An artificial intelligence-based driving behavior real-time monitoring device

Abstract

The application discloses a kind of driving behavior real-time monitoring device based on artificial intelligence, specifically related to traffic control technical field, including mobile terminal, mobile terminal is connected with image acquisition module, database and driving state module;Image acquisition module is used to obtain the image information of vehicle driving position, and image information is transmitted to mobile terminal;Driving state module is used to obtain the state data of vehicle in driving process;Mobile terminal is used to process data information;Database is used to store the data for processing.The present application can regulate and remind the non-standard driving behavior of the driver, reduce the phenomenon that the driver falls asleep, is tired or other distraction in driving process, improve the control of the driver to vehicle to improve driving safety, reduce the possibility of traffic accident.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of traffic control, and more particularly, to a driving behavior real-time monitoring device based on artificial intelligence. BACKGROUND

[0002] Illegal driving behavior and dangerous driving behavior of motor vehicle drivers are one of the main reasons leading to traffic accidents, so pre-identifying illegal driving behavior and dangerous driving behavior and correcting and guiding them can effectively reduce the traffic accident rate.

[0003] In recent years, the number of cars in China has expanded rapidly, and the safety of automobile traffic has attracted increasing attention. How to avoid and reduce traffic accidents has become a topic of active research for scientists, with the main goal being to prevent traffic accidents and reduce injuries in advance. Non-standard driving behavior is one of the main factors leading to traffic accidents. Non-standard driving behavior includes answering the phone with one hand while driving, not looking ahead for a long time, and dozing off while driving, etc. Drivers in a state of fatigue will have their attention dispersed, their thinking activity reduced, and thus their reaction delayed and their vehicle control ability decreased, increasing the likelihood of a traffic accident. SUMMARY

[0004] To achieve the above object, the present application provides the following technical scheme: a driving behavior real-time monitoring device based on artificial intelligence, comprising a mobile terminal, the mobile terminal being connected with an image acquisition module, a database and a driving state module;

[0005] The image acquisition module is used to acquire image information of a vehicle driving position and transmit the image information to the mobile terminal;

[0006] The driving state module is used to acquire state data of the vehicle during driving;

[0007] The mobile terminal is used to process data information;

[0008] The database is used to store processed data.

[0009] In a preferred embodiment, the mobile terminal comprises a face positioning module, a motion capture module and a state analysis module;

[0010] The face positioning module is used to position the face information of the driver in the image information and then extract the face information to the state analysis module;

[0011] The motion capture module is used to capture the motion information of the driver in the image information and then extract the motion information to the state analysis module;

[0012] The state analysis module completes driving behavior analysis of the driver according to the above information.

[0013] In a preferred embodiment, the driving state module obtains vehicle driving data including vehicle speed V and vehicle steering data.

[0014] In a preferred embodiment, the action analysis step of the state analysis module is specifically:

[0015] Step one, according to the driving seat image information obtained by the image acquisition module, three-dimensional modeling is carried out, a three-dimensional model of the driving seat is constructed, a safe driving range is demarcated in the three-dimensional model, and the three-dimensional data of the safe driving range is exported and marked as L1, W1 and H1;

[0016] Step two, read the action information obtained by the action capture module, construct the three-dimensional image information of the user's limbs in the three-dimensional model according to the size data in the image information, obtain the three-dimensional data L2, W2 and H2 of the three-dimensional image of the limbs, compare the three-dimensional data with L1, W1 and H1, and obtain the relative values L1-L2, W1-W2 and H2. 1- H2;

[0017] Step three, extract the three-dimensional data with a relative value less than 0, read the information of the corresponding dimension, and indicate that the driver's limbs exceed the safe range in that dimension.

[0018] In a preferred embodiment, the action analysis step of the state analysis module is specifically:

[0019] Step one, the user records the face information through the image acquisition module to provide standard comparison data for face recognition of the state analysis module;

[0020] Step two, the face positioning module extracts the user's face image dynamic information from the driving image information obtained by the image acquisition module, and constructs a three-dimensional model of the face image dynamic information using the state analysis module.

[0021] Step three, input the recorded standard comparison data parameters into the three-dimensional modeling to realize comparison of the face data.

[0022] In a preferred embodiment, the standard comparison data recorded in step one includes face data when the user is sleeping, face data when the user is drowsy, face data when the user is talking on the phone and distracted, and normal face data of the user.

[0023] In a preferred embodiment, the face data comparison in step three is to mark a certain number of points in the user's face dynamic three-dimensional model, obtain the number of points that coincide with the user's standard comparison data inputted, obtain the percentage of the number of coinciding points, and supervise the driving behavior according to the percentage ratio a.

[0024] Technical effects and advantages of the present application:

[0025] The present application can regulate and remind the non-standard driving behavior of the driver, reduce the phenomenon of the driver falling asleep, fatigue or other distraction in the driving process, improve the control of the driver on the vehicle and improve the driving safety, and reduce the possibility of traffic accidents. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 It is a schematic diagram of the overall structure of the present application. DETAILED DESCRIPTION

[0027] The technical solutions in the embodiments of the present application will be described clearly and completely below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0028] As Figure 1 shown in one kind of driving behavior real-time monitoring device based on artificial intelligence, including mobile terminal, mobile terminal is connected with image acquisition module, database and running state module;

[0029] The image acquisition module is used for acquiring image information of the vehicle driving position, and transmitting the image information to the mobile terminal;

[0030] The running state module is used for acquiring the state data of the vehicle in the driving process;

[0031] The mobile terminal is used for processing data information;

[0032] The database is used for storing the processed data.

[0033] In one embodiment, the mobile terminal includes a face positioning module, an action capture module and a state analysis module, the face positioning module is used for positioning the face information of the driver in the image information, and then extracting the face information to the state analysis module, the action capture module is used for capturing the action information of the driver in the image information, and then extracting the action information to the state analysis module, and the state analysis module completes the driving behavior analysis of the driver according to the above information.

[0034] Further, the running data of the vehicle acquired by the running state module includes the vehicle running speed V and the vehicle steering data.

[0035] In one embodiment, the action analysis step of the state analysis module is specifically:

[0036] Step one, according to the image acquisition module to obtain the driving position image information, three-dimensional modeling, build driving position three-dimensional model, in the three-dimensional model to determine the safe driving range, the three-dimensional data of the safe driving range is exported, marked as L1, W1 and H1;

[0037] Step two, read the action information obtained by the motion capture module, construct the user's limb three-dimensional image information in the three-dimensional model according to the size data in the image information, obtain the three-dimensional data L2, W2 and H2 of the three-dimensional image, compare the three-dimensional data with L1, W1 and H1, and obtain the relative values L1-L2, W1-W2 and H 1- H2;

[0038] Step three, the three-dimensional data with a relative value less than 0 is extracted separately, and the information of its corresponding dimension is read, which indicates that the driver's limbs exceed the safe range in that dimension.

[0039] On the basis of the above, when the relative value W1-W2 is less than 0, it means that the user's limbs extend beyond the normal distance in the horizontal direction, such as extending out of the window, extending to the rear side or extending to the co-driver position;

[0040] Further, due to the existence of some special situations in driving, the user's limbs may extend out of the window, extend to the rear side or extend to the co-driver, so when analyzing the action, the time when the relative value exists is detected, and within a certain time range, the relative value less than 0 is a reasonable state;

[0041] Further, since the vehicle driving state changes, the relative value will also change, specifically: when the driving state module obtains that the vehicle is stationary, even if the driver's limbs extend out of the window, it is reasonable, so according to the state information obtained by the driving state module, the reasonable range of the relative value will also change.

[0042] Specifically, during the process of vehicle parking, the relative value less than 0 is allowed, and during the process of vehicle deceleration, the relative value less than 0 within a certain time is allowed.

[0043] In one embodiment, the action analysis step of the state analysis module is specifically:

[0044] Step one, the user records the face information through the image acquisition module, providing the standard comparison data for face recognition of the state analysis module;

[0045] Step two, the face positioning module extracts the user's face image dynamic information from the driving image information obtained by the image acquisition module, and uses the state analysis module to construct a face dynamic three-dimensional model.

[0046] Step three, input the recorded standard comparison data parameters into the three-dimensional modeling to realize the comparison of facial data.

[0047] The standard comparison data recorded in step one includes: facial data when the user is sleeping, facial data when the user is drowsy, facial data when the user is talking on the phone and distracted, normal facial data of the user;

[0048] The facial data comparison in step three is to mark a certain number of points in the dynamic three-dimensional model of the user's face, and obtain the number of points that coincide with the user's standard comparison data input into the dynamic three-dimensional model, obtain the percentage of the number of coinciding points, and supervise the driving behavior according to the percentage ratio a.

[0049] Further, since the user is in different states, the dangerous driving behavior has different harmfulness, therefore, in the process of facial recognition, the determination of comparison with different standard comparison data is also different;

[0050] When compared with the normal face of the user, the percentage ratio a is not limited in range;

[0051] When compared with the facial data when the user is sleeping, the percentage ratio a is limited in the range of 25%-30%;

[0052] When compared with the facial data when the user is talking on the phone and distracted, the percentage ratio a is limited in the range of 70%-80%;

[0053] When compared with the facial data when the user is drowsy, the percentage ratio a is limited in the range of 40%-50%.

[0054] By the above-mentioned way, the driving behavior of the driver is monitored in real time, and when the irregular phenomenon occurs, the additional alarm device (audio device) is used to remind, which can standardize and remind the irregular driving behavior of the driver, reduce the phenomenon of the driver falling asleep, fatigue or other distraction during driving, improve the control of the driver on the vehicle, thereby improve the driving safety and reduce the possibility of traffic accidents.

[0055] Finally, it should be pointed out that: first of all, in the description of the present application, it should be pointed out that, unless otherwise specified and limited, the terms "installation", "connection", "connection" should be understood broadly, which can be mechanical connection or electrical connection, or the communication between two elements, or direct connection, "up", "down", "left", "right" and the like are only used to indicate the relative positional relationship, when the absolute position of the described object changes, the relative positional relationship may change;

[0056] Secondly: the embodiment of the present application discloses only the structure related to the embodiment of the present application, other structures can refer to the general design, and in the case of no conflict, the same embodiment and different embodiments of the present application can be combined with each other;

[0057] Finally: the above only for the preferred embodiment of the present application, and does not limit the present application, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. An apparatus for real-time monitoring of driving behavior based on artificial intelligence, characterized in that, The mobile terminal is connected with an image acquisition module, a database and a driving state module; The image acquisition module is used for acquiring image information of a driving position of a vehicle and transmitting the image information to the mobile terminal; The driving state module is used for acquiring state data of the vehicle during driving; The mobile terminal is used for processing data information; The database is used for storing the processed data; The mobile terminal comprises a face positioning module, a motion capture module and a state analysis module; The face positioning module is used for positioning face information of a driver in the image information and then extracting the face information to the state analysis module; The motion capture module is used for capturing motion information of the driver in the image information and then extracting the motion information to the state analysis module; The state analysis module completes driving behavior analysis of the driver according to the above information; The driving data acquired by the driving state module comprises a vehicle speed V and vehicle steering data; The motion analysis step of the state analysis module is specifically as follows: Step one, according to the image information of the driving position acquired by the image acquisition module, a three-dimensional model is established, a three-dimensional model of the driving position is constructed, a safe driving range is demarcated in the three-dimensional model, and three-dimensional data of the safe driving range is exported and marked as L1, W1 and H1; Step two, read the action information obtained by the action capture module, construct the three-dimensional image information of the user's limbs in the three-dimensional model according to the size data in the image information, obtain the three-dimensional data L2, W2 and H2 of the three-dimensional image of the limbs, compare the three-dimensional data with L1, W1 and H1, obtain the relative values L1-L2, W1-W2 and H 1- H2; Step three, three-dimensional data with a relative value less than 0 is extracted separately, and information of corresponding dimensions is read, indicating that the driver's limbs exceed the safe range in the dimension; The motion analysis step of the state analysis module is specifically as follows: Step one, the user records face information through the image acquisition module to provide standard comparison data for face recognition of the state analysis module, and the comparison data comprises face data when the user is sleeping, face data when the user is sleepy, face data when the user is talking on the phone and distracted, and normal face data of the user; Step two, the face positioning module extracts face image dynamic information of the user from the driving image information acquired by the image acquisition module, and a three-dimensional model of the face image dynamic information is established by using the state analysis module to construct a face dynamic three-dimensional model; Step three, the recorded standard comparison data parameters are input into the three-dimensional modeling to realize comparison of the face data, specifically, a certain number of points are marked in the face dynamic three-dimensional model of the user, and the number of points that coincide with the input standard comparison data of the user is acquired, the percentage of the number of points is acquired, and driving behavior supervision is performed according to the percentage ratio α.

Images