavior recognition algorithm; Determine the driving environment grouping to which the current driving behavior belongs and the bad driving behavior threshold corresponding to the driving environment grouping according to the current driving behavior category and a pre-trained bad driving behavior recognition model; Obtain the current driving behavior characteristics corresponding to the current driving behavior, and determine whether the current driving behavior is a bad driving behavior according to the current driving behavior characteristics and the bad driving behavior threshold; Among them, the training process of the bad driving behavior recognition model includes: Obtain historical driving-related data, and determine the historical driving behavior category of each historical driving behavior based on the historical driving-related data through the preset driving behavior recognition algorithm; Determine the first driving environment grouping to which each historical driving behavior belongs based on the historical driving behavior category, and obtain the grouping characteristics corresponding to each first driving environment grouping; Determine the candidate influencing factors of the historical driving behavior, and determine the target influencing factors from the candidate influencing factors according to the grouping characteristics; Determine the second driving environment grouping to which each historical driving behavior belongs based on the historical driving behavior and the target influencing factors, and determine the bad driving behavior threshold corresponding to the second driving environment grouping according to the members in each second driving environment grouping.

2. The method according to claim 1, characterized in that, before determining the first driving environment grouping to which each historical driving behavior belongs based on the historical driving-related data, it further includes: Determine a preset screening condition according to the historical driving behavior category, and screen the historical driving behavior according to the preset screening condition.

3. The method according to claim 1, characterized in that, obtaining the grouping characteristics corresponding to each first driving environment grouping includes: Determine the corresponding historical driving behavior characteristics and historical statistical characteristics according to the historical driving behavior category corresponding to the first driving environment grouping; Obtain the grouping characteristics according to the historical driving behavior characteristics and the historical statistical characteristics.

4. The method according to claim 1, characterized in that, determining the candidate influencing factors of the historical driving behavior, and determining the target influencing factors from the candidate influencing factors according to the grouping characteristics includes: Determine the candidate influencing factors of the historical driving behavior corresponding to the driving behavior category of the first driving environment grouping; Analyze the correlation between each candidate influencing factor and the grouping characteristics according to the partial correlation analysis method; Determine the target influencing factors from the candidate influencing factors according to the correlation.

5. The method according to claim 1, characterized in that, determining the bad driving behavior threshold corresponding to each second driving environment grouping according to the members in each second driving environment grouping includes: Determine the bad driving behavior threshold corresponding to each of the second driving environment groups according to the objective analysis method and the members in each of the second driving environment groups.

6. An apparatus for identifying bad driving behavior, characterized in that, comprising: A current behavior category determination module, configured to obtain current driving-related data, and determine the current driving behavior category of the current driving behavior based on the current driving-related data through a preset driving behavior recognition algorithm; A grouping threshold determination module, configured to determine the driving environment group to which the current driving behavior belongs and the bad driving behavior threshold corresponding to the driving environment group according to the current driving behavior category and a pre-trained bad driving behavior recognition model; A bad driving behavior judgment module, configured to obtain the current driving behavior characteristics corresponding to the current driving behavior, and judge whether the current driving behavior is a bad driving behavior according to the current driving behavior characteristics and the bad driving behavior threshold; Wherein, the apparatus further includes a behavior recognition model training module, and the behavior recognition model training module includes: A historical behavior category determination module, configured to obtain historical driving-related data, and determine the historical driving behavior category of each historical driving behavior based on the historical driving-related data through the preset driving behavior recognition algorithm; A grouping feature acquisition module, configured to determine the first driving environment group to which each historical driving behavior belongs based on the historical driving behavior category, and acquire the grouping features corresponding to each first driving environment group; A target influencing factor determination module, configured to determine the candidate influencing factors of the historical driving behavior, and determine the target influencing factor from the candidate influencing factors according to the grouping features; A behavior threshold determination module, configured to determine the second driving environment group to which each historical driving behavior belongs based on the historical driving behavior and the target influencing factor, and determine the bad driving behavior threshold corresponding to each second driving environment group according to the members in each second driving environment group.

7. An electronic device, characterized in that, the electronic device includes: One or more processors; A storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the bad driving behavior recognition method according to any one of claims 1-5.

8. A computer-readable storage medium, on which a computer program is stored, characterized in that, When the program is executed by a processor, it implements the bad driving behavior recognition method according to any one of claims 1-5.

Images