CN112455327A - Vehicle lamp control method and device and computer readable storage medium - Google Patents
Vehicle lamp control method and device and computer readable storage medium Download PDFInfo
- Publication number
- CN112455327A CN112455327A CN202011395286.XA CN202011395286A CN112455327A CN 112455327 A CN112455327 A CN 112455327A CN 202011395286 A CN202011395286 A CN 202011395286A CN 112455327 A CN112455327 A CN 112455327A
- Authority
- CN
- China
- Prior art keywords
- driver
- lamp
- eye
- vehicle
- angle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60Q—ARRANGEMENT OF SIGNALLING OR LIGHTING DEVICES, THE MOUNTING OR SUPPORTING THEREOF OR CIRCUITS THEREFOR, FOR VEHICLES IN GENERAL
- B60Q3/00—Arrangement of lighting devices for vehicle interiors; Lighting devices specially adapted for vehicle interiors
- B60Q3/80—Circuits; Control arrangements
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Lighting Device Outwards From Vehicle And Optical Signal (AREA)
Abstract
The invention provides a car light control method, a car light control device and a computer readable storage medium, which are applied to the technical field of automobiles. The vehicle lamp control method provided by the invention can control the working state of the vehicle lamp according to the eye action and the sight angle of the driver, effectively prevents the hand of the driver from being separated from the steering wheel due to the operation of the lamp switch, and further improves the driving safety.
Description
Technical Field
The invention belongs to the technical field of automobiles, and particularly relates to a method and a device for controlling a car lamp and a computer readable storage medium.
Background
The automobile lamp mainly comprises an automobile inner lamp and an automobile outer lamp, wherein the automobile outer lamp is used for illuminating a driving environment when the automobile runs at night, and the automobile inner lamp is used for illuminating the automobile in a dark environment, such as an automobile inner reading lamp and an automobile inner atmosphere lamp.
In the prior art, the lamps in the vehicle are mostly turned on and off through manual operation, and the working state of the lamps in the vehicle is naturally controlled by manually operating the lamp switch in a static state of the vehicle.
Disclosure of Invention
In view of the above, an object of the present invention is to provide a method and an apparatus for controlling a vehicle lamp, and a computer readable storage medium, which control a working state of a lamp in a vehicle according to an eye movement and a sight angle of a driver, and effectively prevent a hand of the driver from being separated from a steering wheel by operating a lamp switch, so as to improve driving safety, and the specific scheme is as follows:
in a first aspect, the present invention provides a vehicle lamp control method, including:
acquiring a plurality of facial pictures of a driver within a preset time length;
determining eye actions and sight angles of the driver based on the face picture of the driver;
if the eye movement represents that the driver blinks, judging whether the driver stares at an in-vehicle lamp according to the sight angle;
and if the driver watches the lamp in the vehicle, switching the working state of the lamp in the vehicle.
Optionally, the vehicle lamp control method provided by the first aspect of the present invention further includes:
obtaining the irradiation angle of the external lamp;
and controlling the external lamp according to the consistency of the sight angle and the irradiation angle.
Optionally, the determining the eye movement and the sight angle of the driver based on the facial picture of the driver includes:
calling a pre-trained recognition model, wherein the recognition model is obtained by training a neural network by taking a facial picture as a training sample and eye actions and sight angles corresponding to the facial picture as labels, and the eye actions comprise eye opening and eye closing;
inputting the face picture of the driver into the recognition model to obtain an eye action recognition result and the sight angle of the driver;
if the ratio of the number of the face pictures of the driver with eye movement determined to be eye closing to the number of all the face pictures of the driver in the eye movement identification result is greater than or equal to a preset ratio threshold value, determining that the driver blinks;
and if the ratio of the number of the face pictures of the driver with the eye motion being closed to the number of all the face pictures of the driver in the eye motion recognition result is smaller than the preset ratio threshold, determining that the driver does not blink.
Optionally, the neural network includes: an input layer module, a convolution layer network module, a Lenet two-classification network module and an Alexnet network module which are formed by cascading a plurality of Hourglass network modules, wherein,
the input layer module is used for extracting an eye feature map in the face picture of the driver;
the convolution layer network module is used for extracting eye key point data based on the eye feature map;
the Lenet two-classification network module is used for determining the eye action based on the eye key point data;
the Alexnet network module is used for determining the sight angle of the driver based on the eye key point data.
Optionally, the determining whether the driver watches the lamp in the vehicle according to the sight angle includes:
acquiring a reference sight line vector, wherein the reference sight line vector is the sight line vector when the driver looks at the in-vehicle lamp;
calculating the current sight line vector of the driver according to the sight line angle;
if the deviation of the current sight line vector and the reference sight line vector is within a first preset deviation range, judging that the driver watches the in-vehicle lamp;
and if the deviation of the current sight line vector and the reference sight line vector is not within a first preset deviation range, judging that the driver does not watch the in-vehicle lamp.
Optionally, the switching the working state of the lamp in the vehicle includes:
if the in-vehicle lamp is in an on state currently, turning off the in-vehicle lamp;
and if the lamp in the vehicle is in the off state currently, the lamp in the vehicle is turned on.
Optionally, the controlling the exterior lighting fixture according to the consistency of the sight line angle and the irradiation angle includes:
calculating the root mean square value of the sight angle and the irradiation angle;
if the root mean square value is within a second preset deviation range, maintaining the irradiation angle of the external lamp unchanged;
and if the root mean square value is not within the second preset deviation range, adjusting the irradiation angle of the external lamp according to the sight angle.
Optionally, the picture of the face of the driver is an infrared picture.
In a second aspect, the present invention provides a vehicle lamp control device, comprising:
the system comprises a first acquisition unit, a second acquisition unit and a display unit, wherein the first acquisition unit is used for acquiring a plurality of facial pictures of a driver within a preset time length;
the determining unit is used for determining the eye action and the sight angle of the driver based on the face picture of the driver;
the judging unit is used for judging whether the driver watches the in-vehicle lamp according to the sight angle if the eye movement represents that the driver blinks;
and the first control unit is used for switching the working state of the lamp in the vehicle if the driver watches the lamp in the vehicle.
In a third aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, wherein the computer instructions are executed to perform the steps of the vehicle lamp control method according to any one of the first aspect of the present invention.
Based on the technical scheme, the method for controlling the vehicle lamp provided by the invention has the advantages that after a plurality of facial pictures of the driver within the preset time are obtained, the eye action and the sight angle of the driver are determined based on the facial pictures of the driver, if the eye action of the driver represents that the driver blinks, whether the driver gazes at the lamp in the vehicle is judged according to the sight angle of the driver, and if the driver gazes at the lamp in the vehicle, the working state of the lamp in the vehicle is switched. The vehicle lamp control method provided by the invention can control the working state of the vehicle lamp according to the eye action and the sight angle of the driver, effectively prevents the hand of the driver from being separated from the steering wheel due to the operation of the lamp switch, and further improves the driving safety.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a flow chart of a vehicle light control method according to an embodiment of the present invention;
FIG. 2 is a network architecture diagram of a neural network provided by an embodiment of the present invention;
FIG. 3 is a flow chart of another method for controlling a vehicle light according to an embodiment of the present invention;
fig. 4 is a block diagram of a vehicle lamp control device according to an embodiment of the present invention;
fig. 5 is a block diagram of another vehicle lamp control device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, fig. 1 is a flowchart of a vehicle lamp control method provided in an embodiment of the present invention, where the method is applicable to a vehicle-mounted electronic device, and the vehicle-mounted electronic device may be a vehicle-mounted electronic device, such as a vehicle control unit, a vehicle lamp controller, an auxiliary driving system controller, and the like, which has data processing capability and can run a preset program, obviously, the vehicle-mounted electronic device may also be implemented by using a server on a network side in some cases; referring to fig. 1, a process of a vehicle lamp control method according to an embodiment of the present invention may include:
s100, obtaining a plurality of face pictures of the driver within a preset time length.
According to the vehicle lamp control method provided by the embodiment of the invention, the driver face picture is used as a basic judgment basis for controlling the vehicle lamp, and when the vehicle lamp is controlled, a plurality of driver face pictures within a preset time length are acquired, for example, a plurality of driver face pictures within 1s are acquired. In the prior art, most image acquisition devices can acquire 24 frames of pictures within 1s, and even more, the number of the pictures of the face of the driver provided by the image acquisition technology in the prior art in a short time can completely meet the application requirement of the scheme.
Therefore, although the subsequent content of the invention relates to the judgment of the sight angle of the driver and requires the driver to watch the lamp in the vehicle, the process is very short in the actual implementation process and does not require the driver to watch the lamp in the vehicle for a long time, so that the vehicle lamp control method provided by the embodiment of the invention has the requirement on the sight angle of the driver in the implementation process, but the driving safety is not influenced due to very short time.
Optionally, because to the use of light in the car, mostly at night or under the relatively poor condition of ambient light, also can bring the influence to the collection of driver's facial picture under this kind of condition, influence the recognition accuracy degree of follow-up driver's eye action and sight angle, therefore, when gathering driver's facial picture, can adopt infrared camera to realize, compare with ordinary camera, for example RGB camera, infrared camera also can acquire the driver's facial picture of higher definition under the low light condition, help improving the execution effect of follow-up step.
And S110, determining the eye movement and the sight angle of the driver based on the face picture of the driver.
In the embodiment of the present invention, the eye movements of the driver mainly include open eyes and closed eyes. Furthermore, according to the basic structure of human eyes, the eyeball can be regarded as a spherical rigid body, so that three parameters describing the rotation motion of the spherical rigid body can be used for representing the sight angle of the eyes, specifically, the three parameters are a yaw angle, a pitch angle and a roll angle. The yaw angle is used for representing left and right eyes, the pitch angle is used for representing up and down eyes, and the roll angle cannot occur due to the structural limitation of human eyeballs. Therefore, in the present embodiment, the driver's gaze angle is mainly characterized by two parameters, i.e., the yaw angle and the pitch angle of the eyeball.
Optionally, an embodiment of the present invention provides a method for determining an eye movement and a sight angle of a driver based on a facial picture of the driver, and specifically,
and calling a pre-trained recognition model, wherein the recognition model is obtained by training a neural network by taking the face picture as a training sample and taking the eye action and the sight angle corresponding to the face picture as labels. After the recognition model is called, the face picture of the driver is input into the recognition model, and the eye action recognition result of the driver and the sight angle of the driver can be obtained through the processing of the recognition model.
Optionally, the recognition model processes only one driver face picture at a time, and outputs an eye movement recognition result and a driver's sight angle corresponding to the driver face picture. As mentioned above, the eye movement of the driver includes two types, namely, eye opening and eye closing, and accordingly, when the driver closes the eyes, the sight line angle of the driver is not considered, and the sight line angle referred to in the present scheme refers to the sight line angle when the eye movement of the driver is the eye opening. Further, in most application scenarios, there are a plurality of eye movements of the driver as the open-eye pictures, which means that a plurality of viewing angles are obtained, and as an alternative implementation manner, each eye movement may be used as an average value of the viewing angles corresponding to the open-eye driver pictures as the viewing angle used in the subsequent step.
It should be noted that other methods may also be adopted to identify the eye movement and the implementation angle of the driver based on the face picture of the driver, and the method also belongs to the protection scope of the present invention without exceeding the scope of the core idea of the present invention.
Alternatively, referring to fig. 2, fig. 2 is a network architecture diagram of a neural network for training a recognition model according to an embodiment of the present invention. The neural network provided by the embodiment of the invention comprises an input layer module (shown by 5 Hourglass network modules in figure 2) formed by cascading a plurality of Hourglass network modules, a convolutional layer network module, a Lenet two-class network module and an Alexnet network module, wherein,
the input layer module is used for extracting the eye feature map in the face picture of the driver. In practical applications, in order to improve the extraction efficiency of the eye feature map and the accuracy of the subsequent recognition result, the size of the image input to the neural network may be preprocessed, for example, the sizes of the input images may be unified to 256 × 256, and accordingly, the output of the input layer module may also be the eye feature map with the same size.
The input data of the convolutional layer network module is the aforementioned eye feature map, and the output data of the convolutional layer network module is eye key point data, for example, the output data may be a feature map including 18 eye key points, and the 18 eye key points may correspond to 3 periocular key points, 4 iris key points, eyeball center, and iris center of two eyes. It should be noted that, for the eye key points, other selections may be made according to related contents in the prior art, and the specific selection of the eye key points in the present invention is not limited, and the eye key points that can be used for determining the eye movement are also optional and also belong to the scope of the present invention.
After the eye key point data is obtained, an Lenet two-classification network module and an Alexnet network module are respectively input, wherein the Lenet two-classification network module is used for determining the eye action based on the eye key point data, and the Alexnet network module is used for determining the sight angle of the driver based on the eye key point data. For the specific structure of the Alexnet network module and the Lenet two-class network module, the implementation can be realized by referring to the prior art, and details are not described here.
The recognition model used in the embodiment of the present invention can be obtained by training the neural network with facial images including different eye movements and gaze angles as training samples, and the specific training process can be implemented by referring to the training process of the neural network in the prior art, which is not specifically limited by the present invention.
And S120, judging whether the eye movement represents the blinking of the driver, if so, executing S130.
As described above, the eye movement of the driver in the embodiment of the present invention mainly includes eye opening and eye closing, and accordingly, the eye movement recognition result described in the embodiment of the present invention is specifically to distinguish the respective facial pictures of the driver, that is, the facial picture of the driver obtained in S100 is divided into the facial picture of the driver in which the eye movement is determined as eye opening and the facial picture of the driver in which the eye movement is determined as eye closing by a recognition model, and whether the driver blinks is determined based on the recognition result.
Specifically, if the ratio of the number of the face pictures of the driver with eye closing motion to the number of all the face pictures of the driver is judged to be greater than or equal to a preset ratio threshold value in the eye motion recognition result, the driver is determined to blink; on the contrary, if the ratio of the number of the face pictures of the driver, of which the eye movement is determined to be eye closing, to the number of all the face pictures of the driver in the eye movement recognition result is smaller than the preset proportion threshold value, it is determined that the driver does not blink.
For example, 24 face pictures of the driver are acquired, the preset proportion threshold is 50%, and if 12 or more eye movement determination eye movements in the 24 face pictures of the driver are taken as closed-eye face pictures, the driver can be determined to blink.
It should be noted that, for the selection of the preset proportion threshold, the selection may be flexibly selected according to the requirement for the recognition accuracy in the actual application, the computing capability of the controller, and the total number of the facial images of the driver.
It should be noted that, if the eye movement indicates that the driver is not blinking, the current control flow is exited, and the vehicle light control of the next cycle is started.
And S130, judging whether the driver watches the lamp in the vehicle according to the sight angle, and if so, executing S140.
In the case that it is determined that the driver blinks in S120, the vehicle lamp control method according to the embodiment of the present invention further determines whether the driver gazes at the interior lamp according to the sight angle obtained in the foregoing step.
Optionally, in order to determine whether the driver watches the in-vehicle lamp based on the obtained sight angle, the embodiment of the present invention provides a reference sight vector, where the reference sight vector is a sight vector when the driver watches the in-vehicle lamp. Specifically, the midpoint of the connecting line of the 2 eyes of the driver is taken as the origin of the sight line, the central point of a certain lamp in the vehicle is taken as a target point, a world coordinate system is established, the origin O of the world coordinate system can be any point, the OX axis points to the front, the OZ axis direction is vertical downward, and the OY axis can be determined according to the right-hand rule. Based on the positions of the sight line origin and the target point in the world coordinate system, a unit vector from the sight line origin to the target point in the world coordinate system can be calculated, which is marked as [ s1, s2, s3], and is a reference sight line vector used in the embodiment of the present invention. It should be noted that, for the reference sight line vector, other methods may be used for calculation and characterization, and any method capable of providing a unit vector from the origin of the driver's sight line to the lighting fixture in the vehicle is optional, and the method also falls within the protection scope of the present invention without departing from the scope of the core idea of the present invention. Further, the determination of the sight line origin may be implemented by using a method in the prior art, which is not limited by the present invention.
Further, calculating the current sight line vector of the driver according to the sight line angle of the driver obtained in the previous step. Under the world coordinate system, 2 view angles, namely the yaw angle and the pitch angle, which are respectively denoted as m and n, can determine a unique current view vector in the world coordinate system, and specifically, the current view vector can be expressed as [ cos (n) cos (m), cos (n) sin (m), -sin (n) ]. It should be noted that, for a specific algorithm for converting into a current sight line vector in a world coordinate system based on a yaw angle and a pitch angle, the specific algorithm may be implemented with reference to the prior art, and the present invention is not limited to this.
After the current sight line vector and the reference sight line vector are obtained, further comparing the deviation between the current sight line vector and the reference sight line vector, and if the deviation between the current sight line vector and the reference sight line vector is within a first preset deviation range, judging that a driver watches the lamp in the vehicle; on the contrary, if the deviation of the current sight line vector and the reference sight line vector is not within the first preset deviation range, the driver is judged not to watch the lamp in the vehicle.
Alternatively, the deviation between the current sight line vector and the reference sight line vector can be measured by the root mean square error between the two. The first preset deviation range can be flexibly selected according to the requirement of actual control precision, and the specific selection of the first preset deviation range is not limited by the invention.
When it is determined that the driver is not looking at the in-vehicle lamp, the current control process is exited and the control process of the next cycle is started.
And S140, switching the working state of the lamp in the vehicle.
Based on the steps, if the driver blinks and watches the in-vehicle lamp, the working state of the in-vehicle lamp is switched. Specifically, if the in-vehicle lamp is currently in an on state, the in-vehicle lamp is turned off; and if the lamp in the vehicle is in the off state at present, the lamp in the vehicle is turned on.
In summary, according to the car light control method provided by the invention, after a plurality of driver face pictures within a preset time are acquired, the eye movement and the sight angle of the driver are determined based on the driver face pictures, if the eye movement of the driver represents that the driver blinks, whether the driver gazes at the car interior lamp is judged according to the sight angle of the driver, and if the driver gazes at the car interior lamp, the working state of the car interior lamp is switched. The vehicle lamp control method provided by the invention can control the working state of the vehicle lamp according to the eye action and the sight angle of the driver, effectively prevents the hand of the driver from being separated from the steering wheel due to the operation of the lamp switch, and further improves the driving safety.
In practical application, the irradiation angle of an outside lamp installed on an automobile is mostly fixed, particularly, a front lamp is usually emitted to the right front of the automobile, when the automobile runs at a strange complex intersection, a driver usually pays attention to pedestrians and road surface environments on the side of the automobile, and needs a larger-angle view.
To solve the problem, the invention provides another vehicle lamp control method based on the sight line of the driver. According to the method provided by the embodiment, the external lamp of the vehicle is required, and according to the actual application requirements, the driving motor capable of driving the headlamp to rotate is mainly arranged on the headlamp of the vehicle, and meanwhile, an angle sensor capable of feeding back the irradiation angle of the external lamp is also arranged. On this basis, referring to fig. 3, fig. 3 is a flowchart of another vehicle lamp control method provided in the embodiment of the present invention, and on the basis of the embodiment shown in fig. 1, the method further includes:
s200, obtaining the irradiation angle of the external lamp.
As mentioned above, the vehicle applying the method is pre-equipped with the angle sensor for measuring the illumination angle of the lamp, and the illumination angle of the lamp outside the vehicle can be obtained through the angle sensor. Optionally, in practical application, the illumination angle of the exterior lighting fixture mainly refers to a yaw angle and a pitch angle of the exterior lighting fixture.
It is conceivable that the specific execution time of S200 may be executed after S110, or may be executed synchronously when S100 is executed, the specific execution time of S200 is not specifically limited in the present invention, and the flowchart shown in fig. 3 is not limited to the execution time of S200, and is only used as a reference for explaining the present method.
S210, controlling the external lamp according to the consistency of the sight angle and the irradiation angle.
Alternatively, the acquisition process of the driver' S sight angle may be implemented with reference to the implementation process of S110.
For deviations between the viewing angle and the illumination angle, a root mean square difference between the two can be used for characterization. For example, the viewing angle is denoted as a yaw angle a and a pitch angle b, the illumination angle is denoted as a yaw angle c and a pitch angle d, and the root mean square difference between the viewing angle and the illumination angle can be calculated by the following formula:
meanwhile, a second preset deviation range is set, and after the root mean square values of the two are obtained, if the root mean square value is within the second preset deviation range, the sight angle is considered to be consistent with the irradiation angle, and the irradiation angle of the external lamp is kept unchanged; on the contrary, if the root mean square value is not in the second preset deviation range, the fact that the sight angle of the driver is inconsistent with the irradiation angle is judged, and the irradiation angle of the external lamp is further adjusted according to the sight angle of the driver.
It should be noted that, similar to the first preset deviation range, the selection of the second preset deviation range may also be flexibly selected according to the actual control requirement, and the specific setting of the second preset deviation range is not limited in the present invention.
For another example, the difference between the yaw angle a and the yaw angle c, and the difference between the pitch angle b and the pitch angle d may be calculated, and then compared with the preset threshold corresponding to the obtained difference, and if both the differences are smaller than the corresponding thresholds, it is determined that the vehicle exterior light is consistent with the sight line.
Or calculating a sight line direction vector based on the sight line angle, calculating a vehicle lamp irradiation direction vector based on the irradiation angle, then calculating an included angle between the two vectors, and comparing the calculation result with a preset included angle threshold value, thereby judging the consistency of the sight line angle and the irradiation angle.
Of course, in any of these methods, when it is determined that the driver's sight angle does not coincide with the lamp irradiation angle, it is necessary to control the turning of the exterior lamp in accordance with the driver's sight angle and adjust the irradiation angle of the lamp.
In summary, the vehicle lamp control method provided in the embodiments of the present invention can further control the external lamp of the vehicle according to the driver's sight on the premise that the internal lamp can be controlled based on the driver's sight, so that the illumination angle of the external lamp can be changed according to the viewing angle of the driver, thereby providing a wider field of view for the driver, and contributing to improving the driving safety.
In the following, the vehicle lamp control device provided in the embodiment of the present invention is introduced, and the vehicle lamp control device described below may be regarded as a functional module architecture that needs to be set in the central device to implement the vehicle lamp control method provided in the embodiment of the present invention; the following description may be cross-referenced with the above.
Fig. 4 is a block diagram of a vehicle lamp control device according to an embodiment of the present invention, and referring to fig. 4, the device may include:
the first acquisition unit 10 is used for acquiring a plurality of facial pictures of the driver within a preset time length;
a determination unit 20 configured to determine an eye movement and a sight line angle of the driver based on the driver face picture;
the judging unit 30 is configured to judge whether the driver watches the in-vehicle lamp according to the sight angle if the eye movement represents that the driver blinks;
and the first control unit 40 is used for switching the working state of the lamp in the vehicle if the driver watches the lamp in the vehicle.
Optionally, the determining unit 20 is configured to, when determining the eye movement and the sight angle of the driver based on the facial picture of the driver, specifically include:
calling a pre-trained recognition model, wherein the recognition model is obtained by training a neural network by taking a facial picture as a training sample and eye actions and sight angles corresponding to the facial picture as labels, and the eye actions comprise eye opening and eye closing;
inputting the face picture of the driver into the recognition model to obtain an eye action recognition result and the sight angle of the driver;
if the ratio of the number of the face pictures of the driver with eye movement determined to be eye closing to the number of all the face pictures of the driver in the eye movement identification result is greater than or equal to a preset ratio threshold value, determining that the driver blinks;
and if the ratio of the number of the face pictures of the driver with the eye motion being closed to the number of all the face pictures of the driver in the eye motion recognition result is smaller than the preset ratio threshold, determining that the driver does not blink.
Optionally, the neural network includes: an input layer module, a convolution layer network module, a Lenet two-classification network module and an Alexnet network module which are formed by cascading a plurality of Hourglass network modules, wherein,
the input layer module is used for extracting an eye feature map in the face picture of the driver;
the convolution layer network module is used for extracting eye key point data based on the eye feature map;
the Lenet two-classification network module is used for determining the eye action based on the eye key point data;
the Alexnet network module is used for determining the sight angle of the driver based on the eye key point data.
Optionally, the determining unit 30 is configured to determine whether the driver watches the lamp in the vehicle according to the sight angle, and specifically includes:
acquiring a reference sight line vector, wherein the reference sight line vector is the sight line vector when the driver looks at the in-vehicle lamp;
calculating the current sight line vector of the driver according to the sight line angle;
if the deviation of the current sight line vector and the reference sight line vector is within a first preset deviation range, judging that the driver watches the in-vehicle lamp;
and if the deviation of the current sight line vector and the reference sight line vector is not within a first preset deviation range, judging that the driver does not watch the in-vehicle lamp.
Optionally, the first control unit 40 is configured to, when switching the working state of the in-vehicle lamp, specifically include:
if the in-vehicle lamp is in an on state currently, turning off the in-vehicle lamp;
and if the lamp in the vehicle is in the off state currently, the lamp in the vehicle is turned on.
Optionally, referring to fig. 5, fig. 5 is a block diagram of another vehicle lamp control device provided in the embodiment of the present invention, and on the basis of the embodiment shown in fig. 4, the device further includes:
a second obtaining unit 50 for obtaining an irradiation angle of the exterior lamp;
and a second control unit 60 for controlling the exterior lighting fixture according to the consistency of the sight angle and the illumination angle.
Optionally, when the second control unit 60 is configured to control the external lamp according to the consistency between the sight angle and the irradiation angle, the method specifically includes:
calculating the root mean square value of the sight angle and the irradiation angle;
if the root mean square value is within a second preset deviation range, maintaining the irradiation angle of the external lamp unchanged;
and if the root mean square value is not within the second preset deviation range, adjusting the irradiation angle of the external lamp according to the sight angle.
Optionally, the present application further provides a computer-readable storage medium, on which computer instructions are stored, and when the computer instructions are executed, the steps of the vehicle lamp control method in any of the above embodiments are executed.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A vehicle lamp control method characterized by comprising:
acquiring a plurality of facial pictures of a driver within a preset time length;
determining eye actions and sight angles of the driver based on the face picture of the driver;
if the eye movement represents that the driver blinks, judging whether the driver stares at an in-vehicle lamp according to the sight angle;
and if the driver watches the lamp in the vehicle, switching the working state of the lamp in the vehicle.
2. The vehicular lamp control method according to claim 1, characterized by further comprising:
obtaining the irradiation angle of the external lamp;
and controlling the external lamp according to the consistency of the sight angle and the irradiation angle.
3. The vehicle light control method according to claim 1, wherein the determining the eye movement and the sight line angle of the driver based on the driver face picture comprises:
calling a pre-trained recognition model, wherein the recognition model is obtained by training a neural network by taking a facial picture as a training sample and eye actions and sight angles corresponding to the facial picture as labels, and the eye actions comprise eye opening and eye closing;
inputting the face picture of the driver into the recognition model to obtain an eye action recognition result and the sight angle of the driver;
if the ratio of the number of the face pictures of the driver with eye movement determined to be eye closing to the number of all the face pictures of the driver in the eye movement identification result is greater than or equal to a preset ratio threshold value, determining that the driver blinks;
and if the ratio of the number of the face pictures of the driver with the eye motion being closed to the number of all the face pictures of the driver in the eye motion recognition result is smaller than the preset ratio threshold, determining that the driver does not blink.
4. The vehicular lamp control method according to claim 3, wherein the neural network comprises: an input layer module, a convolution layer network module, a Lenet two-classification network module and an Alexnet network module which are formed by cascading a plurality of Hourglass network modules, wherein,
the input layer module is used for extracting an eye feature map in the face picture of the driver;
the convolution layer network module is used for extracting eye key point data based on the eye feature map;
the Lenet two-classification network module is used for determining the eye action based on the eye key point data;
the Alexnet network module is used for determining the sight angle of the driver based on the eye key point data.
5. The vehicle light control method according to claim 1, wherein the determining whether the driver is looking at the in-vehicle light according to the sight angle comprises:
acquiring a reference sight line vector, wherein the reference sight line vector is the sight line vector when the driver looks at the in-vehicle lamp;
calculating the current sight line vector of the driver according to the sight line angle;
if the deviation of the current sight line vector and the reference sight line vector is within a first preset deviation range, judging that the driver watches the in-vehicle lamp;
and if the deviation of the current sight line vector and the reference sight line vector is not within a first preset deviation range, judging that the driver does not watch the in-vehicle lamp.
6. The vehicle lamp control method according to claim 1, wherein the switching of the operating state of the interior lamp includes:
if the in-vehicle lamp is in an on state currently, turning off the in-vehicle lamp;
and if the lamp in the vehicle is in the off state currently, the lamp in the vehicle is turned on.
7. The vehicle lamp control method according to claim 2, wherein the controlling of the exterior lamp according to the coincidence of the sight-line angle and the irradiation angle includes:
calculating the root mean square value of the sight angle and the irradiation angle;
if the root mean square value is within a second preset deviation range, maintaining the irradiation angle of the external lamp unchanged;
and if the root mean square value is not within the second preset deviation range, adjusting the irradiation angle of the external lamp according to the sight angle.
8. The vehicular lamp control method according to any one of claims 1 to 7, wherein the picture of the driver's face is an infrared picture.
9. A vehicle lamp control device characterized by comprising:
the system comprises a first acquisition unit, a second acquisition unit and a display unit, wherein the first acquisition unit is used for acquiring a plurality of facial pictures of a driver within a preset time length;
the determining unit is used for determining the eye action and the sight angle of the driver based on the face picture of the driver;
the judging unit is used for judging whether the driver watches the in-vehicle lamp according to the sight angle if the eye movement represents that the driver blinks;
and the first control unit is used for switching the working state of the lamp in the vehicle if the driver watches the lamp in the vehicle.
10. A computer readable storage medium having computer instructions stored thereon, wherein the computer instructions when executed perform the steps of the vehicle light control method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011395286.XA CN112455327A (en) | 2020-12-03 | 2020-12-03 | Vehicle lamp control method and device and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011395286.XA CN112455327A (en) | 2020-12-03 | 2020-12-03 | Vehicle lamp control method and device and computer readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112455327A true CN112455327A (en) | 2021-03-09 |
Family
ID=74805367
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011395286.XA Pending CN112455327A (en) | 2020-12-03 | 2020-12-03 | Vehicle lamp control method and device and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112455327A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113609958A (en) * | 2021-08-02 | 2021-11-05 | 金茂智慧科技(广州)有限公司 | Light adjusting method and related device |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5031126B1 (en) * | 2011-04-07 | 2012-09-19 | パイオニア株式会社 | Ambient condition detection system for moving objects |
| CN105835761A (en) * | 2016-04-18 | 2016-08-10 | 乐视控股（北京）有限公司 | Method and device for controlling on and off of atmosphere lamps in car |
| CN107380064A (en) * | 2017-07-04 | 2017-11-24 | 上海青研科技有限公司 | A kind of vehicle-mounted Eye-controlling focus device based on augmented reality |
| CN107433898A (en) * | 2017-08-09 | 2017-12-05 | 上海科世达-华阳汽车电器有限公司 | A kind of control method, information processor and the system of car light direction |
| CN111079475A (en) * | 2018-10-19 | 2020-04-28 | 上海商汤智能科技有限公司 | Driving state detection method and device, driver monitoring system and vehicle |
| CN111137208A (en) * | 2019-12-20 | 2020-05-12 | 恒大新能源汽车科技(广东)有限公司 | Method, device and system for controlling illuminating lamp of storage box applied to automobile |
| CN112016457A (en) * | 2020-08-27 | 2020-12-01 | 青岛慕容信息科技有限公司 | Driver distraction and dangerous driving behavior recognition method, device and storage medium |
-
2020
- 2020-12-03 CN CN202011395286.XA patent/CN112455327A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5031126B1 (en) * | 2011-04-07 | 2012-09-19 | パイオニア株式会社 | Ambient condition detection system for moving objects |
| CN105835761A (en) * | 2016-04-18 | 2016-08-10 | 乐视控股（北京）有限公司 | Method and device for controlling on and off of atmosphere lamps in car |
| CN107380064A (en) * | 2017-07-04 | 2017-11-24 | 上海青研科技有限公司 | A kind of vehicle-mounted Eye-controlling focus device based on augmented reality |
| CN107433898A (en) * | 2017-08-09 | 2017-12-05 | 上海科世达-华阳汽车电器有限公司 | A kind of control method, information processor and the system of car light direction |
| CN111079475A (en) * | 2018-10-19 | 2020-04-28 | 上海商汤智能科技有限公司 | Driving state detection method and device, driver monitoring system and vehicle |
| CN111137208A (en) * | 2019-12-20 | 2020-05-12 | 恒大新能源汽车科技(广东)有限公司 | Method, device and system for controlling illuminating lamp of storage box applied to automobile |
| CN112016457A (en) * | 2020-08-27 | 2020-12-01 | 青岛慕容信息科技有限公司 | Driver distraction and dangerous driving behavior recognition method, device and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 李孟歆: "《基于机器视觉的图像识别技术及其应用》", 31 January 2019, 中国矿业大学出版社 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113609958A (en) * | 2021-08-02 | 2021-11-05 | 金茂智慧科技(广州)有限公司 | Light adjusting method and related device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10877485B1 (en) | Handling intersection navigation without traffic lights using computer vision | |
| CN109584507B (en) | Driving behavior monitoring method, device, system, vehicle and storage medium | |
| CN111931579B (en) | Automatic driving assistance system and method using eye tracking and gesture recognition techniques | |
| CN107323338B (en) | Vehicle steering lamp control system, control method and vehicle | |
| US10744936B1 (en) | Using camera data to automatically change the tint of transparent materials | |
| US10810966B1 (en) | Fusion of electronic mirror systems and driver monitoring for increased convenience and added safety | |
| US9418287B2 (en) | Object detection apparatus | |
| WO2021016873A1 (en) | Cascaded neural network-based attention detection method, computer device, and computer-readable storage medium | |
| CN105654753A (en) | Intelligent vehicle-mounted safe driving assistance method and system | |
| WO2021253953A1 (en) | Car window adjustment method and apparatus, electronic device, and readable storage medium | |
| CN104200192A (en) | Driver gaze detection system | |
| CN110781718B (en) | Cab infrared vision system and driver attention analysis method | |
| US10809801B1 (en) | Electronic mirror visual parameter adjustment method | |
| JP2013054735A (en) | Method for classifying eye closures | |
| CN110341617B (en) | Eyeball tracking method, device, vehicle and storage medium | |
| CN111267724B (en) | Method and system for protecting and reminding vehicle door opening | |
| CN109229009A (en) | Control method, device, the vehicles and the storage medium of vehicles headlamp | |
| CN113370901A (en) | Rearview mirror adjusting method, device, equipment and storage medium | |
| CN113661495A (en) | Sight line calibration method, sight line calibration device, sight line calibration equipment, sight line calibration system and sight line calibration vehicle | |
| CN112455327A (en) | Vehicle lamp control method and device and computer readable storage medium | |
| CN113306486A (en) | In-vehicle lighting device control method, storage medium, and electronic apparatus | |
| CN110949272A (en) | Vehicle-mounted display equipment adjusting method and device, vehicle, medium and equipment | |
| CN113140093A (en) | Fatigue driving detection method based on AdaBoost algorithm | |
| CN115424598A (en) | Display screen brightness adjusting method and device and storage medium | |
| Ribarić et al. | A neural-network-based system for monitoring driver fatigue |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210309 |
|
| RJ01 | Rejection of invention patent application after publication |