CN116039559A - Method and device for detecting the safety state of a seat safety device - Google Patents

Abstract

A method and device for identifying the safety state of a seat safety device. The method includes: obtaining vehicle state information; obtaining a seat image when the vehicle state information meets the preset first-level reminder condition; Whether there are passengers on the seat; when there are passengers on the seat, judge whether the vehicle status information meets the preset second-level reminder conditions; when the vehicle status information meets the preset second-level reminder conditions, obtain the seat belt buckle status Information; according to the seat image and seat belt buckle status information to determine whether the passenger has fastened the seat belt; when the passenger has not fastened the seat belt, a secondary reminder will be given. The method and device improve driving safety.

Description

Method and device for identifying safety state of seat safety device

technical field

The invention relates to the field of safe driving of automobiles, in particular to a method and a device for identifying the safety state of a seat safety device.

Background technique

The personal safety of drivers and passengers during vehicle driving is a problem that needs to be continuously solved.

Cars are generally equipped with safety belts, safety belt buckles and automatic alarm devices when drivers and passengers do not wear safety belts. However, many drivers and passengers do not wear seat belts properly due to various reasons and will take some measures to deceive the seat belt alarm device so that the alarm device will be exempted from the alarm. These bad behaviors or habits will bring serious safety hazards to drivers, passengers and other vehicles on the road. .

Therefore, it is necessary to adopt more accurate detection conditions to ensure that the drivers and passengers are correctly fastened their seat belts.

Contents of the invention

The technical problem solved by the present invention is to provide a method and device for identifying the safety state of a seat safety device, so as to improve the personal safety of drivers and passengers.

In order to solve the above technical problems, the technical solution of the present invention provides a method for identifying the safety status of a seat safety device, including: acquiring vehicle status information; when the vehicle status information meets the preset first-level reminder conditions, acquiring a seat image; Judging whether there are passengers on the seat according to the seat image; when there are passengers on the seat, judging whether the vehicle status information meets the preset second-level reminder conditions; when the vehicle status information meets the preset second-level reminder conditions When the seat belt buckle status information is obtained; according to the seat image and the seat belt buckle status information, it is judged whether the passenger has fastened the seat belt; when the passenger has not fastened the seat belt, a secondary reminder is given.

Optionally, the method for judging whether the passenger has fastened the seat belt according to the seat image and the status information of the seat belt buckle includes: judging whether the position of the seat belt satisfies the first preset condition according to the seat image; judging whether the seat belt buckle switch meets the second preset condition; when the position of the seat belt does not meet the first preset condition, or when the seat belt buckle switch does not meet the second preset condition, it is determined that the passenger Not wearing a seat belt.

Optionally, the first preset condition is that the seat belt is on the body of the passenger; the second preset condition is that the seat belt buckle switch is in a disconnected state.

Optionally, based on the deep learning algorithm model, it is judged according to the seat image whether the position of the safety belt satisfies the first preset condition.

Optionally, the condition of the second-level reminder and the way of the second-level reminder meet the safety belt not fastened reminder rule.

Optionally, the conditions for the secondary reminder include: the distance the vehicle has continuously traveled forward is greater than or equal to the preset distance, the speed of the vehicle forward is greater than or equal to the preset speed, and the time the vehicle has traveled forward is greater than or equal to A combination of one or more of the preset times.

Optionally, the preset distance is 300 meters, the preset speed is 15 kilometers per hour, and the preset time is 60 seconds.

Optionally, the secondary reminders include: visual reminders and auditory reminders; the duration of a single visual reminder and auditory reminder is greater than or equal to 30 seconds, and the interval between two adjacent reminders is less than or equal to 3 seconds.

Optionally, the method for judging whether there are passengers on the seat according to the seat image includes: using a deep learning algorithm to preprocess the seat image and extract features; Detecting to obtain a detection result; judging whether there is a portrait in the content of the seat image according to the detection result; when there is a portrait feature in the content of the seat image, it is judged that there is a passenger on the seat.

Optionally, the model of the deep learning algorithm includes a convolutional neural network model, and the convolutional neural network model includes an Inception Resnet v2 model.

Optionally, before using the deep learning algorithm to process and compare the seat images, the method further includes: using several seat images and portrait pictures to train the deep learning algorithm model.

Optionally, the condition of the first-level reminder includes: the vehicle is in a starting state.

Optionally, when the vehicle status information satisfies the preset condition of the first-level reminder but does not meet the preset condition of the second-level reminder, a first-level reminder is performed.

Optionally, the condition of the first-level reminder and the way of the first-level reminder meet the seat belt not fastened reminder rule; the way of the first-level reminder includes a visual reminder, and the duration of a single reminder is greater than or equal to 60 seconds.

Optionally, in the process of performing the first-level reminder, if the vehicle state information satisfies the preset condition of the second-level reminder, the method further includes: switching the first-level reminder to a second-level reminder.

Correspondingly, the technical solution of the present invention also provides a device for identifying the safety status of a seat safety device, including: a vehicle status information judging unit for acquiring vehicle status information; a camera for when the vehicle status information satisfies a preset level When the reminder condition is met, obtain the seat image; the main controller unit is used to judge whether there is a passenger on the seat according to the seat image; and, when there is a passenger on the seat, it is used to judge whether the vehicle state information meets the predetermined The condition of the second-level reminder; the seat belt buckle switch is used to send the seat belt buckle status information when the vehicle status information meets the preset second-level reminder condition; the main controller unit also It is used to obtain the status information of the seat belt buckle, and judge whether the passenger has fastened the seat belt according to the seat image and the status information of the seat belt buckle; and, when the passenger has not fastened the seat belt, it is used to send a secondary reminder signal; the instrument and a loudspeaker for obtaining a secondary reminder signal for a secondary reminder.

Optionally, the main controller unit is further configured to send a primary alert signal when the vehicle state information meets the preset condition of the primary alert but does not meet the preset condition of the secondary alert.

Optionally, the instrument is also used to obtain a first-level reminder signal for a first-level reminder; the way of the first-level reminder includes a visual reminder, and the visual reminder includes an indicator light flashing, and the duration of a single reminder is greater than or equal to 60 seconds.

Optionally, the way of the secondary reminder includes visual reminder and auditory reminder, the visual reminder includes the flashing of the indicator light on the instrument, the auditory reminder includes the alarm bell played by the speaker, the continuous duration of a single visual reminder and auditory reminder The time is greater than or equal to 30 seconds, and the interval between two adjacent reminders is less than or equal to 3 seconds.

Optionally, the main controller unit is further configured to judge whether the vehicle state information satisfies the preset first-level reminder condition according to the vehicle state information.

Optionally, the main controller unit and the vehicle state information judging unit are located on the same on-board chip, or on different on-board chips.

Optionally, the main controller unit is electrically connected to the vehicle state information judging unit through a bus, the main controller unit is electrically connected to the camera through a bus, and the main controller unit is electrically connected to the seat belt buckle switch through a bus , the main controller unit is electrically connected to the meter and the speaker through a bus.

Optionally, the main controller unit is wirelessly connected to the vehicle state information judging unit, the main controller unit is wirelessly connected to the camera, the main controller unit is wirelessly connected to the seat belt buckle switch, and the main controller The unit connects wirelessly with gauges and speakers.

Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

In the method of the technical solution of the present invention, by judging whether there is a passenger on the seat according to the seat image, and judging whether the passenger is wearing a seat belt according to the seat image, multi-conditions are used to determine whether there is a passenger on the seat and whether the passenger is wearing a seat belt. Accurate identification reduces the risk of misjudgment. Improved driving safety.

Further, based on the deep learning algorithm, it can accurately identify whether there are passengers on the seat and whether the passenger on the seat is wearing a seat belt, reducing the risk of misjudging the situation of placing heavy objects on the seat; according to the seat image The two conditions for judging whether the passenger has fastened the seat belt with the status information of the seat belt buckle are used to judge whether the passenger has successfully fastened the seat belt, which solves the risk of passengers installing a third-party buckle to eliminate the reminder that the seat belt is not fastened, and improves driving safety.

According to the device of the technical solution of the present invention, the main controller unit judges whether there is a passenger on the seat according to the seat image, and judges whether the passenger has fastened the seat belt according to the seat image, and uses multiple conditions to determine whether there is a passenger on the seat. Accurate identification of passengers and whether they are wearing seat belts, thereby reducing the risk of misjudgment. Improved driving safety.

Description of drawings

1 to 3 are flowcharts of a method for identifying a safety state of a seat safety device in an embodiment of the present invention;

Fig. 4 is a schematic structural diagram of a device for identifying a safety state of a seat safety device in an embodiment of the present invention.

Detailed ways

As mentioned in the background art, more accurate detection conditions need to be adopted to ensure that the driver and passengers correctly fasten the seat belts.

Specifically, the vehicle usually uses a gravity sensor or a switch to determine whether there are passengers in the seat. When the vehicle reaches a certain driving condition (the first-level condition and the second-level condition required by the law), it will remind the passengers who are not wearing seat belts.

However, the gravity sensor or switch on the vehicle is only based on the weight on the seat exceeding a certain threshold as a judgment condition, and cannot identify whether the seat is a passenger or an object, so false alarms may occur. In addition, after a passenger installs a third-party buckle on the seat belt buckle, it is considered that the passenger has fastened the seat belt, thereby greatly increasing traffic safety hazards.

In order to solve the above problems, the technical solution of the present invention provides a method and device for identifying the safety status of a seat safety device, by judging whether there is a passenger on the seat according to the seat image, and judging whether the passenger is wearing a seat belt according to the seat image , using multiple conditions to accurately identify whether there are passengers on the seat and whether the passengers are wearing seat belts, thereby reducing the risk of misjudgment. Improved driving safety.

In order to make the above objects, features and beneficial effects of the present invention more comprehensible, specific embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings.

1 to 3 are flowcharts of a method for identifying a safety state of a seat safety device in an embodiment of the present invention.

Please refer to Figure 1, the method for identifying the safety state of the seat safety device includes:

Step S1: Obtain vehicle status information;

Step S2: When the vehicle state information meets the preset condition of the first-level reminder, acquire the seat image;

Step S3: judge whether there is a passenger on the seat according to the seat image;

Step S4: When there are passengers on the seat, it is judged whether the vehicle state information satisfies the preset condition of the secondary reminder;

Step S5: When the vehicle state information meets the preset condition of the second-level reminder, acquire the seat belt buckle state information;

Step S6: Determine whether the passenger has fastened the seat belt according to the seat image and the status information of the seat belt buckle;

Step S7: When the passenger has not fastened the seat belt, a secondary reminder is performed.

The method judges whether there is a passenger on the seat according to the seat image, and judges whether the passenger is wearing a seat belt according to the seat image, and uses multiple conditions to accurately identify whether there is a passenger on the seat and whether the passenger is wearing a seat belt, thereby The risk of misjudgment is reduced. Improved driving safety.

Next, each step will be described separately.

Please continue to refer to FIG. 1 and execute step S1: acquire vehicle status information.

The vehicle state information refers to the real-time state of the vehicle, for example, the vehicle is in the starting state, stopped state or driving state, and when the vehicle is in the driving state, the vehicle's driving speed, driving distance or driving time, etc.

In this embodiment, the vehicle state information is acquired by the vehicle state information judging unit. The vehicle state information judging unit is a functional module on the vehicle chip.

Please continue to refer to FIG. 1 , and perform step S2: when the vehicle state information meets the preset condition of the first-level reminder, acquire the seat image.

In this embodiment, after the vehicle state information judging unit obtains the vehicle state information, the main controller unit obtains the vehicle state information from the vehicle state information judging unit, and judges the vehicle state information to determine whether the vehicle state information satisfies a preset condition. Conditions for level alerts. The main controller unit is a separate vehicle-mounted chip or a functional module on the vehicle-mounted chip.

The condition of the first-level reminder satisfies the safety belt not fastened reminder rule.

In this embodiment, the condition of the first-level reminder includes: the vehicle is in a starting state.

In this embodiment, when the vehicle state information satisfies the preset condition of the first-level reminder, the seat image is acquired by the camera, and the image information of the vehicle seat is collected by the camera in real time and transmitted to the main controller unit.

The camera can also be used to collect images of passengers for monitoring and analysis of passenger status, and can also be used to collect facial images of passengers for monitoring and analysis of occupant fatigue.

The camera for collecting the image information of the vehicle seat in real time can be shared with the camera of the occupant fatigue monitoring system or the driver fatigue monitoring system, and the traditional gravity sensor or switch can be unnecessary, thus reducing the cost of the whole vehicle.

Please continue to refer to FIG. 1 and execute step S3: judge whether there is a passenger on the seat according to the seat image.

Please refer to Fig. 2, in this embodiment, the method for judging whether there is a passenger on the seat according to the seat image includes:

Step S31: Using a deep learning algorithm to preprocess and extract features of the seat image;

Step S32: Detecting the seat image based on a deep learning algorithm model to obtain a detection result;

Step S33: According to the detection result, it is judged whether there is a portrait in the content of the seat image;

Step S34: When there is a portrait feature in the content of the seat image, it is determined that there are passengers on the seat.

Based on the deep learning algorithm, it can accurately identify whether there are passengers on the seat, which reduces the risk of misjudging the situation of placing heavy objects on the seat.

Next, analyze and explain each step.

Before performing step S31: using a deep learning algorithm to process and compare the seat images, it also includes performing step S35: using several seat images and portrait pictures to train the deep learning algorithm model.

The seat images and portrait pictures are pictures in the picture library.

A number of seat images and portrait pictures are used to train the deep learning algorithm model to obtain a feature database, and when the preset recognition accuracy rate is reached, a deep learning algorithm model that meets the requirements is obtained. The deep learning algorithm model needs to identify seat features and portrait features on the seat.

In this embodiment, the model of the deep learning algorithm includes a convolutional neural network model, and the convolutional neural network model includes an Inception Resnet v2 model.

Please continue to refer to FIG. 2 , and perform step S31 : perform preprocessing and feature extraction on the seat image by using a deep learning algorithm.

The features include portrait features and seat features.

Please continue to refer to FIG. 2 and perform step S32: detect the seat image based on a deep learning algorithm model to obtain a detection result.

The main controller unit uses a deep learning algorithm model to compare and analyze the features of the portrait and the seat with the features of the feature database, and obtain a detection result. The detection result is whether the seat image has both the seat feature and the feature database. portrait features.

Please continue to refer to FIG. 2 , and execute step S33 : judge whether there is a portrait in the content of the seat image according to the detection result.

Please continue to refer to FIG. 2 and execute step S34: when there is a portrait feature in the content of the seat image, it is determined that there are passengers on the seat.

In the detection result, when there is a portrait feature in the content of the seat image, the main controller unit judges that there are passengers on the seat.

Please continue to refer to FIG. 1 and execute step S4: when there are passengers on the seat, determine whether the vehicle status information satisfies the preset condition of the second-level reminder.

The condition of the secondary reminder satisfies the safety belt not fastened reminder rule.

The conditions for the secondary reminder include: the distance the vehicle travels forward continuously is greater than or equal to the preset distance, the speed of the vehicle driving forward is greater than or equal to the preset speed, and the time the vehicle travels forward is greater than or equal to the preset time A combination of one or more of .

In this embodiment, the preset distance is 300 meters, the preset speed is 15 kilometers per hour, and the preset time is 60 seconds.

When the vehicle status information satisfies the condition of the second-level reminder, it means that the vehicle is in the driving state at this time, and it is necessary to judge in time whether to fasten the seat belt in order to improve driving safety.

In other embodiments, when the seat belt not fastening reminder rule changes, the condition of the secondary reminder also changes accordingly, so as to meet the seat belt not fasten reminder rule.

In this embodiment, when there are passengers on the seat, the main controller unit judges whether the vehicle status information satisfies the preset condition of the secondary reminder.

Please continue to refer to FIG. 1 , and perform step S5: when the vehicle status information meets the preset condition of the second-level reminder, acquire the seat belt buckle status information.

When the main controller unit judges that the vehicle status information meets the preset condition of the second-level reminder, the main controller unit obtains the status information of the seat belt buckle from the seat belt buckle switch. The seat belt buckle state information is whether the seat belt buckle switch is in a disconnected state or in a connected state.

The seat belt buckle switch is in the connected state when the seat belt is not fastened, and the seat belt buckle switch is in the disconnected state after the seat belt buckle is correctly fastened.

Please continue to refer to FIG. 1 and execute step S6: judge whether the passenger has fastened the seat belt according to the seat image and the status information of the seat belt buckle.

Please refer to FIG. 3. In this embodiment, the method for judging whether a passenger has fastened a seat belt according to the seat image and seat belt buckle status information includes:

Step S61: judging whether the position of the seat belt satisfies the first preset condition according to the seat image;

Step S62: judging whether the seat belt buckle switch satisfies the second preset condition according to the status information of the seat belt buckle;

Step S63: When the position of the seat belt does not meet the first preset condition, or when the seat belt buckle switch does not meet the second preset condition, it is determined that the passenger has not fastened the seat belt.

Judging whether the passenger has fastened the seat belt according to the seat image and the status information of the seat belt buckle is used to judge whether the passenger has successfully fastened the seat belt. driving safety.

Next, each step will be described separately.

Please continue to refer to FIG. 3 , step S61 is executed: judging whether the position of the seat belt satisfies the first preset condition according to the seat image.

The main controller unit analyzes the position feature of the seat belt in the seat image based on a deep learning algorithm, and judges whether the position of the seat belt satisfies the first preset condition.

Based on the deep learning algorithm, it can accurately identify whether the passengers on the seat are fastened with seat belts, which reduces the risk of misjudging the situation of placing heavy objects on the seat.

In this embodiment, the first preset condition is that the seat belt is on the body of the passenger.

Please continue to refer to FIG. 3 , step S62 is executed: judging whether the seat belt buckle switch satisfies the second preset condition according to the seat belt buckle state information.

The main controller unit judges whether the seat belt buckle switch satisfies the second preset condition according to the state information of the seat belt buckle.

In this embodiment, the second preset condition is that the seat belt buckle switch is in an off state.

Please continue to refer to FIG. 3 and execute step S63: when the position of the seat belt does not meet the first preset condition, or when the seat belt buckle switch does not meet the second preset condition, it is determined that the passenger is not fastened. bring.

When the main controller unit judges that the seat belt is on the passenger and the seat belt buckle switch is in the off state, it means that the passenger has fastened the seat belt in accordance with regulations, and there is no need for an alarm reminder.

If the main controller unit judges that the seat belt is not located on the passenger, or when the seat belt buckle switch is connected, it means that the passenger has not properly fastened the seat belt and there is a potential safety hazard, and an alarm is required in time .

Please continue to refer to FIG. 1 and execute step S7: when the passenger does not fasten the seat belt, perform a secondary reminder.

The manner of the secondary reminder satisfies the safety belt not fastened reminder rule.

In this embodiment, the two-level reminders include: visual reminders and auditory reminders; the duration of a single visual reminder and auditory reminder is greater than or equal to 30 seconds, and the interval between two adjacent reminders is less than or equal to 3 seconds. Second.

The visual reminder includes the flashing of the indicator light on the instrument, and the auditory reminder includes the alarm bell played by the speaker. The single duration of the flashing indicator light and the alarm bell played by the speaker is greater than or equal to 30 seconds. The interval between the flashing and the alarm tone played by the speaker is less than or equal to 3 seconds.

When the main controller unit judges that the passenger has not fastened the seat belt, it sends the information of the secondary reminder to the speaker and the instrument. After the speaker and the instrument receive the information of the secondary reminder, the speaker plays the alarm bell and the indicator light flashes for The second-level reminder enables passengers to fasten their seat belts in a standard manner, thereby improving driving safety.

In another embodiment, when the vehicle state information satisfies the preset condition of the first-level reminder but does not meet the preset condition of the second-level reminder, a first-level reminder is performed.

When the main controller unit judges that the vehicle state information meets the preset first-level reminder condition, that is, the vehicle is in the starting state, and judges that the vehicle state information does not meet the preset second-level reminder condition, that is, when the vehicle is in the stopped state, this Only need to send a first-level reminder signal to the instrument for a first-level reminder.

The conditions of the first-level reminder and the way of the first-level reminder meet the safety belt not fastened reminder rule; the way of the first-level reminder includes a visual reminder, and the duration of a single reminder is greater than or equal to 60 seconds.

The visual reminder includes the blinking of the indicator light on the instrument, and the duration of a single blink of the indicator light is greater than or equal to 60 seconds.

In another embodiment, if the vehicle state information satisfies the preset condition of the second-level reminder during the first-level reminder, it further includes: switching the first-level reminder to the second-level reminder.

In the process of carrying out the first-level reminder, when the main controller unit judges that the vehicle state information meets the preset conditions of the second-level reminder, it sends a second-level reminder signal to the instrument and the loudspeaker for the second-level reminder, and the second-level reminder replaces the second-level reminder. The above-mentioned first-level reminder.

Fig. 4 is a schematic structural diagram of a device for identifying a safety state of a seat safety device in an embodiment of the present invention.

Please refer to Figure 4, the device for identifying the safety state of the seat safety device includes:

A vehicle state information judging unit 10, configured to obtain vehicle state information;

The camera 20 is used to acquire the seat image when the vehicle status information meets the preset first-level reminder condition;

The main controller unit 30 is used to judge whether there is a passenger on the seat according to the seat image; and, when there is a passenger on the seat, is used to judge whether the vehicle state information meets the preset condition of the secondary reminder;

The seat belt buckle switch 40 is used to send the seat belt buckle status information when the vehicle state information meets the preset secondary reminder condition;

The main controller unit 30 is also used to obtain seat belt buckle status information, and judge whether the passenger has fastened the seat belt according to the seat image and the seat belt buckle status information; and, when the passenger has not fastened the seat belt, for Send a secondary reminder signal;

The instrument 50 and the loudspeaker 60 are used to obtain a secondary reminder signal for secondary reminder.

The main controller unit judges whether there is a passenger on the seat according to the seat image, and judges whether the passenger is wearing a seat belt according to the seat image, and uses multiple conditions to accurately determine whether there is a passenger on the seat and whether the passenger is wearing a seat belt. identification, thereby reducing the risk of misjudgment. Improved driving safety.

Next, each configuration will be described separately.

Please continue to refer to FIG. 4 , the vehicle state information judging unit 10 is configured to acquire vehicle state information.

The vehicle state information judging unit 10 is a functional module on a vehicle chip.

The vehicle state information refers to the real-time state of the vehicle, for example, the vehicle is in the starting state, stopped state or driving state, and when the vehicle is in the driving state, the vehicle's driving speed, driving distance or driving time, etc.

Please continue to refer to FIG. 4 , the camera 20 is used to acquire the seat image when the vehicle status information meets the preset condition of the first-level reminder.

In this embodiment, the main controller unit 30 is further configured to judge whether the vehicle state information satisfies the preset primary alert condition according to the vehicle state information.

The camera can also be used to collect images of passengers for monitoring and analysis of passenger status, and can also be used to collect facial images of passengers for monitoring and analysis of occupant fatigue.

The camera for collecting the image information of the vehicle seat in real time can be shared with the camera of the occupant fatigue monitoring system or the driver fatigue monitoring system, and the traditional gravity sensor or switch can be unnecessary, thus reducing the cost of the whole vehicle.

Please continue to refer to FIG. 4, the main controller unit 30 is used to judge whether there is a passenger on the seat according to the seat image; and, when there is a passenger on the seat, to judge whether the vehicle state information meets the preset The condition of the secondary reminder.

The main controller unit 30 accurately recognizes whether there are passengers on the seat based on the deep learning algorithm, which reduces the risk of misjudging the situation of placing heavy objects on the seat.

The main controller unit 30 judges whether the vehicle state information satisfies the preset condition of the second-level reminder, and obtains the seat belt buckle status information when the preset condition of the second-level reminder is met.

The main controller unit 30 and the vehicle state information judging unit 10 are located on the same on-board chip, or on different on-board chips.

Please continue to refer to FIG. 4 , the seat belt buckle switch 40 is configured to send the seat belt buckle status information when the vehicle status information meets the preset second-level reminder condition.

When the main controller unit 30 judges that the vehicle state information meets the preset condition of the second-level reminder, the main controller unit obtains the seat belt buckle state information from the seat belt buckle switch 40 . The seat belt buckle state information is whether the seat belt buckle switch 40 is in the off state or in the on state.

Please continue to refer to Fig. 4, in this embodiment, the main controller unit 30 is also used to judge whether the passenger has fastened the seat belt according to the seat image and the status information of the seat belt buckle; and, when the passenger does not fasten the seat belt With time, it is used to send the secondary reminder signal.

If the main controller unit 30 judges that the seat belt is not located on the passenger, or when the seat belt buckle switch is in the connected state, it means that the passenger has not fastened the seat belt properly, and there is a potential safety hazard, and an alarm needs to be issued in time remind.

The main controller unit 30 judges whether the passenger has fastened the seat belt according to the seat image and the status information of the seat belt buckle to determine whether the passenger has successfully fastened the seat belt. Reminds you of risks and improves driving safety.

Please continue to refer to FIG. 4 , the meter 50 and the speaker 60 are used to obtain a secondary reminder signal for a secondary reminder.

After the instrument 50 and the speaker 60 obtain the secondary reminder signal from the autonomous control unit 30, they perform a secondary reminder.

The mode of the secondary reminder includes visual reminder and auditory reminder. The visual reminder includes the flashing of the indicator light on the instrument. The auditory reminder includes the alarm bell played by the speaker. The duration of a single visual reminder and auditory reminder is greater than or It is equal to 30 seconds, and the interval between two adjacent reminders is less than or equal to 3 seconds.

Please continue to refer to FIG. 4. In another embodiment, the main controller unit 30 is also configured to: Send a first-level reminder signal.

Please continue to refer to FIG. 4 , in another embodiment, the instrument 50 is also used to obtain a primary reminder signal to perform a primary reminder; the way of the primary reminder includes a visual reminder, and the visual reminder includes an indicator light flashing, The duration of a single reminder is greater than or equal to 60 seconds.

Please continue to refer to FIG. 4 , in this embodiment, the main controller unit 30 is electrically connected to the vehicle state information judging unit 10 through a bus, and the main controller unit 30 is electrically connected to the camera 20 through a bus. The controller unit 30 is electrically connected to the seat belt buckle switch 40 through the bus, and the main controller unit 30 is electrically connected to the instrument 50 and the speaker 60 through the bus.

In another embodiment, the main controller unit is wirelessly connected to the vehicle state information judging unit, the main controller unit is wirelessly connected to the camera, and the main controller unit is wirelessly connected to the seat belt buckle switch. The main controller unit is wirelessly connected to the gauges and speakers.

Although the present invention is disclosed above, the present invention is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, so the protection scope of the present invention should be based on the scope defined in the claims.

Claims (23)

1. A method of identifying a safety condition of a seat safety device, comprising:

acquiring vehicle state information;

when the vehicle state information meets the preset first-level reminding condition, acquiring a seat image;

judging whether passengers sit on the seats according to the seat images;

when a passenger sits on the seat, judging whether the vehicle state information meets the preset secondary reminding condition;

when the vehicle state information meets the preset secondary reminding condition, acquiring the safety belt lock catch state information;

judging whether a passenger ties a safety belt according to the seat image and the safety belt locking state information;

and when the passenger is not belted with the safety belt, carrying out secondary reminding.

2. The method of identifying a safety condition of a seat safety device according to claim 1, wherein the method of determining whether a passenger is belted based on the seat image and the belt buckle status information comprises: judging whether the position of the safety belt meets a first preset condition according to the seat image; judging whether the safety belt lock switch meets a second preset condition according to the safety belt lock state information; and when the position of the safety belt does not meet the first preset condition or the safety belt lock switch does not meet the second preset condition, judging that the passenger does not tie the safety belt.

3. The method of identifying a safety condition of a seat safety device according to claim 2, wherein the first predetermined condition is that the seat belt is located on a passenger; the second preset condition is that the safety belt lock catch switch is in an off state.

4. A method of identifying a safety condition of a seat safety device as claimed in claim 3 wherein the determination of whether the position of the seat belt meets a first predetermined condition is based on a model of a deep learning algorithm based on the seat image.

5. A method of identifying a safety condition of a seat safety device as claimed in claim 1 wherein the secondary alert condition and the secondary alert manner satisfy a seat belt unbuckled alert rule.

6. The method of identifying a safety condition of a seat safety device according to claim 5, wherein the secondary alert condition comprises: the distance that the vehicle continuously travels forward is greater than or equal to a combination of one or more of a preset distance, a speed that the vehicle travels forward is greater than or equal to a preset speed, and a time that the vehicle travels forward is greater than or equal to a preset time.

7. The method of identifying a safety condition of a seat safety device according to claim 6, wherein the predetermined distance is 300 meters, the predetermined speed is 15 km/h, and the predetermined time is 60 seconds.

8. The method of identifying a safety condition of a seat safety device according to claim 5, wherein the secondary alert comprises: visual and audible reminders; the duration of the single visual reminding and the audible reminding is greater than or equal to 30 seconds, and the interval time between two adjacent reminding is less than or equal to 3 seconds.

9. The method of identifying a safety condition of a seat safety device according to claim 1, wherein the method of determining whether a passenger is seated on the seat based on the seat image comprises: preprocessing the seat image and extracting the characteristics by adopting a deep learning algorithm; detecting the seat image based on a deep learning algorithm model to obtain a detection result; judging whether a person exists in the content of the seat image according to the detection result; and when the image content of the seat has the portrait character, judging that the seat has the passenger.

10. The method of identifying a safety condition of a seat safety device according to claim 9, wherein the model of the deep learning algorithm comprises a convolutional neural network model comprising a Inception Resnet v model.

11. The method of identifying a safety condition of a seat safety device according to claim 9, further comprising, prior to processing and comparing the seat images using a deep learning algorithm: and training the deep learning algorithm model by adopting a plurality of seat images and portrait pictures.

12. The method of identifying a safety condition of a seat safety device according to claim 1, wherein the primary alert condition comprises: the vehicle is in a start state.

13. The method of claim 1, wherein the first-level reminder is performed when the vehicle state information satisfies a preset first-level reminder condition and does not satisfy a preset second-level reminder condition.

14. The method of identifying a safety condition of a seat safety device according to claim 13, wherein the primary alert condition and primary alert manner satisfy a seat belt unbuckled alert rule; the primary reminding mode comprises visual reminding, and the duration of single reminding is greater than or equal to 60 seconds.

15. The method of claim 13, further comprising, during the primary alert, if the vehicle status information satisfies a preset secondary alert condition: and switching the primary reminding to the secondary reminding.

16. An apparatus for identifying a safety condition of a seat safety device, comprising:

a vehicle state information judgment unit configured to acquire vehicle state information;

the camera is used for acquiring a seat image when the vehicle state information meets the preset primary reminding condition;

a main controller unit for judging whether a passenger sits on the seat according to the seat image; when a passenger sits on the seat, the vehicle state information is used for judging whether the vehicle state information meets the preset secondary reminding condition or not;

the safety belt lock catch switch is used for sending safety belt lock catch state information when the vehicle state information meets the preset secondary reminding condition;

the main controller unit is also used for acquiring the safety belt locking state information and judging whether a passenger ties a safety belt or not according to the seat image and the safety belt locking state information; when the passenger is not belted with the safety belt, the safety belt is used for sending a secondary reminding signal;

and the instrument and the loudspeaker are used for acquiring the secondary reminding signal to carry out secondary reminding.

17. The apparatus for identifying a safety condition of a seat safety device according to claim 16, wherein the main controller unit is further configured to send a primary alert signal when the vehicle status information satisfies a preset primary alert condition and does not satisfy a preset secondary alert condition.

18. The apparatus for identifying a safety condition of a seat safety device according to claim 17, wherein the meter is further configured to obtain a primary alert signal for primary alert; the primary reminding mode comprises visual reminding, wherein the visual reminding comprises flashing of an indicator light, and the duration of single reminding is greater than or equal to 60 seconds.

19. The apparatus for identifying a safety condition of a seat safety device according to claim 16 wherein the secondary alert means comprises a visual alert comprising a flashing indicator on the meter and an audible alert comprising an alarm bell played by a speaker, the duration of a single visual alert and audible alert being greater than or equal to 30 seconds, the interval between adjacent alerts being less than or equal to 3 seconds.

20. The apparatus for identifying a safety condition of a seat safety device according to claim 16, wherein the main controller unit is further configured to determine whether the vehicle status information satisfies a preset primary alert condition based on the vehicle status information.

21. The apparatus for recognizing the safety condition of a seat safety apparatus according to claim 16, wherein the main controller unit and the vehicle condition information judging unit are located on the same vehicle-mounted chip or on different vehicle-mounted chips.

22. The apparatus for identifying a safety state of a seat safety device according to claim 16, wherein the main controller unit is electrically connected to the vehicle state information judgment unit via a bus, the main controller unit is electrically connected to the camera via a bus, the main controller unit is electrically connected to the belt lock switch via a bus, and the main controller unit is electrically connected to the meter and the speaker via a bus.

23. The apparatus for identifying a safety state of a seat safety device according to claim 16, wherein the main controller unit is wirelessly connected to the vehicle state information judgment unit, the main controller unit is wirelessly connected to the camera, the main controller unit is wirelessly connected to the belt buckle switch, and the main controller unit is wirelessly connected to the meter and the speaker.

Images