KR20240011931A - Occupant classification method and apparatus for vehicle - Google Patents

Abstract

Obtaining images of the interior of the vehicle; Obtaining a seating position; determining a sheet area value; Detecting a seated person and determining a seated area value corresponding to the detected seated person; and determining the class of the seated person based on the area ratio of the seat area value and the seated person area value of the corresponding seat. A method and device for classifying a seated person for a vehicle including a step is introduced.

Description

Vehicle seat classification method and device {OCCUPANT CLASSIFICATION METHOD AND APPARATUS FOR VEHICLE}

The present invention relates to a method and device for classifying vehicle occupants, and more specifically, to a method and device for classifying vehicle occupants based on image information.

As vehicle-related technologies continue to develop, today's vehicles are equipped with a variety of functions, including basic driving functions, functions related to the safety of passengers, and functions related to convenience. Additionally, in order to properly apply these functions to prevent malfunctions and provide improved performance, technologies for detecting or classifying the occupants of vehicles subject to each function have been developed.

An example of such occupant detection or classification technology is SBR (Seat Belt Reminder) technology to encourage the wearing of seat belts. SBR technology is a technology that determines whether the passenger is wearing a seat belt and encourages the seat belt to be worn through notifications when not wearing it. For this purpose, SBR technology uses weight information to determine whether a passenger is on board. More specifically, the load applied is detected through a switch sensor attached to the seating surface at the top of the seat, and this is transmitted to the controller to determine whether or not to ride. However, since this type of SBR technology can generally only detect the presence or absence of a load, there are limitations in that it cannot distinguish between people and objects and cannot classify occupants.

Another example is ODS (Occupant Detection System) technology, which attaches a structure to the seating surface at the top of the seat to determine whether a passenger is on board and classifies the type of occupant according to the applied load. The ODS transmits occupant information to a separate controller, allowing airbag deployment to be suppressed if the occupant is an infant. Unlike the aforementioned SBR, ODS technology can detect the degree of load, making it possible to classify occupants according to the degree of load. However, like the aforementioned SBR technology, ODS technology uses only weight information, so there is a limitation in being unable to distinguish between people and objects.

Accordingly, there is a need to propose a method to detect passengers more accurately and classify them in detail.

The matters described as background technology above are only for the purpose of improving understanding of the background of the present invention, and should not be taken as recognition that they correspond to prior art already known to those skilled in the art.

KR 10-2006-0003193 A

The present invention relates to a method and device for classifying seated occupants for a vehicle that allows more accurate and detailed classification of seated occupants based on image information.

The technical problems to be achieved in the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the description below. You will be able to.

In order to solve the above technical problem, a method for classifying seat occupants for a vehicle according to an embodiment of the present invention includes the steps of acquiring an image of the interior of the vehicle; Obtaining a seating position of a seat; determining a seat area value based on parameters associated with the seat position; Detecting a seated person from the acquired image and determining a seated area value corresponding to the detected seated person; and determining the class of the seated person based on an area ratio according to the seat area value and the seated person's area value of the corresponding seat.

For example, the image may be an image captured by an in-cabin camera installed at the front of the vehicle, facing rearward.

For example, the seat position includes the front and rear positions of the seat, and the step of determining the seat area value includes at least the seat area based on parameters preset to correspond to the frontmost and rearmost positions of the seat. It may include a step of determining a value.

For example, the step of acquiring the seat position may include obtaining a seat position according to the front-to-back position of the seat from a seat track position sensor connected to the seat.

For example, the seat position may include a recline angle of the seatback of the seat, and the step of determining the seat area value may include determining the seat area value by additionally considering the recline angle. You can.

For example, if the seated person cannot be detected from the acquired image, determining the class as a specific class may be further included.

For example, determining the area value of the seated person may include obtaining segmentation of the seated person from the acquired image; and determining the seat area value based on the obtained seat segmentation.

For example, the step of obtaining the seated person segment may include obtaining the seated person segment from the acquired image by inputting the acquired image into a predetermined artificial intelligence learning model.

For example, the artificial intelligence learning model is an artificial intelligence algorithm and may include at least one of machine learning, neural network, deep learning, classification algorithm, or a combination thereof.

For example, the step of determining the class of the seated person may include comparing the area ratio with a preset threshold and determining the class of the seated person based on the comparison result.

For example, the step of determining the class of the seated person may include determining the class of the seated person through a plurality of class rules defined as a range between two adjacent threshold values among a plurality of preset thresholds. You can.

For example, the step of obtaining weight information of the seated person may further include, and the step of determining the class of the seated person may include determining the class of the seated person by further considering the weight information. .

For example, the step of acquiring the weight information of the seated person may include obtaining the weight information of the seated person from a weight sensor connected to the seat.

For example, determining the class of the seated person may include determining the area ratio for each of a plurality of seats; and determining the class of the seated person for each of the plurality of seats based on the determined area ratios.

In order to solve the above technical problem, a vehicle occupant classification device according to an embodiment of the present invention includes at least one seat; Cameras that film the interior of a vehicle; and acquire the seat position of the target seat among the at least one seat, determine the seat area value based on the parameter associated with the seat position, and detect the seated person from the image inside the vehicle obtained through the camera, and detect the seated person. It includes a controller that determines the seat area value corresponding to the seated person and determines the class of the seated person based on the area ratio of the seat area value and the seated area value of the target seat.

The object of the present invention is not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below.

The method for classifying seated occupants in a vehicle according to at least one embodiment of the present invention configured as described above can more accurately detect and classify seated occupants in detail by utilizing image information.

By improving the accuracy of seated occupant detection, problems such as malfunctioning of functions installed in the vehicle or failure to perform as intended due to incorrect detection can be alleviated.

Furthermore, through further refinement of the classification of occupants, each function mounted on the vehicle can be applied to more specifically reflect the characteristics of each occupant, thereby improving performance.

The effects that can be obtained from the present invention are not limited to the effects mentioned above, and other effects not mentioned can be clearly understood by those skilled in the art from the description below. will be.

1 is a diagram showing the configuration of a vehicle seat classification device according to an embodiment of the present invention.
Figure 2 is a diagram representing an image of the interior of a vehicle including a seat according to an embodiment of the present invention.
Figure 3 is a diagram showing determining the area value of a seated person in classifying a seated person according to an embodiment of the present invention.
Figure 4 is a diagram showing classification of seated person classes based on area ratio and threshold value in classifying seated persons according to an embodiment of the present invention.
Figure 5 is a flowchart of a method for classifying seat occupants for a vehicle according to an embodiment of the present invention.

Specific structural and functional descriptions of the embodiments of the present invention disclosed in the present specification or application are merely illustrative for the purpose of explaining the embodiments according to the present invention, and the embodiments according to the present invention are implemented in various forms. It may be possible and should not be construed as being limited to the embodiments described in this specification or application.

Since the embodiments according to the present invention can make various changes and have various forms, specific embodiments will be illustrated in the drawings and described in detail in the specification or application. However, this is not intended to limit the embodiments according to the concept of the present invention to a specific disclosed form, and should be understood to include all changes, equivalents, and substitutes included in the spirit and technical scope of the present invention.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the technical field to which the present invention pertains. Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related technology, and unless clearly defined in this specification, should not be interpreted as having an ideal or excessively formal meaning. .

Hereinafter, embodiments disclosed in the present specification will be described in detail with reference to the attached drawings. However, identical or similar components will be assigned the same reference numbers regardless of reference numerals, and duplicate descriptions thereof will be omitted.

The suffixes “module” and “part” for components used in the following description are given or used interchangeably only for the ease of preparing the specification, and do not have distinct meanings or roles in themselves.

In describing the embodiments disclosed in this specification, if it is determined that detailed descriptions of related known technologies may obscure the gist of the embodiments disclosed in this specification, the detailed descriptions will be omitted. In addition, the attached drawings are only for easy understanding of the embodiments disclosed in this specification, and the technical idea disclosed in this specification is not limited by the attached drawings, and all changes included in the spirit and technical scope of the present invention are not limited. , should be understood to include equivalents or substitutes.

Terms containing ordinal numbers, such as first, second, etc., may be used to describe various components, but the components are not limited by the terms. The above terms are used only for the purpose of distinguishing one component from another.

When a component is said to be "connected" or "connected" to another component, it is understood that it may be directly connected to or connected to the other component, but that other components may exist in between. It should be. On the other hand, when it is mentioned that a component is “directly connected” or “directly connected” to another component, it should be understood that there are no other components in between.

Singular expressions include plural expressions unless the context clearly dictates otherwise.

In this specification, terms such as “comprise” or “have” are intended to indicate the presence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, but are not intended to indicate the presence of one or more other features. It should be understood that this does not exclude in advance the possibility of the existence or addition of elements, numbers, steps, operations, components, parts, or combinations thereof.

A controller is a communication device that communicates with other controllers or sensors to control the function it is in charge of, a memory that stores the operating system, logic commands, input/output information, etc., and a device that performs the judgments, calculations, and decisions necessary to control the function it is in charge of. It may include more than one processor.

In embodiments of the present invention, when classifying a seated person using image information, the seated person is detected based on the relative area of the seat according to the seat position and the seated person's area on the image information to increase detection accuracy and to classify the seated person's type in detail. We propose a classification method and device for vehicle occupants that allows for proper classification.

Before explaining the method for classifying vehicle occupants according to the present invention, a vehicle occupant classification device according to an embodiment of the present invention will first be described with reference to FIG. 1 .

Figure 1 is a diagram showing the configuration of a vehicle seat classification device according to an embodiment of the present invention.

Referring to FIG. 1, the vehicle occupant classification device 100 according to the embodiment may include a seat 110, a position sensor 120, a weight sensor 130, a camera 140, and a controller 150. . Figure 1 mainly shows components related to embodiments of the present invention, and it is obvious to those skilled in the art that more or fewer components may be included in the actual implementation of the classification device 100. Each component is described below.

First, the vehicle occupant classification device 100 according to the embodiment includes at least one seat 110, and the seats 110 may be divided into a driver's seat 111, a passenger seat 112, etc. depending on the placement location. Meanwhile, in the present invention, the seat position of the seat 110 can be adjusted. The seat position may include at least one of a position and a seatback angle along a direction parallel to the driving direction of the seat 110 disposed on the driver's seat 111, the passenger seat 112, etc. Here, the position may refer to the state of the sheet 110, including a position in which the sheet 110 moves in the forward and backward direction through a seat track, etc., while maintaining within a certain range from the placement location. Additionally, the seatback angle may mean the angle of the seatback with respect to the seating surface or the ground depending on the recline or relaxation mode of the seat 110.

Meanwhile, the vehicle occupant classification device 100 according to an embodiment of the present invention may further include a position sensor 120 that is connected to the seat and detects the seat position. For example, the position sensor 120 may include a seat track position sensor that detects the front-to-back position of the seat 110 on the seat track. Through this position sensor 120, the position of the seat 110 can be detected more accurately, and this can be used to classify the seated person. In addition, it is possible to continuously monitor whether the position of the seat 110 changes through the position sensor 120 and update the classification of the seated person based on the changes.

In addition, the vehicle seat classification device 100 according to an embodiment of the present invention is connected to the seat 110, detects the load applied to the seat 110, and includes a weight sensor 130 that transmits weight information of the seated person. ) may further be included. This weight sensor 130 may be implemented through an Occupant Classification System (OCS) or an Occupant Detection System (ODS). When the weight sensor 130 is provided in the seated person classification device 100, the seated person can be classified by additionally utilizing weight information in addition to image information, thereby improving classification accuracy.

Meanwhile, the camera 140 performs the function of photographing the interior of the vehicle and transmitting the captured image. An image captured by the camera 140 may include at least one seat 110 or a person seated on the seat 110 together with the seat 110 . Additionally, it is appropriate for the camera 140 to be installed at the front of the vehicle and facing rearward so that the front of the seat 110 and the seated person can be clearly displayed. For example, the camera 140 may be implemented as an in-cabin camera. In addition, the in-cabin camera is mounted inside the vehicle and can check the status of the passengers in real time, so the occupant classification device 100 according to an embodiment of the present invention continuously monitors the boarding and exiting and changes of the occupants using real-time images. Monitoring becomes possible.

The image captured by the camera 140 may include the seat 110 and the seated person, and the seat 110 may be moved forward or backward or reclined while the camera 140 is fixed. When the position changes, the area occupied by the sheet 110 in the image changes. Even if the position of the seat 110 is changed, it does not affect the area of the seat 110 itself, but when the seat 110 is located in the front, it is located closer to the camera 140 compared to when the seat 110 is located in the rear, so it is not visible in the image. It occupies a large area. Additionally, when the recline angle of the seatback is small, a larger area is projected and occupies a larger area in the image compared to when the recline angle is large. Therefore, when determining the area value of the sheet 110 on the image by considering the position of the sheet 110, more accurate results can be derived.

Meanwhile, the controller 150 may obtain the position of the seat 110 from the position sensor 120, etc., and obtain an image of the inside of the vehicle from the camera 140.

The controller 150 can determine the area value of the seat 110 based on the parameters associated with the seat 110 position, and extracts the parameter corresponding to the current position of the seat 110 through the acquired position of the seat 110. And based on this, the area value of the sheet 110 can be determined. Here, the area value of the sheet 110 can be viewed as meaning the area of the area occupied by the sheet 110 on the image, not the actual area of the sheet 110.

Additionally, the controller 150 may detect a seated person from the acquired image of the inside of the vehicle and determine an area value corresponding to the detected seated person.

When the area value of the seat 110 and the area value of the seated person on the seat 110 are determined, the controller 150 can determine the area ratio through these, and classify the class, that is, the type, of the seated person based on the area ratio. can do. Here, the area ratio can be judged as the area value of the seated person/the area value of the seat 110. According to this, the larger the area ratio, the more the person seated is a relatively large person, and the smaller the area ratio, the smaller the person is.

The position of the seat 110 affects not only the area value of the seat 110, but also the distance and angle between the seated person on the seat 110 and the camera 140, so that the image for the same seated person depends on the seat 110 position. The seating area value also changes. Therefore, when only the area of the seated person is considered, the area value of the seated person in the image varies depending on the position of the seat 110, making it impossible to classify the seated person using a certain standard. On the other hand, when determining the class of a seated person through the above area ratio, the classification standard can be kept constant and the precision of classifying the seated person can be improved.

Specific details regarding the determination of the area value of the seat 110 and the area of the seated person and the determination of the seated person's class based on the area ratio will be described in detail below with reference to FIGS. 2 to 4.

Figure 2 is a diagram representing an image of the interior of a vehicle including a seat according to an embodiment of the present invention.

Referring to Figure 2, an image including the driver's seat 111 and the passenger seat 112 appears. The actual seat 110 sizes of the driver's seat 111 and the passenger seat 112 are the same, but it can be seen that the driver's seat 111 occupies a larger area in the image than the passenger seat 112. This is because the position of the driver's seat 111 is located forward compared to the passenger seat 112 and is relatively close to the camera 140. Considering this, the controller 150 obtains the position of the seat 110 and determines the area value of the seat 110 based on the parameters associated therewith.

Here, the parameter determines the area value of the seat 110 according to the position of the seat 110. The closer the seat 110 is located to the front, the larger the area value of the seat 110 is, and the closer the seat 110 is located to the rear, the larger the seat 110 area value is. (110) It is appropriate that the area value is set to be small. In this case, the parameters may be set to correspond to at least the frontmost position and the rearmost position of the seat 110 so that the controller 150 can determine the area value according to the front and rear positions of the seat 110.

Additionally, the position of the seat 110 may include not only the front and rear positions of the seat 110 but also the seatback recline angle. In this case, the controller 150 determines the area value of the seat 110 by additionally considering the recline angle along with the front and rear positions of the seat 110, thereby determining a value closer to the area according to the area occupied by the seat 110 in the actual image. can be obtained.

FIG. 2 illustrates the area value of the seat 110 according to the position of the seat 110 in the present invention, and the determination of the area value of the seated person will be described below with reference to FIG. 3 .

Figure 3 is a diagram showing how to determine the area value of a seated person in classifying a seated person according to an embodiment of the present invention.

Referring to FIG. 3, an image of the interior of the vehicle including the seat 110 and the occupants can be seen. The controller 150 detects the seated person from the acquired image and determines the seated area value corresponding to the detected seated person.

Here, the detection of the seated person is to recognize the seated person appearing on the image, and can be accomplished using an artificial intelligence learning model. Artificial intelligence learning models may specifically include deep learning. In the case of deep learning, the seated person can be set as ROI (Region Of Interest), and the seated person can be detected through ROI search.

To explore ROI, first, labels containing learning data images, class information, and location information are prepared. In an embodiment of the present invention, an image of the seated person, the class of the seated person in the image (adult, child, etc.), and the position of the seated person on the image may be prepared. In the deep learning process, the learning model learns the characteristics of the seated person through this information, and repeated learning may be involved to improve accuracy. After learning is performed through deep learning, the learning model receives the image shown in Figure 3 and searches for an area that satisfies the characteristics of the seated person learned from the image. If an area that satisfies the characteristics of the seated person is found, the learning model can detect the area as the seated person.

If a seated person is not detected, the controller 150 may determine that there is no seated person in the seat 110 and derive a class such as 'unoccupied state' or 'Empty'. In particular, when a plurality of seats 110 are included in the acquired image and the occupants of some of the seats 110 are not detected, which seat 110 is the occupant present and which seat 110 is the occupant? You can check whether it does not exist.

Meanwhile, the seat area value can be determined by seat segmentation. To this end, the controller 150 may obtain the seated person's segment again from the acquired image and determine the seated person's area value based on this. Here, segmentation means separating the objects that appear in the image by type, and is separating areas according to the meaning of each part at the pixel level of the image. In an embodiment of the present invention, it means separating the seated area from the area other than the seated area, and segmentation of the seated person can be understood as separating the area corresponding to the seated person on the image. In determining the area value of the seated person, by using the seated person segmentation, the area occupied by the seated person on the image can be identified more precisely, thereby improving the accuracy of determining the area value of the seated person on the image. This seated segmentation may be performed by an artificial intelligence learning model, where the artificial intelligence learning model is an artificial intelligence algorithm and may include at least one of machine learning, neural network, deep learning, classification algorithm, or a combination thereof.

Meanwhile, referring to FIG. 3, a plurality of sheets 110 may be included in the acquired image. In this case, the controller 150 determines the area ratio according to the area value of each seat 110 and the seated person on the seat 110 for each seat 110, and based on this, the class of the seated person can be determined for each seat. . For example, when the driver's seat 111 and the passenger seat 112 are included in the acquired image and the driver 113 and the passenger 114 are seated, the controller 150 determines the seat area value of the driver's seat 111, the passenger seat ( The seat area values of the driver 112) are determined, and the seat area values of the driver 113 and the passenger 114 are determined respectively. Thereafter, the area ratio for the driver's seat 111 is the seating area value of the driver 113/the seat area value of the driver's seat 111, and the area ratio for the passenger seat 112 is the seating area value of the passenger 114/passenger seat ( 112) Each is judged based on the sheet area value. Thereafter, the controller 150 determines the classes of the driver 113 and the passenger 114 based on each determined area ratio. Additionally, unlike what is shown in FIG. 3 , assuming that the passenger 114 is not in the vehicle, the process of determining the class of the driver 113 is performed in the same manner. In the case of the passenger seat 112 without a seated person, as the seated person is not detected, the controller 150 determines the seated class as a specific class such as 'unoccupied' or 'Empty', or the seated area value becomes 0 and the controller 150 (150) may determine the seated person's class to be a specific class such as 'unoccupied state' or 'Empty'.

Figures 2 and 3 illustrate the process of deriving the area values of the seat 110 and the seated person in classifying the seated person according to an embodiment of the present invention, and the class of the seated person is determined through the derived area values. The process will be described below with reference to Figure 4.

Figure 4 is a diagram showing classification of seated person classes based on area ratio and threshold value in classifying seated persons according to an embodiment of the present invention.

Referring to FIG. 4, a graph is displayed where one axis represents time and one axis represents area ratio. Here, the area ratio refers to the ratio of the area value of the seat 110 and the area value of the seated person, as described above, and the threshold value can be understood as a standard for dividing the section of the area ratio. The threshold value is a preset value and may include one or more values, and the class of the seated person may be determined according to each section formed around at least one threshold value. For example, in the case where the threshold includes the first to fourth thresholds as shown in FIG. 4, five sections are formed on the graph by each threshold. Through this, the controller 150 sets the class of the seated person to the first class when the area ratio is less than the first threshold, and sets the class of the seated person to the second class when the area ratio is more than the first threshold or less than the second threshold. If the area ratio is above the 2nd threshold but below the 3rd threshold, the class of the seated person is set to the 3rd class, and if the area ratio is above the 3rd threshold but below the 4th threshold, the class of the seated person is set to the 4th class, and the area ratio is set to the 4th threshold. In the above case, the class of the seated person can be judged as the 5th class. Here, assuming that the area ratio is the seat area value/seat 110 area value, it can be understood that the larger the total area ratio, the larger the relative size of the seated person, so the seated person with the first class is the relatively smallest person or non-mounted person. It means status, and the fifth class can be classified as meaning the relatively largest person. For example, seated persons may be classified as shown in Table 1 below.

section class seater Area ratio < first threshold 1st class Large First threshold value ≤ area ratio < second threshold value second class Medium Second threshold ≤ area ratio < third threshold third class Small Third threshold ≤ area ratio < fourth threshold 4th class Child Fourth threshold ≤ area ratio 5th class Empty

Hereinafter, the order in which the process of classifying a seated person according to the embodiments of the present invention described so far is performed will be described with reference to FIG. 5.

Figure 5 is a flowchart of a method for classifying seat occupants for a vehicle according to an embodiment of the present invention.

Referring to FIG. 5, first, the image inside the vehicle captured by the camera 140 is input to the controller 150 (S401). Additionally, the controller 150 obtains the seat 110 position including the front-to-back position of the seat 110, the recline angle of the seatback, etc. from the seat track position sensor 130 (S402).

The controller 150 determines the area value of the seat 110 through the position of the seat 110 (S403). In this case, the area value of the seat 110 is determined based on parameters associated with the position of the seat 110, and the parameters may be preset to be at least equal to the frontmost and rearmost positions of the seat 110. Additionally, the controller 150 may determine the area value of the seat 110 by additionally considering the recline angle of the seatback. In this case, the parameter may be set to be related to the recline angle of the seatback.

Meanwhile, the controller 150 detects the seated person from the acquired image (S404). In this case, if the seated person is not detected (No in S404), the controller 150 may determine the class of the seated person to be a specific class, such as the first class (S405). When a seated person is detected (Yes in S404), the controller 150 determines an area ratio according to the area value of the seat 110 and the seated person's area value of the seat 110 (S406).

The controller 150 determines the class of the seated person based on the determined area ratio. In this case, the class of the seated person may be determined based on a comparison result between the determined area ratio and the threshold value. If the area ratio is greater than or equal to the first threshold and less than the second threshold (Yes in S407), the controller 150 sets the class of the seated person to the second class (S408). If the area ratio is greater than or equal to the second threshold but less than the third threshold ( Yes in S409) If the class of the seated person is set to the third class (S410), and the area ratio is greater than or equal to the third threshold but less than the fourth threshold (Yes in S411), the class of the seated person is set to the fourth class (S412), and the area ratio is set to the fourth class (S412). 4 If it is more than the threshold (Yes in S413), the class of the seated person can be determined as the 5th class (S412), and if the area ratio is less than the first threshold (No in S413), the class of the seated person can be determined as the 1st class (S4150). It can be judged as follows. The process of determining the class of the seated person based on the result of comparing the area ratio and the threshold value may be performed in a series like this, but may also be performed individually depending on whether each condition is satisfied.

The method for classifying seated occupants in a vehicle according to at least one embodiment of the present invention configured as described above can more accurately detect and classify seated occupants in detail by utilizing image information.

By improving the accuracy of seated occupant detection, problems such as malfunctioning of functions installed in the vehicle or failure to perform as intended due to incorrect detection can be alleviated.

Furthermore, through further refinement of the classification of occupants, each function mounted on the vehicle can be applied to more specifically reflect the characteristics of each occupant, thereby improving performance.

Although the present invention has been shown and described in relation to specific embodiments of the present invention as described above, it is understood that the present invention can be variously improved and changed without departing from the technical spirit of the present invention provided by the following claims. This will be self-evident to those skilled in the art.

110: sheet
140: camera
150: controller

Claims (15)

Obtaining images of the interior of the vehicle;
Obtaining a seating position of a seat;
determining a seat area value based on parameters associated with the seat position;
Detecting a seated person from the acquired image and determining a seated area value corresponding to the detected seated person; and
A method of classifying a seated person for a vehicle comprising: determining a class of the seated person based on an area ratio between the seat area value and the seated person area value of the corresponding seat. In claim 1,
The image above is,
A method of classifying seat occupants for a vehicle, characterized in that the image is captured by an in-cabin camera provided at the front of the vehicle interior, facing rearward. In claim 1,
The seat position is,
Including the front-to-back position of the sheet,
The step of determining the sheet area value is,
A method for classifying seat occupants in a vehicle, comprising: determining the seat area value based on parameters preset to correspond to at least the frontmost and rearmost positions of the seat. In claim 3,
The step of acquiring the seat position is,
Obtaining a seat position according to the front-to-back position of the seat from a seat track position sensor connected to the seat. In claim 3,
The seat position is,
Including a seatback recline angle of the seat,
The step of determining the sheet area value is,
A method for classifying seat occupants in a vehicle, comprising: determining the seat area value by additionally considering the recline angle. In claim 1,
If the seated person cannot be detected from the acquired image, determining the class as a specific class. In claim 1,
The step of determining the seat area value is,
Obtaining seated segmentation from the acquired image; and
Determining the seat area value based on the obtained seat segmentation. In claim 7,
The step of obtaining the seated division is,
Obtaining the seated person segment from the acquired image by inputting the acquired image into a predetermined artificial intelligence learning model. A method for classifying a seated person for a vehicle, comprising: In claim 8,
The artificial intelligence learning model is,
An artificial intelligence algorithm, comprising at least one of machine learning, neural network, deep learning, classification algorithm, or a combination thereof. In claim 1,
The step of determining the class of the seated person is,
Comparing the area ratio with a preset threshold and determining the class of the seated person based on the comparison result. In claim 10,
The step of determining the class of the seated person is,
A method for classifying a seated person in a vehicle, comprising: determining the class of the seated person through a plurality of class rules defined as a range between two adjacent thresholds among a plurality of preset thresholds. In claim 1,
Further comprising: acquiring weight information of the seated person,
The step of determining the class of the seated person is,
A method of classifying a seated person for a vehicle, comprising: determining the class of the seated person by further considering the weight information. In claim 12,
The step of obtaining the weight information of the seated person is,
A method for classifying a seated person in a vehicle, comprising: acquiring weight information of the seated person from a weight sensor connected to the seat. In claim 1,
The step of determining the class of the seated person is,
determining the area ratio for each of a plurality of sheets; and
A method for classifying a seated person in a vehicle, comprising: determining the class of the seated person for each of the plurality of seats based on the determined area ratio. at least one sheet;
Cameras that film the interior of a vehicle; and
Obtain the seat position of the target seat among the at least one seat, determine the seat area value based on the parameter associated with the seat position, detect the seated person from the image inside the vehicle obtained through the camera, and detect the seated seat. A controller for determining the seat area value corresponding to the seat and determining the class of the seated person based on the seat area value and the area ratio according to the seat area value of the target seat: A seat for a vehicle including: Sorting device.

Images