Driver behavior recognition method and related device

Abstract

An embodiment of the present application provides a driver behavior recognition method and related devices, the method comprising: obtaining a target image of a driver when driving a target vehicle; obtaining target behavior parameters of the driver through the target image; determining the driver's target driving behavior based on the target behavior parameters; if the target driving behavior is a preset abnormal driving behavior, determining the driver's first target behavior danger level; if the first target behavior danger level is higher than a first preset danger level, issuing an alarm message, which can improve the accuracy of the early warning.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a driver behavior recognition method and related devices. Background Art

[0002] With the continuous development of society, cars have become an integral part of everyday life. With the increasing number of cars, road safety has become a constant concern. Existing solutions often rely solely on the presence of passengers to assess a driver's behavior, resulting in limited accuracy in driver behavior warnings. Summary of the Invention

[0003] The embodiments of the present application provide a driver behavior recognition method and related devices, which can improve the accuracy of driving behavior warnings.

[0004] A first aspect of an embodiment of the present application provides a driver behavior recognition method, the method comprising:

[0005] Acquire a target image when the driver is driving a target vehicle;

[0006] Obtaining target behavior parameters of the driver through the target image;

[0007] determining a target driving behavior of the driver according to the target behavior parameter;

[0008] If the target driving behavior is a preset abnormal driving behavior, determining a first target behavior risk level of the driver;

[0009] If the risk level of the first target behavior is higher than a first preset risk level, an alarm message is issued.

[0010] In conjunction with the first aspect, in one possible implementation, obtaining the target behavior parameter of the driver through the target image includes:

[0011] Performing feature extraction on the target image to obtain feature data;

[0012] determining the driver's action information based on the characteristic data;

[0013] The target behavior parameter is determined according to the action information.

[0014] In conjunction with the first aspect, in one possible implementation, the motion information includes head motion information and hand motion information, and determining the target behavior parameter based on the motion information includes:

[0015] acquiring the driver's head movement information according to the head movement information;

[0016] determining a first reference behavior parameter according to the head movement information;

[0017] determining facial movement information of the driver from the head movement information;

[0018] determining the driver's sight direction and facial expression based on the facial action information;

[0019] determining a second reference behavior parameter according to the gaze direction and the facial expression;

[0020] determining a third reference behavior parameter of the driver according to the hand motion information;

[0021] The target behavior parameter is determined according to the first reference behavior parameter, the second reference behavior parameter, and the first reference behavior parameter.

[0022] In conjunction with the first aspect, in one possible implementation, determining the target driving behavior of the driver based on the target behavior parameter includes:

[0023] determining the driver's behavioral tendency according to the target behavioral parameter;

[0024] determining at least one reference driving behavior based on the behavior tendency;

[0025] Obtaining the target user's driving time and driving starting location;

[0026] A target driving behavior is determined from the at least one reference driving behavior according to the driving duration and the driving starting location.

[0027] In conjunction with the first aspect, in one possible implementation, the method further includes:

[0028] If the first target behavior danger level is higher than the second preset danger level and lower than the first preset danger level, obtaining the location information of the target vehicle, the location information including the location information of the vehicle on the road;

[0029] Determining the current driving lane and lane change information of the target vehicle based on the position information;

[0030] determining a behavior level correction value according to the current driving lane and the lane change information;

[0031] determining a second target behavior risk level according to the first target behavior risk level and the level correction value;

[0032] If the second target behavior danger level is higher than the first preset danger level, the warning information is issued.

[0033] A second aspect of an embodiment of the present application provides a driver behavior recognition device, the device comprising:

[0034] a first acquiring unit, configured to acquire a target image of a driver driving a target vehicle;

[0035] a second acquiring unit, configured to acquire a target behavior parameter of the driver through the target image;

[0036] a first determining unit, configured to determine a target driving behavior of the driver according to the target behavior parameter;

[0037] a second determining unit, configured to determine a first target behavior risk level of the driver if the target driving behavior is a preset abnormal driving behavior;

[0038] The sending unit is configured to send an alarm message if the danger level of the first target behavior is higher than a first preset danger level.

[0039] With reference to the second aspect, in one possible implementation, the second acquiring unit is configured to:

[0040] Performing feature extraction on the target image to obtain feature data;

[0041] determining the driver's action information based on the characteristic data;

[0042] The target behavior parameter is determined according to the action information.

[0043] With reference to the second aspect, in one possible implementation, the motion information includes head motion information and hand motion information. In determining the target behavior parameter based on the motion information, the second acquisition unit is configured to:

[0044] acquiring the driver's head movement information according to the head movement information;

[0045] determining a first reference behavior parameter according to the head movement information;

[0046] determining facial movement information of the driver from the head movement information;

[0047] determining the driver's sight direction and facial expression based on the facial action information;

[0048] determining a second reference behavior parameter according to the gaze direction and the facial expression;

[0049] determining a third reference behavior parameter of the driver according to the hand motion information;

[0050] The target behavior parameter is determined according to the first reference behavior parameter, the second reference behavior parameter, and the first reference behavior parameter.

[0051] With reference to the second aspect, in one possible implementation, the first determining unit:

[0052] determining the driver's behavioral tendency according to the target behavioral parameter;

[0053] determining at least one reference driving behavior based on the behavior tendency;

[0054] Obtaining the target user's driving time and driving starting location;

[0055] A target driving behavior is determined from the at least one reference driving behavior according to the driving duration and the driving starting location.

[0056] In conjunction with the second aspect, in one possible implementation, the apparatus is further configured to:

[0057] If the first target behavior danger level is higher than the second preset danger level and lower than the first preset danger level, obtaining the location information of the target vehicle, the location information including the location information of the vehicle on the road;

[0058] Determining the current driving lane and lane change information of the target vehicle based on the position information;

[0059] determining a behavior level correction value according to the current driving lane and the lane change information;

[0060] determining a second target behavior risk level according to the first target behavior risk level and the level correction value;

[0061] If the second target behavior danger level is higher than the first preset danger level, the warning information is issued.

[0062] A third aspect of an embodiment of the present application provides a terminal, comprising a processor, an input device, an output device, and a memory, wherein the processor, input device, output device, and memory are interconnected, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions and execute the step instructions in the first aspect of the embodiment of the present application.

[0063] The fourth aspect of the embodiments of the present application provides a computer-readable storage medium, wherein the above-mentioned computer-readable storage medium stores a computer program for electronic data exchange, wherein the above-mentioned computer program enables a computer to execute some or all of the steps described in the first aspect of the embodiments of the present application.

[0064] A fifth aspect of the embodiments of the present application provides a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program, wherein the computer program is operable to cause a computer to perform some or all of the steps described in the first aspect of the embodiments of the present application. The computer program product may be a software installation package.

[0065] Implementing the embodiments of the present application has at least the following beneficial effects:

[0066] By acquiring a target image of the driver when driving a target vehicle, the target behavior parameters of the driver are acquired through the target image, and the target driving behavior of the driver is determined based on the target behavior parameters. If the target driving behavior is a preset abnormal driving behavior, the first target behavior danger level of the driver is determined. If the first target behavior danger level is higher than the first preset danger level, an alarm message is issued. Compared with the existing solution, in which whether the driver's driving behavior is abnormal is manually determined, the behavior parameters can be determined by including a target image of the target driver, the driving behavior can be determined based on the behavior parameters, and an alarm message can be issued if the driving behavior is abnormal, thereby improving the accuracy of the driver's behavior judgment. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0068] Figure 1 A schematic diagram of a driver behavior recognition method is provided for an embodiment of the present application;

[0069] Figure 2 A flowchart of a driver behavior recognition method is provided for an embodiment of the present application;

[0070] Figure 3 A schematic diagram of the structure of a terminal provided in an embodiment of the present application;

[0071] Figure 4 A schematic structural diagram of a driver behavior recognition device is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0072] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0073] The terms "first," "second," and the like in the specification and claims of this application and the accompanying drawings are used to distinguish between different objects, not to describe a particular order. Furthermore, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or elements is not limited to the listed steps or elements but may optionally include steps or elements not listed, or may optionally include other steps or elements inherent to the process, method, product, or apparatus.

[0074] References to "embodiments" in this application mean that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described in this application may be combined with other embodiments.

[0075] See also Figure 1 , Figure 1 A schematic diagram of a driver behavior recognition method is provided for an embodiment of the present application. Figure 1 As shown, the behavior recognition system includes a camera and a processor. The camera obtains a target image of a target vehicle, and the camera sends the target image to the processor. The processor obtains the target behavior parameters of the driver through the target image. The processor determines the driver's driving behavior based on the target behavior parameters. If the processor determines that the target driving behavior is a preset abnormal driving behavior, the driver's first target danger level is determined. If the first target danger level is higher than the first preset danger level, the processor issues an alarm message. Therefore, compared with the existing solution, whether the driver's driving behavior is abnormal is determined manually, the behavior parameters can be determined by including the target image of the target driver, the driving behavior can be determined based on the behavior parameters, and an alarm message can be issued if the driving behavior is abnormal, thereby improving the accuracy of the driver's behavior judgment.

[0076] See also Figure 2 , Figure 2The present invention provides a flowchart of a method for identifying a driver's behavior. Figure 2 As shown, the behavior recognition method includes:

[0077] 201. Acquire a target image of a driver driving a target vehicle.

[0078] The target image may include the target vehicle, or may include the target vehicle and multiple other vehicles. The target image may be acquired by a camera, for example, by a traffic camera on the road, or by other cameras, for example, by a camera built into the target vehicle.

[0079] After capturing the target image, the target image may be preliminarily processed. For example, a small image of the target vehicle may be extracted from the target image. The small image of the target vehicle may be understood as an image including only the target vehicle, which may include the target driver and one or more passengers.

[0080] 202. Obtain target behavior parameters of the driver through the target image.

[0081] The target behavior parameter may be a driving behavior parameter, such as driving posture, mental state while driving, sight direction while driving, etc.

[0082] According to the target image, a method for obtaining the target behavior parameter may be to extract features from the target image to obtain feature data, and then determine the target behavior parameter according to the feature data.

[0083] The method for extracting features from a target image can be performed using a pre-defined feature extraction network trained using sample data. The target image can be directly input into the feature extraction network, which then extracts features to obtain feature data. This can improve the efficiency and accuracy of feature data acquisition.

[0084] 203. Determine a target driving behavior of the driver according to the target behavior parameter.

[0085] The driver's behavior tendency can be determined based on the target behavior parameter, and the target driving behavior can be determined based on the behavior tendency. The driving behavior can be normal driving and abnormal driving, etc. Normal driving can be understood as driving in accordance with traffic rules, etc.

[0086] 204. If the target driving behavior is a preset abnormal driving behavior, determine a first target behavior danger level of the driver.

[0087] Different abnormal driving behaviors correspond to different danger levels. The higher the danger level, the more dangerous the driver is and the more likely a traffic accident is to occur.

[0088] 205. If the danger level of the first target behavior is higher than a first preset danger level, an alarm message is issued.

[0089] Warning information can be sent through the car's audio equipment, etc. The warning information can be pre-set warning information, which is set based on historical data or experience. When sending warning information, it can be sent at different frequencies. The higher the danger level, the higher the sending frequency, and the lower the danger level, the lower the sending frequency.

[0090] In this example, a target image of the driver driving a target vehicle is obtained, and the target behavior parameters of the driver are obtained through the target image. The target driving behavior of the driver is determined based on the target behavior parameters. If the target driving behavior is a preset abnormal driving behavior, the first target behavior danger level of the driver is determined. If the first target behavior danger level is higher than the first preset danger level, an alarm message is issued. Compared with the existing solution, in which whether the driver's driving behavior is abnormal is determined manually, the behavior parameters can be determined by including a target image of the target driver, the driving behavior can be determined based on the behavior parameters, and an alarm message can be issued if the driving behavior is abnormal, thereby improving the accuracy of the driver's behavior judgment.

[0091] In one possible implementation, a possible method for obtaining the target behavior parameter of the driver through the target image includes:

[0092] A1. Extracting features from the target image to obtain feature data;

[0093] A2. determining the driver's action information based on the characteristic data;

[0094] A3. Determine the target behavior parameters based on the action information.

[0095] The method for extracting features from the target image can be to extract features through a feature extraction network to obtain feature data. The feature extraction network is a pre-trained network that has a feature extraction function.

[0096] Based on the characteristic information, displacement information of different joints of the driver can be determined, and action information can be determined based on the displacement information. The displacement information can be relative to a reference point, which can be the point on the target user's body when standing at attention. The displacement information can reflect the deviation of the joints, and the action information can be determined based on the deviation.

[0097] The motion information of the body joints can be determined based on the action information, the reference behavior parameters can be determined based on the motion information, and the target behavior parameters can be determined based on the reference behavior parameters.

[0098] In one possible implementation, the motion information includes head motion information and hand motion information. A possible method for determining the target behavior parameter based on the motion information includes:

[0099] B1. acquiring the driver's head movement information according to the head movement information;

[0100] B2. determining a first reference behavior parameter based on the head movement information;

[0101] B3. determining the driver's facial movement information from the head movement information;

[0102] B4. determining the driver's sight direction and facial expression based on the facial action information;

[0103] B5. determining a second reference behavior parameter based on the gaze direction and the facial expression;

[0104] B6. determining a third reference behavior parameter of the driver based on the hand motion information;

[0105] B7. Determine the target behavior parameter according to the first reference behavior parameter, the first reference behavior parameter, and the first reference behavior parameter.

[0106] The head movement information can be determined based on the head movement. For example, if the head movement is a head shaking movement, the movement information can be the movement amplitude of the head shaking, etc. Specifically, it can be the maximum head shaking amplitude when the head shakes.

[0107] Different head motion information may correspond to different reference behavior parameters, and the head motion information and the reference behavior parameters have a one-to-one correspondence.

[0108] The head movement information may also include the driver's facial movements, and the facial movement information may be directly extracted from the head movement information. Facial movements may include eye movements and facial movements.

[0109] The driver's gaze direction and facial expression can be determined based on the facial action information. The method for determining the second reference behavior parameter based on the gaze direction and facial expression can be to determine the second reference behavior parameter using a neural network model. The neural network model is a pre-trained network model used to determine the reference behavior parameter based on the gaze direction and facial expression.

[0110] Based on the hand motion information, the third reference behavior parameter can be determined by, for example, if the hand motion is one hand holding the steering wheel while the other hand is on a phone call, the parameter value of the behavior parameter is low; if the hand motion is two hands holding the steering wheel, the parameter value of the behavior parameter is high. A specific evaluation method can be based on the correlation between the hand motion and the driving behavior: a higher correlation results in a higher parameter value, and a lower correlation results in a lower parameter value.

[0111] A weighted operation can be performed on the first reference behavior parameter, the first reference behavior parameter, and the second reference behavior parameter to obtain a target behavior parameter. The behavior parameter can be represented by a behavior parameter value. A higher behavior parameter value indicates a higher correlation with vehicle driving, and a lower behavior parameter value indicates a lower correlation with vehicle driving.

[0112] In one possible implementation, a possible method for determining the target driving behavior of the driver based on the target behavior parameter includes:

[0113] C1. determining the driver's behavioral tendency based on the target behavioral parameter;

[0114] C2. determining at least one reference driving behavior based on the behavioral tendency;

[0115] C3. Obtaining the target user's driving time and driving starting location;

[0116] C4. Determine a target driving behavior from the at least one reference driving behavior based on the driving duration and the driving starting location.

[0117] The higher the parameter value of the target behavior parameter is, the more likely the behavior tendency is to be normal driving, and the lower the parameter value is, the more likely the behavior tendency is to be abnormal driving.

[0118] A driver's behavioral tendency can also be determined based on a preset mapping relationship based on the target behavioral parameter. This mapping relationship can be set based on historical data. For example, a target behavioral parameter value of 1 corresponds to one behavioral tendency, while a value of 2 corresponds to another behavioral tendency. Behavioral tendencies represent the driver's preferred behavior, such as normal driving or abnormal driving.

[0119] For example, if the behavioral tendency is abnormal driving, it can be determined whether the abnormal driving is fatigue driving based on the driving time. The longer the driving time, the lower the safety level, and the shorter the driving time, the higher the safety level.

[0120] The starting point of travel can determine the driver's driving safety level. For example, if the starting point is in the urban area, the safety level is low, and if the starting point is in the suburbs, the safety level is high.

[0121] According to the safety level, a driving behavior corresponding to the safety level of the reference driving behavior may be determined as the target driving behavior.

[0122] In one possible implementation, the present invention further provides the following method:

[0123] D1. If the first target behavior danger level is higher than the second preset danger level and lower than the first preset danger level, obtaining the location information of the target vehicle, the location information including the location information of the vehicle on the road;

[0124] D2. Determine the current lane and lane change information of the target vehicle based on the position information;

[0125] D3. Determining a behavior level correction value based on the current driving lane and the lane change information;

[0126] D4. determining a second target behavior risk level based on the first target behavior risk level and the level correction value;

[0127] D5. If the danger level of the second target behavior is higher than the first preset danger level, the warning information is issued.

[0128] The location information of the target vehicle can be obtained through the positioning device.

[0129] Driving lane and lane change information can reflect the driver's safety.

[0130] The first reference correction value may be determined according to the degree of deviation between the driving lane and the navigation lane. The greater the degree of deviation, the greater the first reference correction value, and the smaller the degree of deviation, the smaller the first reference correction value.

[0131] The lane information may include the lane change direction and number of lane changes of the driver. A higher lane change number corresponds to a larger second reference correction value, while a lower lane change number corresponds to a smaller second reference correction value. A greater deviation of the lane change direction from the normal lane change direction corresponds to a larger second reference correction value, while a smaller deviation corresponds to a smaller second reference correction value.

[0132] The behavior level correction value is determined based on the first reference correction value and the second reference correction value. Specifically, the average of the two values ​​can be determined as the behavior level correction value. Alternatively, the maximum value can be determined as the behavior level correction value.

[0133] The sum of the first target behavior risk level and the level correction value may be determined as the second target behavior risk level.

[0134] The first preset risk level may be set based on experience or historical data.

[0135] In one possible implementation, the car also has a projection device that can be used to project images, etc. The projection device can be a projector, etc., which includes a rotating device. By controlling the rotating device, the projector can be focused to adjust the clarity of the projection. A possible projection method can be:

[0136] E1. Obtaining a first projection clarity when the rotating device is adjusted from the initial position to the first position, and obtaining a second projection clarity when the rotating device is adjusted from the initial position to the second position, wherein a second rotation step length from the initial position to the second position is greater than a first rotation step length from the initial position to the first position;

[0137] E2. Obtaining a clarity deviation value between the first projection clarity and the second projection clarity;

[0138] E3. Determine the first target position according to the clarity deviation value and the initial position;

[0139] E4. Adjusting the rotating device to the first target position;

[0140] E5. Obtaining a third projection clarity of the first target position;

[0141] E6. If the third projection clarity is lower than a preset clarity threshold, obtaining a first distance value and a first angle value between the projection device and the projection screen when the rotating device is adjusted from the initial position to the first position, and obtaining a second distance value and a second angle value between the projection device and the projection screen when the rotating device is adjusted from the initial position to the second position;

[0142] E7. Determine a first distance deviation value according to the first distance value and the second distance value, and determine a first angle deviation value according to the first angle value and the second angle value;

[0143] E8. Determine a first reference correction parameter according to the first distance deviation value, and determine a second reference correction parameter according to the first angle deviation value;

[0144] E9. Determine a first position offset according to the first reference correction parameter and the second reference correction parameter;

[0145] E10. Obtain the target user's expression information;

[0146] E11. Determine location satisfaction based on the facial expression information;

[0147] E12. Determine a second position offset according to the position satisfaction level and the first position offset;

[0148] E13. Determine a second target position according to the second position offset and the first target position;

[0149] E14. Adjust the rotating device to the second target position.

[0150] The method for obtaining the first projection clarity when the rotating device is adjusted to the first position may be to obtain the first projection clarity from a server or from an instruction input by a user, and the method for obtaining the second projection clarity is the same as the method for obtaining the first projection clarity.

[0151] The first rotation step length is smaller than the second rotation step length, and the difference between the second rotation step length and the first rotation step length is smaller than a preset threshold, for example, the difference between the second rotation step length and the first rotation step length is smaller than 10 steps. Since the difference between the second rotation step length and the first rotation step length is smaller than the preset threshold, the first target position can be determined more accurately, thereby improving the accuracy of the first target position.

[0152] The method for obtaining the first projection clarity from the server may be: a user uses a mobile phone or other electronic device with a camera to capture a projection image when the rotating device is rotated to the first position; the electronic device uploads the projection image to the server, and the server processes the projection image to obtain the first projection clarity. Of course, the server may also obtain the projection image through other means, such as from a surveillance camera within a preset range of the projection device, which is not specifically limited here. The preset range is set based on empirical values ​​or historical data.

[0153] The first target position may be determined according to the deviation value between the first projection definition and the second projection definition and the initial position.

[0154] The rotating device can be adjusted from the initial position to the first target position, or from the second position to the first target position, or from the first position to the first target position, or from other positions to the first target position, which is not specifically limited here.

[0155] The first target position may be the position of the rotating device when the projection device performs automatic focusing and then performs projection. The projection device includes a rotating device.

[0156] The method for obtaining the third projection clarity of the first target position can refer to the method for obtaining the first projection clarity in the above embodiment, which will not be described in detail here. The preset clarity threshold can be set based on experience or historical data.

[0157] The first distance value may be the distance between the center of the projection lens of the projection device and the plane where the projection screen is located, and the first angle may be the angle between the plane where the mirror surface of the projection lens is located and the plane where the projection screen is located.

[0158] The first distance value can be obtained by a distance measuring device, and the first angle can be obtained by extending the plane.

[0159] The first distance deviation value may be understood as a fluctuation value of the second distance value relative to the first distance value, and the deviation value may be a positive value or a negative value.

[0160] A method for determining a first reference correction parameter based on the first distance deviation value may include: different distance deviation values ​​correspond to different reference correction parameters, and the first reference correction parameter can be determined based on the first distance deviation value. Different angle deviation values ​​correspond to different reference correction parameters, and the second reference correction parameter can be determined based on the first angle deviation value. The reference correction parameter can be used to determine the position offset.

[0161] The average of the first reference correction parameter and the second reference correction parameter can be obtained, and the first position offset can be determined based on this average. Different reference correction values ​​correspond to different position offsets. A larger reference correction value increases the first position offset, and a smaller reference correction value decreases the first position offset. Determining the first position offset by the average of the first reference correction parameter and the second reference correction parameter can improve the accuracy of the first position offset.

[0162] The method for obtaining the target user's facial expression information may be to obtain a facial image of the target user through the user's electronic device and determine the user's facial expression information based on the facial image. The target user may be a user debugging the projection device, a user viewing the projection image using the projection device, or other related users. This is merely an example.

[0163] Different facial expressions have different satisfaction levels. These expressions can include a smiling face, a crying face, or a normal expression. A normal expression is defined as a face with no facial fluctuations. The satisfaction level for a smiling face is higher than that for a normal expression, which in turn is higher than that for a crying face.

[0164] A correction value for the first position offset can be determined based on the satisfaction level. A higher satisfaction level results in a smaller correction value, while a lower satisfaction level results in a larger correction value. The second position offset is determined by multiplying the first position offset by the correction value. Alternatively, the first position offset can be corrected using the correction value to obtain the second position offset, such as the sum of the first position offset and the correction value.

[0165] The first target position can be offset by the second position offset to obtain the second target position. The offset method can be: if the second position offset is positive, the offset can be added to the first target position to obtain the second target position; if the second position offset is negative, the offset can be subtracted from the first target position to obtain the second target position. Adding or subtracting the offset to the first target position can be understood as adding or subtracting the second position offset to the position offset between the first target position and the initial position.

[0166] In this example, when the projection clarity at the first target position reaches the preset standard, the projection device is adjusted again, thereby improving the accuracy of adjusting the projection device.

[0167] In a possible implementation, a possible method for determining the first set of to-be-learned targets based on the N reference feature information may be:

[0168] F1. Determine a keyword corresponding to each reference feature information in N reference feature information to obtain N keywords;

[0169] F2. Determine the criticality of each keyword in the plurality of keywords to obtain a plurality of criticalities;

[0170] F3. Determine a first set of to-be-learned objectives based on multiple criticalities.

[0171] Keywords can be extracted from the reference feature information to obtain multiple keywords.

[0172] The first set of targets to be learned may be determined according to the average values ​​of the multiple criticalities. The average values ​​of the criticalities have a corresponding relationship with the set to be learned, and the corresponding relationship is set by empirical values.

[0173] A possible method of determining the criticality of each keyword in the multiple keywords to obtain multiple criticalities may include the following steps:

[0174] G1. Determine a target position of keyword a in the text information and a reference key level of keyword a, where keyword a is any keyword among the multiple keywords;

[0175] G2. Determine the reference criticality corresponding to the reference criticality according to a preset mapping relationship between criticality and criticality.

[0176] G3. Determine a target first optimization factor corresponding to the target position according to a mapping relationship between a preset position and a first optimization factor;

[0177] G4. Obtain the volume parameter of the keyword a;

[0178] G5. Determine a target second optimization factor corresponding to the volume parameter of the keyword a according to a preset mapping relationship between the volume parameter and the second optimization factor;

[0179] G6. Optimize the reference criticality according to the target first optimization factor and the target second optimization factor to obtain the criticality of the keyword a.

[0180] The preset threshold value may be set by the user or by the system as a default, and may be an empirical value. The intelligent robot may pre-store a mapping relationship between a preset critical level and a criticality, a mapping relationship between a preset position and a first optimization factor, and a mapping relationship between a preset volume parameter and a second optimization factor.

[0181] In the specific implementation, taking keyword a as an example, keyword a is any keyword among multiple keywords. The intelligent robot can determine the target position of keyword a in the text information and the reference key level of keyword a, and determine the reference key level corresponding to the reference key level according to the mapping relationship between the preset key level and the key degree. According to the mapping relationship between the preset position and the first optimization factor, the target first optimization factor corresponding to the target position is determined. The value range of the first optimization factor can be -1 to 1. For example, the first optimization factor can be -0.08 to 0.08.

[0182] Furthermore, since the volume parameters of the keywords are different, it means that the users attach different importance to them. The volume parameter can be volume or pitch. According to the preset mapping relationship between the volume parameter and the second optimization factor, the target second optimization factor corresponding to the volume parameter of keyword a is determined. The value range of the second optimization factor can be -1 to 1. For example, the second optimization factor can be -0.032 to 0.032. The reference criticality is adjusted according to the target first optimization factor and the target second optimization factor to obtain the criticality of keyword a. The specific calculation formula is as follows:

[0183] Keyword a's criticality = Keyword a's reference criticality * (1 + target first optimization factor) * (1 + target second optimization factor)

[0184] Furthermore, the criticality of the keyword can be accurately determined based on the keyword's position and volume parameters, which helps to improve the accuracy of command recognition.

[0185] For the same example as above, please refer to Figure 3 , Figure 3A schematic structural diagram of a terminal provided in an embodiment of the present application, as shown in the figure, includes a processor, an input device, an output device, and a memory, the processor, the input device, the output device, and the memory being interconnected, wherein the memory is used to store a computer program, the computer program including program instructions, the processor being configured to call the program instructions, and the program including instructions for executing the following steps;

[0186] Acquire a target image when the driver is driving a target vehicle;

[0187] Obtaining target behavior parameters of the driver through the target image;

[0188] determining a target driving behavior of the driver according to the target behavior parameter;

[0189] If the target driving behavior is a preset abnormal driving behavior, determining a first target behavior risk level of the driver;

[0190] If the risk level of the first target behavior is higher than a first preset risk level, an alarm message is issued.

[0191] The above mainly introduces the scheme of the embodiment of the present application from the perspective of the execution process on the method side. It is understandable that, in order to implement the above functions, the terminal includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of the various examples described in the embodiments provided herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in a hardware or computer software driven hardware manner depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0192] The embodiment of the present application can divide the terminal into functional units according to the above method example. For example, each functional unit can be divided according to each function, or two or more functions can be integrated into one processing unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. It should be noted that the division of units in the embodiment of the present application is schematic and is only a logical functional division. In actual implementation, there may be other division methods.

[0193] In line with the above, please see Figure 4 , Figure 4 The present invention provides a schematic diagram of a driver's behavior recognition device. Figure 4 As shown, the device includes:

[0194] The first acquisition unit 401 is used to acquire a target image when the driver is driving a target vehicle;

[0195] A second acquiring unit 402 is configured to acquire a target behavior parameter of the driver through the target image;

[0196] A first determining unit 403 is configured to determine a target driving behavior of the driver according to the target behavior parameter;

[0197] A second determining unit 404 is configured to determine a first target behavior risk level of the driver if the target driving behavior is a preset abnormal driving behavior;

[0198] The sending unit 405 is configured to send an alarm message if the danger level of the first target behavior is higher than a first preset danger level.

[0199] In a possible implementation, the second acquiring unit 402 is configured to:

[0200] Performing feature extraction on the target image to obtain feature data;

[0201] determining the driver's action information based on the characteristic data;

[0202] The target behavior parameter is determined according to the action information.

[0203] In one possible implementation, the motion information includes head motion information and hand motion information. In determining the target behavior parameter based on the motion information, the second acquiring unit 402 is configured to:

[0204] acquiring the driver's head movement information according to the head movement information;

[0205] determining a first reference behavior parameter according to the head movement information;

[0206] determining facial movement information of the driver from the head movement information;

[0207] determining the driver's sight direction and facial expression based on the facial action information;

[0208] determining a second reference behavior parameter according to the gaze direction and the facial expression;

[0209] determining a third reference behavior parameter of the driver according to the hand motion information;

[0210] The target behavior parameter is determined according to the first reference behavior parameter, the second reference behavior parameter, and the first reference behavior parameter.

[0211] In a possible implementation, the first determining unit 403:

[0212] determining the driver's behavioral tendency according to the target behavioral parameter;

[0213] determining at least one reference driving behavior based on the behavior tendency;

[0214] Obtaining the target user's driving time and driving starting location;

[0215] A target driving behavior is determined from the at least one reference driving behavior according to the driving duration and the driving starting location.

[0216] In one possible implementation, the device is further configured to:

[0217] If the first target behavior danger level is higher than the second preset danger level and lower than the first preset danger level, obtaining the location information of the target vehicle, the location information including the location information of the vehicle on the road;

[0218] Determining the current driving lane and lane change information of the target vehicle based on the position information;

[0219] determining a behavior level correction value according to the current driving lane and the lane change information;

[0220] determining a second target behavior risk level according to the first target behavior risk level and the level correction value;

[0221] If the second target behavior danger level is higher than the first preset danger level, the warning information is issued.

[0222] An embodiment of the present application also provides a computer storage medium, wherein the computer storage medium stores a computer program for electronic data exchange, and the computer program enables a computer to execute part or all of the steps of any driver behavior recognition method recorded in the above method embodiments.

[0223] An embodiment of the present application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program enables a computer to execute some or all steps of any driver behavior recognition method recorded in the above method embodiments.

[0224] It should be noted that for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application.

[0225] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0226] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0227] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0228] In addition, the functional units in the various embodiments of the application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software program modules.

[0229] If the integrated unit is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory, including a number of instructions for enabling a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned memory includes: various media that can store program codes, such as a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0230] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable memory, which may include: a flash drive, a read-only memory, a random access memory, a magnetic disk or an optical disk, etc.

[0231] The above is a detailed introduction to the embodiments of the present application. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, for those skilled in the art, according to the idea of ​​the present application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A driver behavior recognition method, characterized in that: The method comprises: Acquire a target image when the driver is driving a target vehicle; Obtaining target behavior parameters of the driver through the target image; determining a target driving behavior of the driver according to the target behavior parameter; If the target driving behavior is a preset abnormal driving behavior, determining a first target behavior risk level of the driver; If the risk level of the first target behavior is higher than the first preset risk level, an alarm message is issued; The target vehicle further comprises a projection device, and the method further comprises: obtaining a first projection clarity when the rotating device is adjusted from an initial position to a first position, and obtaining a second projection clarity when the rotating device is adjusted from the initial position to a second position, wherein a second rotation step length from the initial position to the second position is greater than a first rotation step length from the initial position to the first position; Obtaining a clarity deviation value between the first projection clarity and the second projection clarity; determining a first target position according to the clarity deviation value and the initial position; adjusting the rotating device to the first target position; Acquiring a third projection clarity of the first target position; If the third projection clarity is lower than a preset clarity threshold, obtaining a first distance value and a first angle value between the projection device and the projection screen when the rotating device is adjusted from the initial position to the first position, and obtaining a second distance value and a second angle value between the projection device and the projection screen when the rotating device is adjusted from the initial position to the second position; determining a first distance deviation value according to the first distance value and the second distance value, and determining a first angle deviation value according to the first angle value and the second angle value; determining a first reference correction parameter according to the first distance deviation value, and determining a second reference correction parameter according to the first angle deviation value; determining a first position offset according to the first reference correction parameter and the second reference correction parameter; Get the target user's expression information; determining location satisfaction based on the facial expression information; determining a second position offset according to the position satisfaction level and the first position offset; determining a second target position according to the second position offset and the first target position; adjusting the rotating device to the second target position; Determining the target driving behavior of the driver according to the target behavior parameter includes: determining the driver's behavioral tendency according to the target behavioral parameter; determining at least one reference driving behavior based on the behavior tendency; Obtaining the target user's driving time and driving starting location; A target driving behavior is determined from the at least one reference driving behavior according to the driving duration and the driving starting location.

2. The method according to claim 1, characterized in that The step of obtaining the target behavior parameter of the driver through the target image includes: Performing feature extraction on the target image to obtain feature data; determining the driver's action information based on the characteristic data; The target behavior parameter is determined according to the action information.

3. The method according to claim 2, characterized in that The motion information includes head motion information and hand motion information, and determining the target behavior parameter according to the motion information includes: acquiring the driver's head movement information according to the head movement information; determining a first reference behavior parameter according to the head movement information; determining facial movement information of the driver from the head movement information; determining the driver's sight direction and facial expression based on the facial action information; determining a second reference behavior parameter according to the gaze direction and the facial expression; determining a third reference behavior parameter of the driver according to the hand motion information; The target behavior parameter is determined according to the first reference behavior parameter, the second reference behavior parameter, and the first reference behavior parameter.

4. The method according to any one of claims 1 to 3, characterized in that The method further comprises: If the first target behavior danger level is higher than the second preset danger level and lower than the first preset danger level, obtaining the location information of the target vehicle, the location information including the location information of the vehicle on the road; Determining the current driving lane and lane change information of the target vehicle based on the position information; determining a behavior level correction value according to the current driving lane and the lane change information; determining a second target behavior risk level according to the first target behavior risk level and the level correction value; If the second target behavior danger level is higher than the first preset danger level, the warning information is issued.

5. A driver behavior recognition device, characterized in that: The device comprises: a first acquiring unit, configured to acquire a target image of a driver driving a target vehicle; a second acquiring unit, configured to acquire a target behavior parameter of the driver through the target image; a first determining unit, configured to determine a target driving behavior of the driver according to the target behavior parameter; a second determining unit, configured to determine a first target behavior risk level of the driver if the target driving behavior is a preset abnormal driving behavior; a sending unit, configured to send an alarm message if the danger level of the first target behavior is higher than a first preset danger level; The target vehicle also has a projection device, which is further used to: obtaining a first projection clarity when the rotating device is adjusted from an initial position to a first position, and obtaining a second projection clarity when the rotating device is adjusted from the initial position to a second position, wherein a second rotation step length from the initial position to the second position is greater than a first rotation step length from the initial position to the first position; Obtaining a clarity deviation value between the first projection clarity and the second projection clarity; determining a first target position according to the clarity deviation value and the initial position; adjusting the rotating device to the first target position; Acquiring a third projection clarity of the first target position; If the third projection clarity is lower than a preset clarity threshold, obtaining a first distance value and a first angle value between the projection device and the projection screen when the rotating device is adjusted from the initial position to the first position, and obtaining a second distance value and a second angle value between the projection device and the projection screen when the rotating device is adjusted from the initial position to the second position; determining a first distance deviation value according to the first distance value and the second distance value, and determining a first angle deviation value according to the first angle value and the second angle value; determining a first reference correction parameter according to the first distance deviation value, and determining a second reference correction parameter according to the first angle deviation value; determining a first position offset according to the first reference correction parameter and the second reference correction parameter; Get the target user's expression information; determining location satisfaction based on the facial expression information; determining a second position offset according to the position satisfaction level and the first position offset; determining a second target position according to the second position offset and the first target position; adjusting the rotating device to the second target position; The first determining unit is specifically configured to: determining the driver's behavioral tendency according to the target behavioral parameter; determining at least one reference driving behavior based on the behavior tendency; Obtaining the target user's driving time and driving starting location; A target driving behavior is determined from the at least one reference driving behavior according to the driving duration and the driving starting location.

6. The device according to claim 5, characterized in that The second acquiring unit: Performing feature extraction on the target image to obtain feature data; determining the driver's action information based on the characteristic data; The target behavior parameter is determined according to the action information.

7. The device according to claim 6, characterized in that The motion information includes head motion information and hand motion information. In determining the target behavior parameter based on the motion information, the second acquiring unit is configured to: acquiring the driver's head movement information according to the head movement information; determining a first reference behavior parameter according to the head movement information; determining facial movement information of the driver from the head movement information; determining the driver's sight direction and facial expression based on the facial action information; determining a second reference behavior parameter according to the gaze direction and the facial expression; determining a third reference behavior parameter of the driver according to the hand motion information; The target behavior parameter is determined according to the first reference behavior parameter, the second reference behavior parameter, and the first reference behavior parameter.

8. A terminal, characterized in that: The method comprises a processor, an input device, an output device and a memory, wherein the processor, the input device, the output device and the memory are interconnected, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute the method according to any one of claims 1 to 4.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, wherein the computer program includes program instructions. When the program instructions are executed by a processor, the processor is caused to perform the method according to any one of claims 1 to 4.

Images