KR20220032074A - Driver behavior detection method, apparatus, electronic device, storage medium and program - Google Patents

Abstract

Embodiments of the present application provide a method, an apparatus, an electronic device, a computer storage medium, and a computer program for detecting a driver's behavior. The method comprises: acquiring an image to be detected of a driving position area within the cabin; performing detection on the to-be-detected image to obtain a target detection result including a steering wheel detection result and a hand detection result; determining a driving behavior category of the driver according to the target detection result; and sending a warning message when the driver's driving behavior category is dangerous driving.

Description

Driver behavior detection method, apparatus, electronic device, storage medium and program

[Cross Citation of Related Applications]

This application is filed on the basis of the Chinese patent application filed on August 7, 2020 and with an application number of 202010790208.3, claiming the priority of the Chinese patent application, the entire content of the Chinese patent application is incorporated herein by reference. .

[Technical field]

This application relates to the field of deep learning technology, and more specifically, to a method, an apparatus, an electronic device, a computer storage medium, and a computer program for detecting a driver's behavior.

With the rapid development of the vehicle, the vehicle has become an important transportation tool when users go out, and therefore, safe driving of the vehicle has become one of the important tasks of the current automobile industry. Safe driving of a vehicle is determined by various factors such as a driver's driving behavior, road conditions, and weather conditions.

In general, dangerous driving behavior is one of the main factors that cause most traffic accidents. Accordingly, it is possible to detect the driving behavior of the driver to improve driving safety and ensure the safety of passengers and drivers.

Embodiments of the present application provide at least one method, apparatus, electronic device, computer storage medium, and computer program for detecting driver behavior.

An embodiment of the present application provides a method for detecting a driver's behavior, the method comprising:

acquiring an image to be detected of a driving position area within the cabin;

performing detection on the to-be-detected image to obtain a target detection result including a steering wheel detection result and a hand detection result;

determining a driving behavior category of the driver according to the target detection result; and

and sending a warning message when the driving behavior category of the driver is dangerous driving.

By using the above method, detection is performed on the image to be detected of the obtained driving position area to obtain a target detection result including a steering wheel detection result and a hand detection result. In addition, the driver's driving behavior category is determined based on the target detection result, and when the driver's driving behavior category is dangerous driving, a warning message is sent to realize the detection of the driver's driving behavior, which is convenient to give the driver safety attention. and improve vehicle driving safety.

In some embodiments of the present application, when the steering wheel detection result includes a steering wheel and the hand detection result includes a hand, determining the driving behavior category of the driver according to the target detection result includes:

determining a positional relationship between the steering wheel and the hand according to the steering wheel detection result and the hand detection result; and

and determining a driving behavior category of the driver according to the positional relationship.

In some embodiments of the present application, the step of determining the driving behavior category of the driver according to the location relationship includes:

and when the positional relationship indicates that the driver's hand is holding the steering wheel, determining that the driver's driving behavior category is safe driving.

In some embodiments of the present application, the step of determining the driving behavior category of the driver according to the location relationship includes:

and determining that the driving behavior category of the driver is dangerous driving when the positional relationship indicates that both hands of the driver have left the steering wheel.

In some embodiments of the present application, the determining of the positional relationship between the steering wheel and the hand according to the steering wheel detection result and the hand detection result includes:

When one hand is included in the hand detection result, if there is a region where a detection box corresponding to a hand in the hand detection result overlaps a detection box corresponding to the steering wheel in the steering wheel detection result, the steering wheel and the hand The positional relationship between the two determines that the driver's hand is holding the steering wheel, and if a detection box corresponding to the hand in the hand detection result overlaps with a detection box corresponding to the steering wheel in the steering wheel detection result does not exist. otherwise, determining that the positional relationship between the steering wheel and the hand is that the driver's hands are off the steering wheel.

In some embodiments of the present application, the determining of the positional relationship between the steering wheel and the hand according to the steering wheel detection result and the hand detection result includes:

When both hands are included in the hand detection result, if a detection box corresponding to both hands in the hand detection result does not overlap with a detection box corresponding to the steering wheel in the steering wheel detection result, the steering wheel The positional relationship between the hand and the hand determines that both hands of the driver leave the steering wheel, and if a detection box corresponding to at least one hand among the hand detection results is a detection box corresponding to the steering wheel among the steering wheel detection results, and determining that the positional relationship between the steering wheel and the hand is that the driver's hand is holding the steering wheel if an overlapping area exists.

In some embodiments of the present application, the determining of the positional relationship between the steering wheel and the hand according to the steering wheel detection result and the hand detection result includes:

generating an intermediate feature map corresponding to the to-be-detected image based on the to-be-detected image;

performing convolution processing on the intermediate feature map at least once to generate a classification feature map of two channels corresponding to the intermediate feature map; Corresponds to category -;

extracting two feature values of a feature location matching the center point location information from the classification feature map based on center point location information indicated by detection box information corresponding to a hand among the hand detection results; selecting a maximum feature value from among two feature values, and determining a category of a channel feature map corresponding to the maximum feature value among the classification feature maps as a category corresponding to the center point location information; and

and determining a positional relationship between the steering wheel and the hand based on a category corresponding to each center point position information indicated by the detection box information corresponding to the hand.

In the above embodiment, after determining the steering wheel detection result and performing convolution processing on the intermediate feature map at least once to generate a classification feature map, the generated center point position information of the driver's hand is combined, and the steering wheel and the hand The positional relationship between them can be determined more accurately.

In some embodiments of the present application, the determining of the positional relationship between the steering wheel and the hand based on a category corresponding to each center point position information indicated by the detection box information corresponding to the hand comprises:

determining a category corresponding to the center point position information as a positional relationship between the steering wheel and the hand when the detection box information corresponding to the hand includes one center point position information; and

When the detection box information corresponding to the hand includes two center point location information, and the category corresponding to the two center point location information is a category in which the driver's hand leaves the steering wheel, the position between the steering wheel and the hand The relationship determines that both hands of the driver leave the steering wheel, and when the category corresponding to at least one center point location information among the categories corresponding to the two center point location information is a category in which the driver's hands are holding the steering wheel, determining that the positional relationship between the steering wheel and the hand is that the driver's hand is holding the steering wheel.

In some embodiments of the present application, when the steering wheel is included in the steering wheel detection result and the hand is not included in the hand detection result, determining the driving behavior category of the driver according to the target detection result includes:

and determining that the driving behavior category of the driver is dangerous driving according to the target detection result.

In some embodiments of the present application, the step of obtaining a target detection result by performing detection on the image to be detected includes:

generating an intermediate feature map corresponding to the to-be-detected image based on the to-be-detected image;

performing a target convolution process on the intermediate feature map at least once to generate a detection feature map of a plurality of channels corresponding to the intermediate feature map;

generating a target channel feature map after transformation by performing a feature value conversion process on each feature value of a target channel feature map indicating a location among the detection feature maps of a plurality of channels using an activation function;

Obtaining a plurality of pooling values and a position index corresponding to each pooling value among the plurality of pooling values by performing maximum pooling processing on the target channel feature map after transformation according to a preset pooling size and pooling stride - The position index is the used to identify the location of the pooling value in the target channel feature map after transformation;

generating target detection box information based on a plurality of pooling values and a location index corresponding to each pooling value among the plurality of pooling values; and

and determining the target detection result according to the target detection box information.

In the above embodiment, by performing maximum pulling processing on the target channel feature map to obtain a plurality of pooling values and a position index corresponding to each pooling value among the plurality of pooling values, and by generating target detection box information, the target detection result is determined Provides data support for creation.

In some embodiments of the present application, generating the target detection box information based on a plurality of pooling values and a location index corresponding to each pooling value among the plurality of pooling values includes:

When at least one pooling value among the plurality of pooling values is greater than a set pooling threshold, a target pooling value of the center point of the target detection box is determined from among the plurality of pooling values based on the plurality of pooling values and a pooling threshold to do; and

and generating the target detection box information based on a location index corresponding to the target pooling value.

In the above embodiment, a pulling value larger than a pulling threshold value among a plurality of pulling values is determined as a target pulling value belonging to a center point of a target detection box of a steering wheel or a driver's hand, and steering is based on a position index corresponding to the target pulling value At least one target detection box information of a wheel or a driver's hand is more accurately generated.

In some embodiments of the present application, generating the target detection box information based on a plurality of pooling values and a location index corresponding to each pooling value among the plurality of pooling values includes:

and determining that the target detection box information is empty when the plurality of pooling values are less than or equal to a set pooling threshold.

Hereinafter, for descriptions of the effects of the apparatus, the electronic device, and the like, reference is made to the description of the method, and a redundant description will be omitted herein.

An embodiment of the present application provides an apparatus for detecting a driver's behavior, wherein the apparatus for detecting a driver's behavior includes:

an acquiring module, configured to acquire an image to be detected of a driving position area within the cabin;

a detection module, configured to perform detection on the to-be-detected image to obtain a target detection result including a steering wheel detection result and a hand detection result;

a determining module, configured to determine a driving behavior category of the driver according to the target detection result; and

and a warning module, configured to send a warning message when the driving behavior category of the driver is dangerous driving.

An embodiment of the present application provides an electronic device including a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is executed, the processor operates through the bus In communication with the memory, when the machine-readable instruction is executed by the processor, the method for detecting driver behavior according to any one of the above embodiments is executed.

An embodiment of the present application provides a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the method for detecting driver behavior according to any one of the above embodiments run

An embodiment of the present application further provides a computer program including a computer readable code, wherein when the computer readable code is executed in an electronic device, a processor in the electronic device is configured to detect a driver's behavior according to any one of the above embodiments used to run

In order to make it easier to understand the above objects, features and advantages of the present application, preferred embodiments will be described in detail below in conjunction with the accompanying drawings.

In order to more clearly explain the technical solutions of the embodiments of the present application, the drawings used in the embodiments will be briefly introduced below. BRIEF DESCRIPTION OF THE DRAWINGS The drawings herein are incorporated in and constitute a part of this specification, and the drawings illustrate technical solutions suitable for embodiments of the present application and used together with the specification to describe the present application. It should be understood that the following drawings illustrate only some embodiments of the present application, and should not be regarded as limitations on the scope, and those skilled in the art may obtain other related drawings according to these drawings without creative efforts.
1A shows a schematic diagram of one application scenario according to an embodiment of the present application;
1B shows a schematic flowchart of a driver behavior detection method provided in an embodiment of the present application;
2 is a schematic flowchart of a specific method of obtaining a target detection result by performing detection on an image to be detected among the driver behavior detection methods provided by the embodiments of the present application;
3 shows a structural schematic diagram of a driver behavior detection device provided in an embodiment of the present application;
4 shows a structural schematic diagram of an electronic device provided in an embodiment of the present application.

In order to more clearly understand the purpose, technical solution and advantages of the embodiments of the present application, the technical solutions according to the embodiments of the present application will be clearly and completely described below in conjunction with the drawings of the embodiments of the present application. The described embodiments are only some embodiments of the present application, not all embodiments. In general, the components of the embodiment of the present application described and illustrated in the drawings may be arranged and designed in various different configurations. Accordingly, the detailed description of the embodiments of the present application provided in the following drawings is only for showing selected embodiments of the present application, and is not intended to limit the scope of the present application to be protected. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative efforts also fall within the protection scope of the present application.

Dangerous driving behavior is one of the main factors that cause most traffic accidents. Accordingly, it is possible to detect the driving behavior of the driver to improve driving safety and ensure the safety of passengers and drivers. Accordingly, in order to solve the above problems, an embodiment of the present application provides a method for detecting a driver's behavior.

For convenient understanding of the embodiments of the present application, first, the driver behavior detection method disclosed in the embodiments of the present application will be described in detail.

In some embodiments of the present application, the driver behavior detection method may be executed by a driver behavior detection apparatus, and the driver behavior detection apparatus includes a user equipment (UE), a mobile device, a user terminal, a terminal, a cellular phone, a wireless It may be a telephone, personal digital assistant (PDA), handheld device, computing device, in-vehicle device, wearable device, or the like. The method may be implemented by a processor invoking computer readable instructions stored in a memory. Alternatively, a server may perform the method.

Hereinafter, the present application will be further described by combining one application scenario.

The method for detecting a driver's behavior according to an embodiment of the present application may be applied to application scenarios such as driver driving. 1A is a schematic diagram of one application scenario according to an embodiment of the present application. As shown in FIG. 1A , an image to be detected 11 of the driving position area may be acquired through the driver behavior detection device 10 , and the image 11 to be detected is a steering wheel 110 and a hand 111 . includes In the driver behavior detecting apparatus 10 , the detection results of the steering wheel 110 and the hand 111 may be obtained by performing the driver behavior detection described in the above-described embodiment. Furthermore, the driver's driving behavior category can be determined according to the detection result of the steering wheel 110 and the hand 111, and when the driver's driving behavior category is dangerous driving, the driver behavior detecting device 10 sends a warning message to By implementing the detection of the driver's driving behavior, it is convenient to give the driver safety attention and the safety of vehicle driving is improved.

1B shows a schematic flowchart of a method for detecting driver behavior provided by an embodiment of the present application, the method including S101 to S104 as follows:

S101: Acquire an image to be detected of the driving position area in the cabin.

S102: Perform detection on the image to be detected, to obtain a target detection result including a steering wheel detection result and a hand detection result.

S103: A driver's driving behavior category is determined according to the target detection result.

S104: When the driving behavior category of the driver is dangerous driving, a warning message is sent.

In the method, detection is performed on the to-be-detected image of the obtained driving position area to obtain a target detection result including a steering wheel detection result and a hand detection result. In addition, the driver's driving behavior category is determined based on the target detection result, and when the driver's driving behavior category is dangerous driving, a warning message is sent to realize the detection of the driver's driving behavior, which is convenient to give the driver safety attention. and improve vehicle driving safety.

About S101:

Here, a photographing device may be installed in the cabin, and an image to be detected of the driving position area may be acquired in real time through the photographing device installed in the cabin. Here, the mounting position of the photographing device may be a position capable of photographing the steering wheel and the driver's seat area within the driving position area.

For S102 and S103:

Here, the target detection result can be obtained by inputting the image to be detected into the trained neural network, and performing detection on each image to be detected. Here, the target detection result includes a steering wheel detection result and a hand detection result. The steering wheel detection result includes information on whether or not there is a steering wheel in the image to be detected, and when there is a steering wheel, the steering wheel detection result includes detection box information of the steering wheel. The hand detection result includes information on whether or not there is a hand detection box in the image to be detected, and if there is a hand, the hand detection result includes hand detection box information.

In one selectable embodiment, referring to FIG. 2 , performing detection on an image to be detected to obtain a target detection result may include:

S201: Generate an intermediate feature map corresponding to the image to be detected based on the image to be detected.

S202: Perform target convolution processing on the intermediate feature map at least once to generate a detection feature map of a plurality of channels corresponding to the intermediate feature map.

S203: Perform a feature value conversion process on each feature value of a target channel feature map indicating a location among detection feature maps of a plurality of channels by using an activation function to generate a target channel feature map after transformation.

S204: Perform maximum pooling processing on the target channel feature map after transformation according to the preset pooling size and pooling stride to obtain a plurality of pooling values and a position index corresponding to each pooling value among the plurality of pooling values. The position index is used to identify the position of the pooling value in the target channel feature map after transformation.

S205: Generate target detection box information based on a plurality of pooling values and a location index corresponding to each pooling value among the plurality of pooling values.

S206: Determine a target detection result according to target detection box information.

In the above embodiment, a plurality of pooling values and a position index corresponding to each pooling value among the plurality of pooling values are obtained by performing maximum pooling processing on the target channel feature map, and target detection box information is generated, thereby generating a target detection result. It provides data support for

An image to be detected is input to a trained neural network, and the backbone network in the trained neural network performs convolution processing on the image to be detected several times to generate an intermediate feature map corresponding to the image to be detected. Here, the structure of the backbone network among the neural networks can be set according to actual demand.

Here, the intermediate feature map may be input to the steering wheel detection branch network and the hand detection branch network of the neural network, respectively, to generate a steering wheel detection result and a hand detection result. Hereinafter, the generation of the steering wheel detection result will be described in detail.

Here, first, the first convolution process (ie, target convolution process) is performed on the intermediate feature map at least once to generate a detection feature map of a plurality of channels corresponding to the steering wheel, and the detected feature map is The number of corresponding channels may be three. Here, the detection characteristic map includes a first channel characteristic map indicating a location (the first channel characteristic map is a target channel characteristic map), a second channel characteristic map indicating detection box length information, and a third channel indicating detection box width information. Includes feature maps.

In addition, a feature value conversion process is performed on a target channel feature map indicating a position among detection feature maps of a plurality of channels by using an activation function to generate a target channel feature map after transformation. Each feature value in the target channel feature map after transformation is a numerical value between 0 and 1. Here, the activation function may be a sigmoid function. With respect to the feature value of any one feature point in the target channel feature map after transformation, the closer the feature value is to 1, the higher the probability that the feature point corresponding to the feature value belongs to the center point of the detection box of the steering wheel.

Next, maximum pooling processing is performed on the target channel feature map after transformation according to the preset pooling size and pooling stride, to obtain a pooling value corresponding to each feature position in the target channel feature map and a position index corresponding to each pooling value . The position index may be used to identify the position of the pooling value in the target channel feature map after transformation. Furthermore, a plurality of pooling values corresponding to the target channel feature map and a location index corresponding to each pooling value among the plurality of pooling values may be obtained by merging the same location index among location indexes corresponding to each feature location. Here, the preset size and the pulling stride may be set according to actual demand, for example, the preset pooling size may be 3×3, and the preset pulling stride may be 1.

Furthermore, first detection box information (ie, target detection box information) corresponding to the steering wheel may be generated based on a plurality of pulling values and a position index corresponding to each of the plurality of pulling values.

In some embodiments of the present application, a maximum pooling process with a size of 3×3 and a stride of 1 may be performed on the target channel feature map. At the time of pooling, with respect to the feature value of each 3×3 feature point in the target channel feature map, the maximum response value of the 3×3 feature point (ie, the pooling value) and the position index of the maximum response value on the target channel feature map decide In this case, the number of maximum response values is related to the size of the target channel feature map. For example, if the size of the target channel feature map is 80×60×3, the maximum response value obtained after performing the maximum pooling process on the target channel feature map is 80×60 in total. Also, for each maximum response value, there may be at least one other maximum response value equal to its location index. Then, the maximum response values having the same location index are merged to obtain a location index corresponding to each of the M maximum response values and the M maximum response values. Finally, the first detection box information corresponding to the steering wheel is generated based on the M maximum response values (pooling values) and the position index corresponding to each maximum response value.

Here, the determination process of the second detection box information corresponding to the hand may refer to the determination process of the first detection box information corresponding to the steering wheel, and a redundant description will be omitted herein.

After obtaining the first detection box information corresponding to the steering wheel, the first detection box information may be determined as the steering wheel detection result. When the first detection box information corresponding to the steering wheel is not obtained, it is determined that the steering wheel detection result does not include the steering wheel. In addition, after obtaining the second detection box information corresponding to the hand, the second detection box information may be determined as the hand detection result. When the second detection box information corresponding to the hand is not obtained, the hand detection result is determined as not including the hand.

In one selectable embodiment, generating target detection box information based on a plurality of pooling values and a position index corresponding to each pooling value among the plurality of pooling values may include:

Step A1: when at least one pooling value among the plurality of pooling values is greater than the set pooling threshold, a target pooling value of the center point of the target detection box is determined from among the plurality of pooling values based on the plurality of pooling values and the pooling threshold .

Step A2: Generate target detection box information based on the position index corresponding to the target pooling value.

Continuing, the steering wheel will be described as an example. Here you can set the pooling threshold. When at least one pooling value among the plurality of pooling values is greater than the set pooling threshold, a plurality of pooling values are selected based on the set pooling threshold to obtain one target pooling value greater than the pooling threshold among the plurality of pooling values. . When each of the plurality of pulling values is less than or equal to the set pulling threshold, the target pulling value does not exist, that is, the first detection box information of the steering wheel does not exist.

Furthermore, center point location information of the first detection box corresponding to the steering wheel may be generated based on the location index corresponding to the target pooling value. Here, the pulling threshold corresponding to the steering wheel and the pulling threshold corresponding to the driver's hand may be the same or different. Specifically, the pulling threshold corresponding to the steering wheel and the pulling threshold corresponding to the driver's hand may be determined according to actual cases. For example, a multi-frame sample image collected by a photographing device corresponding to an image to be detected is obtained, and an adaptive algorithm is used according to the collected multi-frame sample image to correspond to a pulling threshold corresponding to the steering wheel and a driver's hand Each pooling threshold can be created.

Continuing with the example above, after obtaining the position index corresponding to each maximum response value among the M maximum response values and the M maximum response values, each maximum response value among the M maximum response values can be compared with the pooling threshold. there is; When a certain maximum response value is greater than the pooling threshold, the maximum response value is determined as a target pooling value. The position index corresponding to the target pulling value is the center point position information of the first detection box of the steering wheel.

Here, the maximum pulling process may be directly performed on the target channel feature map before conversion to obtain center point position information of the first detection box of the steering wheel.

In some embodiments of the present application, after obtaining the center point position information of the first detection box of the steering wheel, a second feature value of a feature location matching the center point location information in a second channel feature map based on the center point location information selects, determines the selected second feature value as a length corresponding to the first detection box of the steering wheel, and selects a third feature value of a feature location matching the center point location information from a third channel feature map, Size information of the first detection box of the steering wheel may be obtained by determining the selected third feature value as a width corresponding to the first detection box of the steering wheel.

Here, information on one or two second detection boxes may be obtained for the driver's hand. That is, the second detection box information corresponding to the left hand and/or the right hand, respectively, may be obtained. Specifically, the process of determining the second detection box information corresponding to the driver's hand may refer to the process of determining the first detection box information of the steering wheel, and a redundant description will be omitted herein.

In the above embodiment, a pulling value larger than a pulling threshold value among a plurality of pulling values is determined as a target pulling value belonging to a center point of a target detection box of a steering wheel or a driver's hand, and steering is based on a position index corresponding to the target pulling value At least one target detection box information of a wheel or a driver's hand is more accurately generated.

In one selectable embodiment, the generating of target detection box information based on a plurality of pooling values and a position index corresponding to each of the plurality of pooling values comprises: If equal, determining that the target detection box information is empty.

Here, when the plurality of pulling values corresponding to the steering wheel are less than or equal to the set pulling threshold value, it is determined that the first detection box information of the steering wheel is empty, and at least one of the plurality of pulling values corresponding to the steering wheel is selected. When the pulling value is greater than the set pulling threshold, it is determined that the first detection box information of the steering wheel is not empty.

After the steering wheel detection result and the hand detection result are obtained, the driver's driving behavior category may be determined based on the steering wheel detection result and the hand detection result.

In some embodiments of the present application, when the steering wheel is included in the steering wheel detection result and the hand is included in the hand detection result, determining the driving behavior category of the driver according to the target detection result includes:

determining a positional relationship between the steering wheel and the hand according to the steering wheel detection result and the hand detection result; and

The method may include determining a driving behavior category of the driver according to the location relationship.

Here, first, the positional relationship between the steering wheel and the hand is determined according to the steering wheel detection result and the hand detection result, and the driver's driving behavior category is determined according to the determined positional relationship. That is, the driver decides whether he is driving safely or driving dangerously.

In some embodiments of the present application, the determining of the driver's driving behavior category according to the positional relationship includes: when the positional relationship indicates that the driver's hand is holding the steering wheel, the driver's driving behavior category is safe driving It may include the step of determining that

Here, when detecting that the positional relationship indicates that the driver's hand is holding the steering wheel, it is determined that the driver's behavior category is safe driving. The case in which the driver's hand is holding the steering wheel includes a case in which the driver holds the steering wheel with the left hand, the case where the driver holds the steering wheel with the right hand, or the case where the driver holds the steering wheel with both hands.

In some embodiments of the present application, the determining of the driving behavior category of the driver according to the positional relationship may include determining that the driver's driving behavior category is dangerous driving when the positional relationship indicates that both hands of the driver have left the steering wheel. may include steps.

Here, if the positional relationship determines that the driver's both hands have left the steering wheel, it is determined that the driver's driving behavior category is dangerous driving.

In some embodiments of the present application, when the steering wheel is included in the steering wheel detection result and the hand is not included in the hand detection result, determining the driving behavior category of the driver according to the target detection result includes: according to the target detection result determining that the driving behavior category of the driver is dangerous driving.

Here, if the steering wheel is detected from the steering wheel detection result and the hand is not detected from the hand detection result, it indicates that both hands of the driver have left the steering wheel, and it is determined that the driver's driving behavior category is dangerous driving.

In some embodiments of the present application, if the steering wheel is not detected in the steering wheel detection result, an image to be detected is determined as an abnormal image, and the driving behavior category of the driver is determined as an abnormal state.

In some embodiments of the present application, the step of determining the positional relationship between the steering wheel and the hand according to the steering wheel detection result and the hand detection result may include: If a region in which the corresponding detection box overlaps the detection box corresponding to the steering wheel exists among the steering wheel detection results, the positional relationship between the steering wheel and the hand is determined as that the driver's hand is holding the steering wheel, and if the hand detection result Determining the positional relationship between the steering wheel and the hand that the driver's both hands are out of the steering wheel if there is no area in which the detection box corresponding to the middle hand overlaps the detection box corresponding to the steering wheel among the steering wheel detection results may include

In some embodiments of the present application, the step of determining the positional relationship between the steering wheel and the hand according to the steering wheel detection result and the hand detection result may include, if both hands are included in the hand detection result, both hands among the hand detection results If there is no region in which the detection box corresponding to the detection box overlaps the detection box corresponding to the steering wheel among the steering wheel detection results, the positional relationship between the steering wheel and the hand determines that the driver's both hands have left the steering wheel, and if If there is an area in which the detection box corresponding to at least one hand of the hand detection result overlaps the detection box corresponding to the steering wheel among the steering wheel detection results, the positional relationship between the steering wheel and the hand is determined by the driver's hand holding the steering wheel. It may include the step of determining that

Here, the positional relationship between the steering wheel and the hand may be determined using a detection box corresponding to the steering wheel among the steering wheel detection results and a detection box corresponding to the hand among the hand detection results.

When one hand is included in the hand detection result, if there is an area where the detection box corresponding to the hand overlaps the detection box corresponding to the steering wheel, the positional relationship between the steering wheel and the hand means that the hand is holding the steering wheel decide If there is an area in which the detection box corresponding to the hand does not overlap the detection box corresponding to the steering wheel, the positional relationship between the steering wheel and the hand determines that the hand leaves the steering wheel.

When both hands are included in the hand detection result, if there is an area where the detection box corresponding to at least one hand overlaps the detection box corresponding to the steering wheel, the positional relationship between the steering wheel and the hand is decide that If there is an area in which the detection boxes respectively corresponding to both hands do not overlap with the detection boxes corresponding to the steering wheel, the positional relationship between the steering wheel and the hand determines that the hand leaves the steering wheel.

In one selectable embodiment, determining the positional relationship between the steering wheel and the hand according to the steering wheel detection result and the hand detection result comprises:

generating an intermediate feature map corresponding to the image to be detected based on the image to be detected; and

generating a classification feature map of two channels corresponding to the intermediate feature map by performing convolution processing on the intermediate feature map at least once - Each channel feature map in the classification feature map of the two channels corresponds to one hand category - includes

Based on the center point position information indicated by the detection box information corresponding to the hand among the hand detection results, two feature values of the feature location matching the center point location information are extracted from the classification feature map, and the maximum feature value among the two feature values is extracted. is selected, and a category of the channel feature map corresponding to the maximum feature value among the classification feature maps is determined as a category corresponding to the center point location information.

A positional relationship between the steering wheel and the hand is determined based on a category corresponding to each center point position information indicated by the detection box information corresponding to the hand.

Here, when the second detection box information indicated by the hand detection result is not empty, convolution processing may be performed on the intermediate feature map at least once to generate a classification feature map of two channels corresponding to the intermediate feature map. Here, each channel feature map among the classification feature maps of the two channels corresponds to one hand category. For example, in the classification feature map, the category corresponding to the channel feature map of channel 0 may be a category in which the driver's hand leaves the steering wheel, and the category corresponding to the channel feature map of the first channel is the category in which the driver's hand leaves the steering wheel. It could be the category holding the steering wheel.

Furthermore, based on the center point location information indicated by the detection box information corresponding to the hand, extract two feature values of the feature location matching the center point location information from the classification feature map, and select the maximum feature value among the two feature values, A category of the channel feature map corresponding to the maximum feature value among the classification feature maps may be determined as a category corresponding to the center point location information.

When two center point position information is included in the detection box information corresponding to the hand (that is, the center point position information corresponding to the left hand and the center point position information corresponding to the right hand are included), for each center point position information, the center point position information is Determine the corresponding category.

For example, in the classification feature map, the category corresponding to the channel feature map of channel 0 may be a category in which the driver's hand leaves the steering wheel, and the category corresponding to the channel feature map of the first channel is the driver's hand. If it is a category holding the steering wheel, two feature values 0.8 and 0.2 are extracted from the classification feature map with respect to the center point location information corresponding to the left hand, and the 0th channel feature map corresponding to 0.8 in the classification feature map The category of is determined as a category of center point location information corresponding to the left hand, that is, the category of center point location information corresponding to the left hand is a category in which the driver's hand leaves the steering wheel. At the same time, the category of center point position information corresponding to the right hand can be obtained.

In the above embodiment, after determining the steering wheel detection result and performing convolution processing on the intermediate feature map at least once to generate a classification feature map, the generated center point position information of the driver's hand is combined, and the steering wheel and the hand The positional relationship between them can be determined more accurately.

In one selectable embodiment, determining the positional relationship between the steering wheel and the hand based on a category corresponding to each center point position information indicated by the detection box information corresponding to the hand may include:

Method 1: When the detection box information corresponding to the hand includes one center point position information, a category corresponding to the center point position information is determined as a positional relationship between the steering wheel and the hand;

Method 2: When two center point location information is included in the detection box information corresponding to the hand, and the category corresponding to the two center point location information is a category in which the driver's hand leaves the steering wheel, the positional relationship between the steering wheel and the hand determines that the driver's hand is off the steering wheel; When the category corresponding to at least one center point position information among the categories corresponding to the two center point position information is a category in which the driver's hand is holding the steering wheel, the positional relationship between the steering wheel and the hand is decide to hold

In the case of method 1, when one center point location information is included in the detection box information corresponding to the hand, that is, the center point location information corresponding to the left hand or the center point location information corresponding to the right hand is included in the detection box information corresponding to the hand. In this case, a category corresponding to the center point position information among detection box information corresponding to the hand may be determined as a positional relationship between the steering wheel and the hand. For example, when the center point location information corresponding to the left hand is included in the detection box information corresponding to the hand, and the category of the center point location information corresponding to the left hand is a category in which the driver's hand is holding the steering wheel, the steering wheel and The positional relationship between the hands is that the driver's hand is holding the steering wheel.

In the case of method 2, when two center point position information is included in the detection box information corresponding to the hand, that is, the center point position information corresponding to the left hand and the center point position information corresponding to the right hand are included in the detection box information corresponding to the hand. In this case, when the categories corresponding to the two center point position information are all categories in which the driver's hand deviates from the steering wheel, it is determined that the positional relationship between the steering wheel and the hand is that the driver's hand deviates from the steering wheel, and the left hand When the category of the center point position information corresponding to is a category in which the driver's hand is holding the steering wheel, and/or the category of the center point position information corresponding to the right hand is the category in which the driver's hand is holding the steering wheel, the steering wheel The positional relationship between and hand determines that the driver's hand is holding the steering wheel.

About S104:

Here, when it is determined that the driver's driving behavior category is dangerous driving, a warning message may be generated to the driver based on the driver's driving behavior category. Here, the warning message may be reproduced in a voice format. For example, the warning message generated could be "Dangerous. Hold the steering wheel."

In the above method according to the specific embodiment, the writing order of each step does not imply a strict execution order and does not make any limitation on the implementation process, and that the specific execution order of each step should be determined according to its function and possible implicit logic The point can be understood by those skilled in the art.

Based on the same concept as above, an embodiment of the present application further provides an apparatus for detecting a driver's behavior. Referring to FIG. 3 , the structural schematic diagram of the driver behavior detection device provided in the embodiment of the present application includes an acquisition module 301 , a detection module 302 , a determination module 303 , and a warning module 304 . do. Specifically:

the acquiring module 301 is configured to acquire an image to be detected of the driving position area in the cabin;

the detection module 302 is configured to perform detection on the image to be detected, to obtain a target detection result including a steering wheel detection result and a hand detection result;

the determining module 303 is configured to determine the driving behavior category of the driver according to the target detection result;

The warning module 304 is configured to send a warning message when the driving behavior category of the driver is dangerous driving.

In some embodiments of the present application, when the steering wheel detection result includes a steering wheel and the hand detection result includes a hand, when the determination module 303 determines the driver's driving behavior category according to the target detection result ,

determine a positional relationship between the steering wheel and the hand according to the steering wheel detection result and the hand detection result;

and determine the driving behavior category of the driver according to the positional relationship.

In some embodiments of the present application, when the determining module 303 determines the driving behavior category of the driver according to the positional relationship,

and determine that the driving behavior category of the driver is safe driving when the positional relationship indicates that the driver's hand is holding the steering wheel.

In some embodiments of the present application, when the determining module 303 determines the driving behavior category of the driver according to the positional relationship,

and determine that the driving behavior category of the driver is dangerous driving when the positional relationship indicates that the driver's both hands have left the steering wheel.

In some embodiments of the present application, when the steering wheel is included in the steering wheel detection result and the hand is not included in the hand detection result, the determination module 303 determines the driver's driving behavior category according to the target detection result when deciding,

and determine that the driving behavior category of the driver is dangerous driving according to the target detection result.

In some embodiments of the present application, when the detection module 302 performs detection on the to-be-detected image to obtain a target detection result,

generate an intermediate feature map corresponding to the to-be-detected image based on the to-be-detected image;

performing target convolution processing on the intermediate feature map at least once to generate a detection feature map of a plurality of channels corresponding to the intermediate feature map;

performing a feature value conversion process on each feature value of a target channel feature map indicating a location among the detection feature maps of a plurality of channels by using an activation function to generate a target channel feature map after transformation;

A maximum pooling process is performed on the target channel feature map after transformation according to a preset pooling size and pooling stride to obtain a plurality of pooling values and a location index corresponding to each pooling value among the plurality of pooling values, wherein the location index is the used to identify the location of the pooling value in the target channel feature map after transformation;

generate target detection box information based on a plurality of pooling values and a position index corresponding to each pooling value among the plurality of pooling values;

and determine the target detection result according to the target detection box information.

In some embodiments of the present application, when the detection module 302 generates target detection box information based on a plurality of pooling values and a position index corresponding to each pooling value among the plurality of pooling values,

When at least one pooling value among the plurality of pooling values is greater than a set pooling threshold, a target pooling value of the center point of the target detection box is determined from among the plurality of pooling values based on the plurality of pooling values and a pooling threshold and;

and generate the target detection box information based on a location index corresponding to the target pooling value.

In some embodiments of the present application, when the detection module 302 generates the target detection box information based on a plurality of pooling values and a position index corresponding to each pooling value among the plurality of pooling values,

and determine that the target detection box information is empty when the plurality of pooling values are less than or equal to a set pooling threshold.

In some embodiments of the present application, when the determining module 303 determines the positional relationship between the steering wheel and the hand according to the steering wheel detection result and the hand detection result,

When one hand is included in the hand detection result, if there is a region where a detection box corresponding to a hand in the hand detection result overlaps a detection box corresponding to the steering wheel in the steering wheel detection result, the steering wheel and the hand The positional relationship between the two determines that the driver's hand is holding the steering wheel, and if a detection box corresponding to the hand in the hand detection result overlaps with a detection box corresponding to the steering wheel in the steering wheel detection result does not exist. otherwise, the positional relationship between the steering wheel and the hand is configured to determine that both hands of the driver are off the steering wheel.

In some embodiments of the present application, when the determining module 303 determines the positional relationship between the steering wheel and the hand according to the steering wheel detection result and the hand detection result,

When both hands are included in the hand detection result, if a detection box corresponding to both hands in the hand detection result does not overlap with a detection box corresponding to the steering wheel in the steering wheel detection result, the steering wheel The positional relationship between the hand and the hand determines that both hands of the driver leave the steering wheel, and if a detection box corresponding to at least one hand among the hand detection results is a detection box corresponding to the steering wheel among the steering wheel detection results, If an overlapping region exists, the positional relationship between the steering wheel and the hand is configured to determine that the driver's hand is holding the steering wheel.

In some embodiments of the present application, when the determining module 303 determines the positional relationship between the steering wheel and the hand according to the steering wheel detection result and the hand detection result,

generate an intermediate feature map corresponding to the to-be-detected image based on the to-be-detected image;

performing convolution processing on the intermediate feature map at least once to generate a classification feature map of two channels corresponding to the intermediate feature map; wherein each channel feature map in the classification feature map of the two channels corresponds to one hand category;

Based on the center point location information indicated by the detection box information corresponding to the hand among the hand detection results, two feature values of a feature location matching the center point location information are extracted from the classification feature map, and the maximum feature among the two feature values selecting a value, and determining a category of a channel feature map corresponding to the maximum feature value among the classification feature maps as a category corresponding to the center point location information;

and determine a positional relationship between the steering wheel and the hand based on a category corresponding to each center point position information indicated by the detection box information corresponding to the hand.

In some embodiments of the present application, when the determining module 303 determines the positional relationship between the steering wheel and the hand based on a category corresponding to each center point position information indicated by the detection box information corresponding to the hand,

when the detection box information corresponding to the hand includes one center point position information, determining a category corresponding to the center point position information as a positional relationship between the steering wheel and the hand;

The detection box information corresponding to the hand includes two center point location information, and when the category corresponding to the two center point location information is a category in which the driver's hand leaves the steering wheel, the position between the steering wheel and the hand The relationship determines that both hands of the driver leave the steering wheel, and when the category corresponding to at least one center point location information among the categories corresponding to the two center point location information is a category in which the driver's hands are holding the steering wheel, The positional relationship between the steering wheel and the hand is configured to determine that the driver's hand is holding the steering wheel.

In some embodiments, the function provided by the device provided in the embodiments of the present application or the template included may be used to execute the method described in the method embodiment, and the specific implementation described in the method embodiment Reference may be made to the bar, and for the sake of brevity, redundant descriptions are omitted here.

Based on the same technical concept, embodiments of the present application further provide an electronic device. Referring to FIG. 4 , a structural schematic diagram of an electronic device provided in an embodiment of the present application includes a processor 401 , a memory 402 , and a bus 403 . Here, the memory 402 is configured to store an execution instruction, and includes a RAM 4021 and an external memory 4022 . Here, the RAM 4021 is also referred to as an internal memory, and is configured to temporarily store arithmetic data in the processor 401 and data exchanged with an external memory 4022 such as a hard disk. The processor 401 performs data exchange with the external memory 4022 through the RAM 4021 . When the electronic device 400 is executed, the processor 401 and the memory 402 communicate via the bus 403 to cause the processor 401 to execute the following commands:

instructions for obtaining an image to be detected of a driving position area within the cabin;

instructions for performing detection on the image to be detected to obtain a target detection result including a steering wheel detection result and a hand detection result;

a command for determining a driving behavior category of the driver according to the target detection result; and

A command to send a warning message when the driving behavior category of the driver is dangerous driving.

In addition, an embodiment of the present application further provides a computer-readable storage medium. A computer program is stored in the computer readable storage medium, and when the computer program is executed by a processor, the steps of the method for detecting driver behavior according to the embodiment of the method are executed.

The computer program product of the driver behavior detection method provided in the embodiment of the present application includes a computer-readable storage medium in which the program code is stored. The instructions included in the program code may be used to execute the steps of the method for detecting driver behavior according to the embodiment of the method, and for specific details, reference may be made to the embodiment of the method, and a redundant description will be omitted herein.

For a convenient and concise description, a person skilled in the art can clearly understand that a specific operation procedure of the above-described system and apparatus may refer to a corresponding procedure among the above-described method embodiments, and redundant description will be omitted herein. In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods may be implemented in other ways. The device embodiment described in the above text is only for explanation, for example, the division of the unit is only a kind of logical function division, and there may be a separate division method in actual implementation. For example, a plurality of units or components may be combined or integrated in separate systems, or some features may be omitted or not implemented, and further, the shown or discussed coupling or direct coupling to each other, or communication connection may be some It may be an indirect coupling or communication connection through a communication interface, device or unit, and may be in an electrical, mechanical or other form.

The unit described as a separate part may or may not be physically separate, and the part shown as a unit may or may not be a physical unit, that is, it may be located in one place or distributed in a plurality of network units. there is. According to actual needs, some or all of the units may be selected to implement the purpose of the present embodiment method.

In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, each unit may physically exist alone, or two or more than two units may be integrated into one unit. .

When the above function is implemented in the form of a software function unit and sold or used as an independent product, it may be stored in a non-volatile computer-readable storage medium executable by a processor. Based on this understanding, an essential part of the technical solution of the present application, that is, a part contributing to the prior art or a part of the technical solution may be implemented in the form of a software product. The computer software product is stored in a storage medium, and includes some instructions so that one computer device (which may be a personal computer, a server, or a network device, etc.) performs all or part of the steps of the method according to each embodiment of the present application. make it run The above-mentioned storage medium includes various media capable of storing program codes, such as U disk, external hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk. include

The above is only a specific embodiment of the present application, and the protection scope of the present application is not limited thereto. Any changes or replacements easily conceivable by those skilled in the art within the technical scope disclosed in this application should be included within the protection scope of the present application. Accordingly, the protection scope of the present application should be based on the protection scope of the claims.

In the embodiments of the present application, a method, an apparatus, an electronic device, a computer storage medium, and a computer program for detecting a driver's behavior are disclosed. The method includes: obtaining a program code used to control a first controlled device operation through a local agent program; and sending the program code to the first controlled device through the local agent program, so that the first controlled device executes the program code. According to the technical solution of the embodiment of the present application, it is possible to acquire and push the program code based on the local agent program, and further implement control of the first controlled device based on the program code. In addition, since the program code can be flexibly edited, it is possible to more flexibly control the first controlled device.

Claims (16)

In the driver behavior detection method,
acquiring an image to be detected of a driving position area within the cabin;
performing detection on the to-be-detected image to obtain a target detection result including a steering wheel detection result and a hand detection result;
determining a driving behavior category of the driver according to the target detection result; and
and sending a warning message when the driver's driving behavior category is dangerous driving.
Driver behavior detection method, characterized in that. According to claim 1,
When the steering wheel detection result includes a steering wheel and the hand detection result includes a hand, determining the driving behavior category of the driver according to the target detection result includes:
determining a positional relationship between the steering wheel and the hand according to the steering wheel detection result and the hand detection result; and
determining a driving behavior category of the driver according to the positional relationship
Driver behavior detection method, characterized in that. 3. The method of claim 2,
The step of determining the driving behavior category of the driver according to the positional relationship comprises:
determining that the driver's driving behavior category is safe driving when the positional relationship indicates that the driver's hand is holding the steering wheel
Driver behavior detection method, characterized in that. 3. The method of claim 2,
The step of determining the driver's driving behavior category according to the location relationship includes:
determining that the driving behavior category of the driver is dangerous driving when the positional relationship indicates that the driver's both hands have left the steering wheel;
Driver behavior detection method, characterized in that. 3. The method of claim 2,
The step of determining the positional relationship between the steering wheel and the hand according to the steering wheel detection result and the hand detection result,
When one hand is included in the hand detection result, if there is an area where a detection box corresponding to a hand in the hand detection result overlaps a detection box corresponding to the steering wheel in the steering wheel detection result, the steering wheel and the hand The positional relationship between the two determines that the driver's hand is holding the steering wheel, and if a detection box corresponding to the hand in the hand detection result overlaps with a detection box corresponding to the steering wheel in the steering wheel detection result does not exist. otherwise, determining that the positional relationship between the steering wheel and the hand is that both hands of the driver have left the steering wheel.
Driver behavior detection method, characterized in that. 3. The method of claim 2,
The step of determining the positional relationship between the steering wheel and the hand according to the steering wheel detection result and the hand detection result,
When both hands are included in the hand detection result, if a detection box corresponding to both hands in the hand detection result does not overlap with a detection box corresponding to the steering wheel in the steering wheel detection result, the steering wheel The positional relationship between the hand and the hand determines that the driver's both hands have left the steering wheel, and if a detection box corresponding to at least one hand among the hand detection results is a detection box corresponding to the steering wheel among the steering wheel detection results, determining that the positional relationship between the steering wheel and the hand is that the driver's hand is holding the steering wheel if an overlapping area exists.
Driver behavior detection method, characterized in that. 3. The method of claim 2,
The step of determining the positional relationship between the steering wheel and the hand according to the steering wheel detection result and the hand detection result,
generating an intermediate feature map corresponding to the image to be detected based on the image to be detected;
generating a classification feature map of two channels corresponding to the intermediate feature map by performing convolution processing on the intermediate feature map at least once Corresponds to category -;
extracting two feature values of a feature location matching the center point location information from the classification feature map based on center point location information indicated by detection box information corresponding to a hand among the hand detection results;
selecting a maximum feature value among the two feature values and determining a category of a channel feature map corresponding to the maximum feature value among the classification feature maps as a category corresponding to the center point location information; and
determining a positional relationship between the steering wheel and the hand based on a category corresponding to each center point position information indicated by the detection box information corresponding to the hand
Driver behavior detection method, characterized in that. 8. The method of claim 7,
The step of determining a positional relationship between the steering wheel and the hand based on a category corresponding to each center point position information indicated by the detection box information corresponding to the hand comprises:
determining a category corresponding to the center point position information as a positional relationship between the steering wheel and the hand when the detection box information corresponding to the hand includes one center point position information; and
The detection box information corresponding to the hand includes two center point location information, and when the category corresponding to the two center point location information is a category in which the driver's hand leaves the steering wheel, the distance between the steering wheel and the hand When the positional relationship determines that both hands of the driver leave the steering wheel, and a category corresponding to at least one center point position information among the categories corresponding to the two center point location information is a category in which the driver's hands are holding the steering wheel , determining that the positional relationship between the steering wheel and the hand is that the driver's hand is holding the steering wheel.
Driver behavior detection method, characterized in that. According to claim 1,
When the steering wheel is included in the steering wheel detection result and the hand is not included in the hand detection result, determining the driving behavior category of the driver according to the target detection result includes:
determining that the driving behavior category of the driver is dangerous driving according to the target detection result
Driver behavior detection method, characterized in that. 10. The method according to any one of claims 1 to 9,
The step of obtaining a target detection result by performing detection on the image to be detected includes:
generating an intermediate feature map corresponding to the to-be-detected image based on the to-be-detected image;
performing a target convolution process on the intermediate feature map at least once to generate a detection feature map of a plurality of channels corresponding to the intermediate feature map;
generating a target channel feature map after transformation by performing a feature value conversion process on each feature value of a target channel feature map indicating a location among the detection feature maps of a plurality of channels using an activation function;
Obtaining a plurality of pooling values and a position index corresponding to each pooling value among the plurality of pooling values by performing maximum pooling processing on the target channel feature map after transformation according to a preset pooling size and pooling stride - The position index is the used to identify the location of the pooling value in the target channel feature map after transformation;
generating target detection box information based on a plurality of pooling values and a location index corresponding to each pooling value among the plurality of pooling values; and
determining the target detection result according to the target detection box information
Driver behavior detection method, characterized in that. 11. The method of claim 10,
Generating the target detection box information based on a plurality of pooling values and a position index corresponding to each pooling value among the plurality of pooling values,
When at least one pooling value among the plurality of pooling values is greater than a set pooling threshold, a target pooling value of the center point of the target detection box is determined from among the plurality of pooling values based on the plurality of pooling values and a pooling threshold to do; and
generating the target detection box information based on a location index corresponding to the target pooling value
Driver behavior detection method, characterized in that. 11. The method of claim 10,
Generating the target detection box information based on a plurality of pooling values and a position index corresponding to each pooling value among the plurality of pooling values,
When the plurality of pooling values are less than or equal to a set pooling threshold, determining that the target detection box information is empty
Driver behavior detection method, characterized in that. In the driver behavior detection device,
an acquiring module, configured to acquire an image to be detected of a driving position area within the cabin;
a detection module, configured to perform detection on the to-be-detected image to obtain a target detection result including a steering wheel detection result and a hand detection result;
a determining module, configured to determine a driving behavior category of the driver according to the target detection result; and
a warning module, configured to send a warning message when the driving behavior category of the driver is dangerous driving;
Driver behavior detection device, characterized in that. In an electronic device,
a processor, a memory and a bus, wherein the memory stores machine readable instructions executable by the processor, and when an electronic device is executed, the processor communicates with the memory via a bus, the machine readable instructions which executes the method of detecting driver behavior according to any one of claims 1 to 12 when it is executed by the processor
Electronic device, characterized in that. A computer-readable storage medium comprising:
A computer program is stored in the computer readable storage medium, and when the computer program is executed by a processor, the method for detecting driver behavior according to any one of claims 1 to 12 is executed.
A computer-readable storage medium, characterized in that. In a computer program,
A computer readable code comprising computer readable code, wherein when the computer readable code is executed in an electronic device, a processor in the electronic device is configured to execute the method for detecting driver behavior according to any one of claims 1 to 12.
A computer program characterized in that.

Images