Automobile seat adjustment method and device

Abstract

This invention provides a method and apparatus for adjusting a car seat. The method includes: acquiring a facial image of the driver when the driver sits in the car seat; determining the driver's three-dimensional facial dimensions, age, and gender based on the facial image; predicting the driver's body shape based on the three-dimensional facial dimensions, age, and gender; determining seat position parameters based on the driver's body shape; and adjusting the position of the car seat based on the seat position parameters. The method and apparatus provided by this invention can automatically adjust the car seat to a position suitable for the driver's body shape, eliminating the need for manual adjustment by the driver, reducing pre-driving preparation time, and providing a better user experience.

Description

Technical Field

[0001] This invention relates to the field of automotive electronics technology, and in particular to a method and apparatus for adjusting a car seat. Background Technology

[0002] To ensure drivers have adequate visibility and operating space, the car seat needs to be adjusted to a suitable position to accommodate their body type before driving. Currently, most car seat adjustment systems are manual, requiring the driver to adjust the seat manually. However, manual seat adjustment is inconvenient and provides a poor user experience. Summary of the Invention

[0003] In view of this, the present invention provides a method and apparatus for adjusting a car seat, to solve the problems of poor convenience and unsatisfactory user experience in existing car seat adjustment solutions. The technical solution is as follows:

[0004] A method for adjusting a car seat, comprising:

[0005] When the driver sits in the car seat, the driver's facial image is captured;

[0006] Based on the facial image, determine the driver's three-dimensional facial dimensions, age, and gender;

[0007] From the human body proportion template library, obtain a human body proportion template that matches the driver's age and gender as the target human body proportion template, wherein the target human body proportion template contains standard three-dimensional face size information and standard body size information.

[0008] Based on the three-dimensional face size information and the target human body proportion template, the driver's body size information is predicted as the driver's body shape information;

[0009] Based on the body shape information, determine the seat position parameters;

[0010] The position of the car seat is adjusted based on the seat position parameters.

[0011] Optionally, based on the facial image, determining the driver's three-dimensional facial size information includes:

[0012] Facial feature detection is performed on the face image to obtain the position information of facial feature points in the pixel coordinate system;

[0013] Based on the position information of the facial feature points in the pixel coordinate system and the predetermined transformation relationship between the world coordinate system and the pixel coordinate system, the position information of the facial feature points in the world coordinate system is determined.

[0014] Based on the position information of the facial feature points in the world coordinate system, the three-dimensional facial size information of the driver is determined.

[0015] Optionally, the car seat adjustment method further includes:

[0016] Based on the body shape information and the seat position parameters, determine the rearview mirror angle parameters;

[0017] Adjust the angle of the car's rearview mirrors based on the aforementioned rearview mirror angle parameters.

[0018] Optionally, before determining the driver's three-dimensional facial dimensions, age, and gender based on the facial image, the car seat adjustment method further includes:

[0019] Based on the facial image, the driver's identity information is obtained as the target identity information;

[0020] The system queries whether the target identity information exists in the driver information database, wherein the driver information database includes several identity information and seat position parameters and rearview mirror angle parameters corresponding to the several identity information respectively;

[0021] If the target identity information exists in the driver information database, the position of the car seat is adjusted based on the seat position parameters corresponding to the target identity information in the driver information database, and the angle of the car rearview mirror is adjusted based on the rearview mirror angle parameters corresponding to the target identity information in the driver information database.

[0022] If the target identity information is not found in the driver information database, the process of determining the driver's three-dimensional facial dimensions, age, and gender based on the facial image is performed.

[0023] Optionally, the car seat adjustment method further includes:

[0024] If the target identity information is not found in the driver information database, the target identity information, along with the determined seat position parameters and the determined rearview mirror angle parameters, are stored in the driver information database.

[0025] Optionally, determining the transformation relationship between the world coordinate system and the pixel coordinate system includes:

[0026] Acquire a chessboard image sequence, wherein the chessboard image sequence consists of several chessboard images corresponding to different seat positions, and the chessboard images are images acquired by the image acquisition device from a chessboard calibration plate placed horizontally on the horizontal surface of the car seat when the car seat is in the corresponding seat position.

[0027] Based on the chessboard image sequence, the transformation relationship between the world coordinate system and the pixel coordinate system is determined.

[0028] Optionally, determining the transformation relationship between the world coordinate system and the pixel coordinate system based on the checkerboard image sequence includes:

[0029] Corner point recognition is performed on each chessboard image in the chessboard image sequence to obtain the corner point recognition results corresponding to each chessboard image in the chessboard image sequence. The corner point recognition results include the positions of several corner points in the corresponding chessboard image in the pixel coordinate system.

[0030] Based on the positions of several corner points on the chessboard calibration board in the world coordinate system, and the corner point recognition results corresponding to each chessboard image in the chessboard image sequence, the transformation relationship between the world coordinate system and the pixel coordinate system is determined.

[0031] Optionally, determining the transformation relationship between the world coordinate system and the pixel coordinate system based on the positions of several corner points on the checkerboard calibration board in the world coordinate system and the corner point recognition results corresponding to each checkerboard image in the checkerboard image sequence includes:

[0032] Based on the positions of several corner points on the chessboard calibration board in the world coordinate system, and the corner point recognition results corresponding to the two chessboard images in the chessboard image sequence, the initial intrinsic and extrinsic parameters of the image acquisition device are determined, wherein the extrinsic parameter is the relative positional relationship between the image acquisition device and the car seat.

[0033] Construct an objective function, wherein the objective function represents the sum of the absolute values ​​of the position differences corresponding to each corner point in other chessboard images in the chessboard image sequence, wherein the position difference is the difference between the actual position of the corresponding corner point in the world coordinate system and the predicted position in the world coordinate system, and the predicted position of a corner point in the world coordinate system is based on the position of the corner point in the pixel coordinate system, as well as the current intrinsic parameters, extrinsic parameters and distortion coefficients of the image acquisition device.

[0034] By minimizing the objective function, the intrinsic parameters, extrinsic parameters, and distortion coefficients of the image acquisition device are updated until the value of the objective function is less than a preset error threshold.

[0035] Based on the final intrinsic parameters, extrinsic parameters, and distortion coefficients, the transformation relationship between the world coordinate system and the pixel coordinate system is determined.

[0036] A car seat adjustment device includes: an image acquisition module, a driver information determination module, a human body proportion template determination module, a body shape information prediction module, a seat position parameter determination module, and a seat position adjustment module;

[0037] The image acquisition module is used to acquire the driver's facial image when the driver sits in the car seat;

[0038] The driver information determination module is used to determine the driver's three-dimensional facial size information, as well as the driver's age and gender, based on the facial image.

[0039] The human body proportion template determination module is used to obtain a human body proportion template that matches the driver's age and gender from the human body proportion template library as a target human body proportion template, wherein the target human body proportion template contains standard three-dimensional face size information and standard body size information.

[0040] The body shape information prediction module is used to predict the driver's body size information based on the three-dimensional face size information and the target human body proportion template, as the driver's body shape information;

[0041] The seat position parameter determination module is used to determine the seat position parameters based on the body shape information;

[0042] The seat position adjustment module is used to adjust the position of the car seat based on the seat position parameters.

[0043] Optionally, the car seat adjustment device provided by the present invention further includes: a rearview mirror angle parameter determination module and a rearview mirror angle adjustment module;

[0044] The rearview mirror angle parameter determination module is used to determine the rearview mirror angle parameters based on the body shape information and the seat position parameters;

[0045] The rearview mirror angle adjustment module is used to adjust the angle of the car rearview mirror based on the rearview mirror angle parameters.

[0046] The car seat adjustment method and apparatus provided by the present invention first acquires the driver's facial image when the driver sits in the car seat, then determines the driver's three-dimensional facial size information, age, and gender based on the driver's facial image, then obtains a human proportion template matching the driver's age and gender from a human proportion template library as a target human proportion template, then predicts the driver's body shape information based on the driver's three-dimensional facial size information and the target human proportion template, and finally determines the seat position parameters based on the driver's body shape information, and adjusts the position of the car seat based on the determined seat position parameters. The car seat adjustment method and device provided by this invention can determine the driver's three-dimensional facial size information, age, and gender based on the driver's facial image. It can also obtain a human proportion template matching the driver's age and gender from a human proportion template library. Furthermore, it can predict the driver's body shape based on the driver's three-dimensional facial size information and the obtained human proportion template, thereby adjusting the car seat to a position suitable for the driver's body shape. As can be seen, the car seat adjustment method and device provided by this invention can automatically adjust the car seat to a position suitable for the driver's body shape without requiring manual adjustment by the driver, reducing the driver's preparation time before driving and providing a better user experience. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0048] Figure 1 This is a flowchart illustrating a method for adjusting a car seat according to an embodiment of the present invention.

[0049] Figure 2 A flowchart illustrating another method for adjusting a car seat provided in an embodiment of the present invention;

[0050] Figure 3 A schematic flowchart illustrating another method for adjusting a car seat provided in an embodiment of the present invention;

[0051] Figure 4 A schematic diagram illustrating the process of determining the transformation relationship between the world coordinate system and the pixel coordinate system based on a checkerboard image sequence, as provided in an embodiment of the present invention.

[0052] Figure 5 This is a schematic diagram of the coordinate system involved in an embodiment of the present invention;

[0053] Figure 6This is a schematic diagram of the structure of the car seat adjustment device provided in an embodiment of the present invention;

[0054] Figure 7 This is a schematic diagram of the structure of an automotive seat adjustment device provided in an embodiment of the present invention. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] Current car seat adjustment methods are mostly manual. Given the poor convenience and unpleasant user experience of manual adjustment, the inventors of this invention sought to propose an automatic car seat adjustment method. Through continuous research, a superior car seat adjustment method was ultimately proposed. This method automatically adjusts the car seat to a position suitable for the driver's body shape, perfectly overcoming the shortcomings of existing car seat adjustment methods. The following embodiments will further illustrate the car seat adjustment method provided by this invention.

[0057] First Embodiment

[0058] Please see Figure 1 The diagram illustrates a flowchart of a car seat adjustment method according to an embodiment of the present invention, which may include:

[0059] Step S101: When the driver sits in the car seat, acquire the driver's facial image.

[0060] When the driver opens the car door and sits in the car seat, the driver's facial image can be obtained based on the image acquisition device.

[0061] Optionally, when the driver opens the car door and sits in the car seat, the Driver Monitoring System (DMS) can be activated to obtain the driver's facial image based on the camera of the Driver Monitoring System.

[0062] Step S102: Based on the driver's facial image, determine the driver's three-dimensional facial size information, as well as the driver's age and gender.

[0063] Optionally, the process of determining the driver's three-dimensional facial dimensions based on the driver's facial image may include:

[0064] Step S1021: Perform facial feature detection on the driver's face image to obtain the position of the facial feature points in the pixel coordinate system.

[0065] Facial feature points can include some or all of the following feature points: the upper and lower edges of the face contour (upper and lower edge points), the left and right edges of the face contour (left and right edge points), the center point of the left eye, the center point of the right eye, etc. The position of the facial feature points in the pixel coordinate system is the two-dimensional pixel position of the facial feature points.

[0066] Step S1022: Based on the position of the facial feature points in the pixel coordinate system and the pre-determined transformation relationship between the world coordinate system and the pixel coordinate system, determine the position of the facial feature points in the world coordinate system.

[0067] The process of determining the transformation relationship between the world coordinate system and the pixel coordinate system will be described in subsequent embodiments.

[0068] Step S1023: Determine the driver's three-dimensional face size information based on the position of the facial feature points in the world coordinate system.

[0069] Optionally, the driver's three-dimensional facial size information may include, but is not limited to, one or more of the following: face length, face width, and eye height in sitting position.

[0070] The face length can be determined by the positions of the upper edge point of the face contour (P1) and the lower edge point (P2) in the world coordinate system. Specifically, the distance between P1 and P2 is the face length. The face width can be determined by the positions of the left edge point of the face contour (P3) and the right edge point (P4) in the world coordinate system. Specifically, the distance between P3 and P4 is the face width. The seated eye height can be determined by the position of the left eye's centroid (P5) or the right eye's centroid (P6) in the world coordinate system. Specifically, the distance from P5 or P6 to the current seat bottom surface is the seated eye height.

[0071] When determining a driver's age and gender based on their facial image, existing methods for predicting age and gender based on facial images can be used to process the driver's facial image to obtain the driver's age and gender. For example, the driver's facial image can be input into a pre-trained facial attribute prediction model to obtain the driver's facial attributes output by the facial attribute prediction model, where the facial attributes include at least age and gender.

[0072] Step S103: Obtain a human proportion template that matches the driver's age and gender from the human proportion template library, and use it as the target human proportion template.

[0073] The human body proportion template library contains human body proportion templates corresponding to various combinations of age and gender, such as human body proportion templates for 26-year-old men, 26-year-old women, 35-year-old men, and 35-year-old women, etc.

[0074] Each human body proportion template in the human body proportion template library contains standard 3D face size information and standard body size information. For example, the standard 3D face size information includes standard face length, standard face width, standard seated eye height, etc., and the standard body size information includes standard height, standard lower leg length, standard thigh length, standard torso length, standard upper arm length, standard forearm length, standard arch length, etc.

[0075] Step S104: Based on the driver's three-dimensional face size information and the target human body proportion template, determine the driver's body size information as the driver's body shape information.

[0076] Specifically, the size ratio r can first be determined based on the driver's three-dimensional face size information and the standard three-dimensional face size information in the target human body proportion template. Then, the driver's body size information can be determined based on the size ratio r and the standard body size information in the target human body proportion template.

[0077] If the standard 3D face size information in the target human body proportion template includes standard face length L0, standard face width W0, and standard seated eye height E0, and the driver's 3D face size information includes driver's face length L1, face width W1, and seated eye height E1, then the driver's body size information can be determined using any of the following implementation methods:

[0078] First implementation method:

[0079] r equals E1 / E0, or L1 / L0, or W1 / W0, and the driver's body size information equals standard body size information * r.

[0080] The second implementation method:

[0081] r equals (E1 / E0 + L1 / L0) / 2, or equals (E1 / E0 + W1 / W0) / 2, or equals (L1 / L0 + ...

[0082] W1 / W0) / 2, the driver's body size information is equal to the standard body size information *r.

[0083] The third implementation method:

[0084] r equals (E1 / E0+L1 / L0+W1 / W0) / 3, and the driver's body size information equals the standard body size information * r.

[0085] It should be noted that the driver's body dimensions are equal to the standard body dimensions in the target human body proportion template *r. That is, the driver's height is equal to the standard height in the target human body proportion template *r, the driver's lower leg length is equal to the standard lower leg length in the target human body proportion template *r, the driver's thigh length is equal to the standard thigh length in the target human body proportion template *r, the driver's torso length is equal to the standard torso length in the target human body proportion template *r, the driver's upper arm length is equal to the standard upper arm length in the target human body proportion template *r, the driver's forearm length is equal to the standard forearm length in the target human body proportion template *r, and the driver's arch length is equal to the standard arch length in the target human body proportion template *r.

[0086] Step S105: Determine the seat position parameters based on the driver's body shape information.

[0087] After obtaining the driver's body shape information, the seat position parameters that match the driver's body shape information can be determined, such as seat height, horizontal position, and backrest angle.

[0088] Step S106: Adjust the position of the car seat based on the seat position parameters.

[0089] Since the determined seat position is a seat position parameter that matches the driver's body shape information, the car seat can be adjusted to a position that adapts to the driver's body shape based on the determined seat position parameter, that is, to make the driver's hip joint, knee joint, elbow joint and ankle joint in a comfortable angle position.

[0090] The car seat adjustment method provided in this invention first acquires the driver's facial image when the driver sits in the car seat. Then, based on the driver's facial image, it determines the driver's three-dimensional facial dimensions, age, and gender. Next, it retrieves a human proportion template matching the driver's age and gender from a human proportion template library. Then, based on the driver's three-dimensional facial dimensions and the acquired human proportion template, it predicts the driver's body shape information. Finally, it determines the seat position parameters based on the driver's body shape information and adjusts the car seat position based on the determined seat position parameters. This car seat adjustment method can automatically adjust the car seat to a position suitable for the driver's body shape, eliminating the need for manual adjustment by the driver, reducing pre-driving preparation time, and providing a better user experience.

[0091] Furthermore, the car seat adjustment method provided in this embodiment of the invention can accurately obtain a human body proportion template based on the driver's age and gender, and then obtain high-precision body shape information based on the driver's three-dimensional facial size information and the obtained human body proportion template. This allows the car seat to be adjusted to a position that matches the driver's body height. Since the car seat adjustment device provided in this embodiment of the invention does not require the driver's information to be pre-entered, it can be well adapted to situations where drivers frequently change, such as car rental (frequent changes of the driver in the driver's seat) and taxi (frequent changes of the passenger driver in the passenger seat).

[0092] Second Embodiment

[0093] To avoid repeatedly determining seat position parameters for the same driver, and to enable the driver to quickly adjust the car seat to a suitable position when sitting down again, this embodiment provides another car seat adjustment method. Please refer to [link to relevant documentation]. Figure 2 The diagram illustrates a flowchart of the car seat adjustment method provided in this embodiment, which may include:

[0094] Step S201: When the driver sits in the car seat, acquire the driver's facial image.

[0095] When the driver opens the car door and sits in the car seat, the driver's facial image can be acquired using an image acquisition device. Optionally, when the driver opens the car door and sits in the car seat, the driver monitoring system can be activated to acquire the driver's facial image using the system's camera.

[0096] Step S202: Based on the driver's facial image, obtain the driver's identity information as the target identity information.

[0097] Existing facial image-based identity recognition methods can be used to identify the driver's facial image to obtain the driver's identity information.

[0098] Step S203: Query whether the target identity information exists in the driver information database. If the target identity information does not exist in the driver information database, execute steps S204a to S209a. If the target identity information exists in the driver information database, execute step S204b.

[0099] The driver information database stores several identity information and corresponding seat position parameters for each identity information. The seat position parameters in the driver information database are seat position parameters that match the body shape information of the driver corresponding to the identity information.

[0100] Step S204a: Based on the driver's face image, determine the driver's three-dimensional face size information, as well as the driver's age and gender.

[0101] Step S205a: Obtain a human proportion template that matches the driver's age and gender from the human proportion template library, and use it as the target human proportion template.

[0102] Step S206a: Based on the driver's three-dimensional face size information and the target human body proportion template, determine the driver's body size information as the driver's body shape information.

[0103] The specific implementation process and related explanations of steps S204a to S206a can be found in the specific implementation process and related explanations of steps S102 to S104 in the first embodiment, which will not be repeated here.

[0104] Step S207a: Determine seat position parameters based on the driver's body shape information.

[0105] If the target identity information is not found in the driver information database, the seat position parameters are determined using the steps S204a to S207a described above.

[0106] Step S208a: Adjust the position of the car seat based on the determined seat position parameters.

[0107] Step S209a: Store the target identity information and the determined seat position parameters in the driver information database.

[0108] The target's identity information and the determined seat position parameters are stored in the driver information database, so that when the driver sits in the car seat again, the car seat can be quickly adjusted to a position that suits the driver's body shape based on the seat position parameters stored in the driver information database.

[0109] Step S204b: Adjust the position of the car seat based on the seat position parameters corresponding to the target identity information in the driver information database.

[0110] If the target's identity information exists in the driver information database, the car seat position is adjusted directly based on the seat position parameters corresponding to that information. Since there's no need to re-determine the seat position parameters, the car seat can be quickly adjusted to a position suitable for the driver's body type based on these parameters.

[0111] The car seat adjustment method provided in this invention can automatically adjust the car seat to a position suitable for the driver's body shape, eliminating the need for manual adjustment by the driver, reducing pre-driving preparation time, and providing a better user experience. Furthermore, when the current driver's identity information exists in the driver information database, the method can quickly adjust the car seat to a position suitable for the driver's body shape without needing to re-determine the seat position parameters. When the current driver's identity information is not in the database, a human body proportion template can be accurately obtained based on the driver's age and gender. Then, based on the driver's three-dimensional facial dimensions and the obtained human body proportion template, high-precision body shape information can be obtained, thereby adjusting the car seat to a position highly compatible with the driver's body shape.

[0112] Third Embodiment

[0113] Please see Figure 3 The diagram illustrates a flowchart of another car seat adjustment method provided by an embodiment of the present invention, which may include:

[0114] Step S301: When the driver sits in the car seat, acquire the driver's facial image.

[0115] When the driver opens the car door and sits in the car seat, the driver's facial image can be acquired using an image acquisition device. Optionally, when the driver opens the car door and sits in the car seat, the driver monitoring system can be activated to acquire the driver's facial image using the system's camera.

[0116] Step S302: Based on the driver's facial image, obtain the driver's identity information as the target identity information.

[0117] Existing facial image-based identity recognition methods can be used to identify the driver's facial image to obtain the driver's identity information.

[0118] Step S303: Query whether the target identity information exists in the driver information database. If the target identity information does not exist in the driver information database, execute step S304a and subsequent steps. If the target identity information exists in the driver information database, execute step S304b.

[0119] The driver information database stores several identity information and corresponding seat position parameters and rearview mirror angle parameters for each identity information. The seat position parameters in the driver information database are seat position parameters that match the body type information of the driver with the corresponding identity information, and the rearview mirror angle parameters in the driver information database are rearview mirror angle parameters suitable for the driver with the corresponding identity information.

[0120] Step S304a: Based on the driver's face image, determine the driver's three-dimensional face size information, as well as the driver's age and gender.

[0121] Step S305a: Obtain a human proportion template that matches the driver's age and gender from the human proportion template library, and use it as the target human proportion template.

[0122] Step S306a: Based on the driver's three-dimensional face size information and the target human body proportion template, determine the driver's body size information as the driver's body shape information.

[0123] The specific implementation process and related explanations of steps S304a to S306a can be found in the specific implementation process and related explanations of steps S102 to S104 in the first embodiment, which will not be repeated here.

[0124] Step S307a: Determine the seat position parameters based on the driver's body shape information, and determine the rearview mirror angle parameters based on the driver's body shape information and seat position parameters.

[0125] If the target identity information is not found in the driver information database, the seat position parameters and rearview mirror angle parameters are determined using the steps S304a to S307a described above.

[0126] Step S308a: Based on the determined seat position parameters, adjust the position of the car seat, and based on the determined rearview mirror angle parameters, adjust the angle of the car rearview mirror.

[0127] Step S309a: By interacting with the driver, determine whether to fine-tune the position of the car seat and / or the angle of the car rearview mirror. If not, proceed to step S310a-a; if yes, proceed to step S310a-b and subsequent steps.

[0128] The driver can be asked in some way (such as by voice) whether he / she needs to make minor adjustments to the position of the car seat and / or the angle of the car rearview mirror. If no minor adjustments are needed, then step S310a-a is executed; otherwise, steps S310a-b and subsequent steps are executed.

[0129] Step S310a-a: Store the target identity information, the determined seat position parameters, and the determined rearview mirror angle parameters into the driver information database.

[0130] The target's identity information, the determined seat position parameters, and the determined rearview mirror angle parameters are stored in the driver information database. This allows the car seat to be quickly adjusted to a position suitable for the driver's body shape based on the seat position parameters corresponding to the target's identity information stored in the driver information database the next time the driver sits in the car. The car seat can also be quickly adjusted to a suitable angle for the driver based on the rearview mirror angle parameters corresponding to the target's identity information stored in the driver information database.

[0131] Steps S310a-b: Based on the driver's adjustment operation, fine-tune the position of the car seat and / or the angle of the car rearview mirror.

[0132] Steps S311a-b: Store the target identity information and the final seat position parameters and rearview mirror angle parameters in the driver information database.

[0133] Step S304b: Based on the seat position parameters and rearview mirror angle parameters corresponding to the target identity information in the driver information database, adjust the position of the car seat and the angle of the car rearview mirror.

[0134] If the seat position parameters and rearview mirror angle parameters corresponding to the target identity information exist in the driver information database, there is no need to re-determine the seat position parameters and rearview mirror angle parameters. Based on the seat position parameters and rearview mirror angle parameters corresponding to the target identity information in the driver information database, the car seat can be quickly adjusted to a position that adapts to the driver's body shape, and the car rearview mirror angle can be quickly adjusted to a suitable angle for the driver.

[0135] The car seat adjustment method provided in this invention can automatically adjust the car seat to a position that suits the driver's body shape and automatically adjust the angle of the rearview mirror to a suitable angle for the driver, without requiring manual adjustment by the driver, reducing the driver's preparation time before driving and providing a better user experience. Furthermore, the car seat adjustment method provided in this embodiment of the invention, when the current driver's identity information exists in the driver information database, does not require re-determining the seat position parameters and rearview mirror angle parameters. Based on the seat position parameters corresponding to the current driver's identity information in the driver information database, the car seat can be quickly adjusted to a position suitable for the current driver's body shape. Based on the rearview mirror angle parameters corresponding to the current driver's identity information in the driver information database, the angle of the car rearview mirror can be quickly adjusted to a suitable angle for the driver. After adjusting the position of the car seat and the angle of the rearview mirror, the car seat can be adjusted to a position more suitable for the current driver's body shape and the angle of the rearview mirror can be adjusted to a more suitable angle for the current driver through interaction with the current driver. When the current driver's identity information does not exist in the driver information database, an accurate human body proportion template can be obtained based on the driver's age and gender. Then, high-precision body shape information can be obtained based on the driver's three-dimensional face size information and the obtained human body proportion template. On this basis, seat position parameters that highly match the driver's body shape information can be determined, thereby adjusting the car seat to a position highly matching the driver's body shape.

[0136] Fourth embodiment

[0137] The first embodiment mentions that when determining the driver's three-dimensional face size information based on the driver's face image, after obtaining the position of the face feature points in the pixel coordinate system, it is necessary to "determine the position of the face feature points in the world coordinate system based on the position of the face feature points in the pixel coordinate system and the pre-determined transformation relationship between the world coordinate system and the pixel coordinate system". This embodiment introduces the process of determining the transformation relationship between the world coordinate system and the pixel coordinate system.

[0138] In one possible implementation, the process of determining the transformation relationship between the world coordinate system and the pixel coordinate system may include: acquiring a chessboard image sequence, and determining the transformation relationship between the world coordinate system and the pixel coordinate system based on the chessboard image sequence.

[0139] The checkerboard image sequence consists of checkerboard images corresponding to several seat positions. The checkerboard image sequence includes at least two checkerboard images. The checkerboard image corresponding to a seat position is an image acquired by the image acquisition device from a checkerboard calibration plate placed horizontally on the horizontal surface of the car seat when the car seat is in that position.

[0140] It should be noted that the process of acquiring a checkerboard image using an image acquisition device is as follows: The position of the car seat is designated as position P. a At that time, an image was acquired of the checkerboard calibration plate on the car seat, and then the position of the car seat was adjusted to position P. b For example, an image is acquired for the checkerboard calibration plate on the car seat, ..., in this way, a checkerboard image sequence can be obtained, which consists of checkerboard images corresponding to several seat positions.

[0141] Please see Figure 4 This illustrates a flowchart of determining the transformation relationship between the world coordinate system and the pixel coordinate system based on a checkerboard image sequence, which may include:

[0142] Step S401: Perform corner point recognition on each chessboard image contained in the chessboard image sequence to obtain the corner point recognition results corresponding to each chessboard image contained in the chessboard image sequence.

[0143] The corner point recognition results include the positions of several corner points in the corresponding chessboard image in the pixel coordinate system.

[0144] Step S402: Based on the positions of several corner points on the checkerboard calibration board in the world coordinate system, and the corner point recognition results corresponding to each checkerboard image in the checkerboard image sequence, determine the transformation relationship between the world coordinate system and the pixel coordinate system.

[0145] Specifically, the process of determining the transformation relationship between the world coordinate system and the pixel coordinate system based on the positions of several corner points on the checkerboard calibration board in the world coordinate system and the corner point recognition results corresponding to each checkerboard image in the checkerboard image sequence can include:

[0146] Step S4021: Based on the positions of several corner points on the chessboard calibration board in the world coordinate system, and the corner point recognition results corresponding to the two chessboard images in the chessboard image sequence, determine the initial intrinsic and extrinsic parameters of the image acquisition device.

[0147] Among them, the intrinsic parameters of the image acquisition device are the parameters of the image acquisition device itself, and the extrinsic parameters of the image acquisition device are the relative positional relationship between the image acquisition device and the car seat.

[0148] To determine the transformation relationship between the world coordinate system and the pixel coordinate system (i.e., the pixel coordinates of a point Q on the horizontal surface of the car seat in the pixel coordinate system: Qp = [uv]) T and coordinates Q in the world coordinate system w =[x w y w z w ] TThe mapping relationship requires the use of camera coordinate system and image coordinate system as intermediate quantities, and coordinate system transformation is performed in sequence.

[0149] Before introducing coordinate system transformations, we will first introduce the definitions of these coordinate systems, such as... Figure 5 As shown, the world coordinate system is the three-dimensional coordinate system where the checkerboard calibration plate is located. Its zero point passes through the upper left corner of the checkerboard. The X and Y axes are parallel to the edge of the checkerboard calibration plate, and the Z axis is perpendicular to the plane of the checkerboard calibration plate. The unit is mm. The camera coordinate system is a three-dimensional coordinate system whose origin passes through the optical center of the camera. Its X and Y axes are parallel to the two sides of the photosensitive chip, and the Z axis is the principal optical axis of the camera. The unit is mm. The image coordinate system is a two-dimensional coordinate system whose origin passes through the center of the photosensitive chip. The unit is mm. The pixel coordinate system is a two-dimensional coordinate system that reflects the pixel coordinate position of the image point within the acquired image. The unit is pixel.

[0150] Transforming from the world coordinate system to the camera coordinate system can be done through rotation and translation. Since point Q on the horizontal plane of the car seat is at X... w Y w On the plane, therefore z w =0. Let the rotation matrix be defined as R = [r1 r2 r3], and the translation vector as t. Then the transformation relationship between the world coordinate system and the camera coordinate system is as follows:

[0151]

[0152] The transformation from the camera coordinate system to the image coordinate system is accomplished by referencing the pinhole imaging model; that is, a point P in the camera coordinate system... c P c The corresponding image point P on the photosensitive chip s And light center O c These three points are in a straight line, and the optical center O is... c The distance to the image sensor plane is the equivalent focal length f of the lens. Ignoring camera distortion, the transformation relationship between the camera coordinate system and the image coordinate system is as follows:

[0153]

[0154] It should be noted that manufacturing and assembly errors in the image acquisition device's photosensitive chip and optical lens introduce nonlinear distortion into the image. To improve detection accuracy, a nonlinear distortion model needs to be established to compensate for image distortion. Let r... s 2 =(x s 2 +y s 2 Then the radial distortion δ r It can be expressed as follows:

[0155]

[0156] Tangential distortion can be expressed by the following formula:

[0157]

[0158] Let Q be the actual image coordinates of point Q. s *=[x s *y s *] T Its coordinates are related to the ideal image coordinates Q. s =[x s y s ] T The transformation relationship is as follows:

[0159]

[0160] It should be noted that r s 2 =(x s 2 +y s 2 The subscript s in equations (3) to (5) above represents the image coordinate system.

[0161] Let the coordinates of the origin of the image coordinate system in the pixel coordinate system be [u0 v0]. T If the width of a pixel in the photosensitive chip is dx and the pixel width is dy, then the transformation relationship between the image coordinate system and the pixel coordinate system is as follows:

[0162]

[0163] Without considering the nonlinear distortion introduced into the image by the image acquisition device, the transformation relationship between the world coordinate system and the pixel coordinate system is as follows:

[0164]

[0165] in, The intrinsic parameter matrix of the image acquisition device is usually denoted as f / dx. x f / dy is denoted as f y B = [r1r2t] is the extrinsic parameter matrix of the image acquisition device, which represents the relative positional relationship between the image acquisition device and the car seat, and H = AB is the homography matrix.

[0166] Expanding H in equation (7) yields:

[0167]

[0168] Expanding equation (8) yields:

[0169]

[0170] Substituting the third expression in equation (9) into the first and second expressions, we get:

[0171]

[0172] Based on equation (10), we can obtain:

[0173]

[0174] To calculate H, we need to obtain the positions (u) of the N corner points on the chessboard calibration board in the world coordinate system. 1 ,v 1 ) T ), ..., (u N ,v N ) T And, the positions of the N corner points in the pixel coordinate system (x 1 w ,y 1 w ) T ... (x N w ,y N w ) T The positions of N corner points in the world coordinate system are known, while their positions in the pixel coordinate system can be obtained by corner point recognition of the checkerboard image. The positions of the N corner points are then obtained by... 1 ,v 1 ) T ,(x 1 w ,y 1 w ) T ), ..., ((u N ,v N ) T ,(x N w ,y N w ) T After that, substituting the positions of the N corner points into (11) yields the following system of equations:

[0175]

[0176] Solving the above system of equations will yield the elements of the H matrix.

[0177] Write H as a column vector:

[0178] H=[h1h2h3]=M[r1r2t] (13)

[0179] but:

[0180]

[0181] Since the rotation vectors are mutually orthogonal and their magnitude is 1, we have:

[0182]

[0183]

[0184] Without considering the distortion of the image acquisition device, the intrinsic parameter matrix M has f x f y There are four parameters: u0, v0, and u0. Each chessboard image corresponding to H can provide two equations (15) and (16). Theoretically, the intrinsic parameter matrix M can be solved using two chessboard images. The solved M is used as the initial intrinsic parameter matrix of the image acquisition device. After obtaining the initial intrinsic parameter matrix M, the initial extrinsic parameter matrix can be obtained based on M and H.

[0185] Step S4022: Construct the objective function.

[0186] In this embodiment, the objective function represents the sum of the absolute values ​​of the position differences corresponding to each corner point in other chessboard images within the chessboard image sequence. The position difference is the difference between the actual position of the corresponding corner point in the world coordinate system and its predicted position in the world coordinate system. The predicted position of a corner point in the world coordinate system is based on its position in the pixel coordinate system, as well as the current intrinsic and extrinsic parameters and distortion coefficients of the image acquisition device. The objective function is as follows:

[0187]

[0188] Where M is the number of other chessboard images in the chessboard image sequence, N is the number of corner points in a single chessboard image, A is the intrinsic parameter matrix of the image acquisition device, k1, k2, p1, and p2 are the distortion coefficients of the image acquisition device, and B... i (i from 1 to M) is the extrinsic parameter matrix of the image acquisition device, B i The position of the image acquisition device and the seat is P. i The relative positional relationship of the car seats when (the seat position corresponding to the i-th checkerboard image).

[0189] Step S4023: By minimizing the objective function, update the intrinsic parameters, extrinsic parameters, and distortion coefficients of the image acquisition device until the value of the objective function is less than a preset error threshold.

[0190] The objective function is optimized to minimize it (for example, using the Levenberg-Marquardt method). This optimization process involves continuously updating the parameters based on the initial intrinsic, extrinsic, and distortion coefficients. The final intrinsic, extrinsic, and distortion coefficients represent the values ​​that optimize the objective function (i.e., values ​​less than a preset error threshold).

[0191] Step S4024: Based on the final intrinsic parameter matrix, extrinsic parameter matrix, and distortion coefficients, determine the transformation relationship between the world coordinate system and the pixel coordinate system.

[0192] The transformation relationship between the world coordinate system and the pixel coordinate system can be obtained through the above process.

[0193] Fifth Embodiment

[0194] This invention also provides a car seat adjustment device. The car seat adjustment device provided in this invention will be described below. The car seat adjustment device described below can be referred to in correspondence with the car seat adjustment method described above.

[0195] Please see Figure 6 The diagram shows a schematic of the structure of a car seat adjustment device provided in an embodiment of the present invention, which may include: an image acquisition module 601, a driver information determination module 602, a human body proportion template determination module 603, a body shape information prediction module 604, a seat position parameter determination module 605, and a first seat position adjustment module 606.

[0196] The image acquisition module 601 is used to acquire the driver's face image when the driver sits in the car seat.

[0197] The driver information determination module 602 is used to determine the driver's three-dimensional facial size information, as well as the driver's age and gender, based on the facial image.

[0198] The human body proportion template determination module 603 is used to obtain a human body proportion template that matches the driver's age and gender from the human body proportion template library, and use it as the target human body proportion template.

[0199] The target human body proportion template includes standard three-dimensional face size information and standard body size information.

[0200] The body shape information prediction module 604 is used to predict the driver's body size information based on the driver's three-dimensional face size information and the target human body proportion template, as the driver's body shape information.

[0201] The seat position parameter determination module 605 is used to determine the seat position parameters based on the driver's body shape information.

[0202] The first seat position adjustment module 606 is used to adjust the position of the car seat based on the seat position parameters.

[0203] Optionally, the driver information determination module 602 may include: a face feature detection submodule, a location determination submodule, and a three-dimensional face size information determination submodule.

[0204] The face feature detection submodule is used to perform face feature detection on the driver's face image to obtain the position of the face feature points in the pixel coordinate system.

[0205] The location determination submodule is used to determine the position of the facial feature points in the world coordinate system based on the position of the facial feature points in the pixel coordinate system and the pre-determined transformation relationship between the world coordinate system and the pixel coordinate system.

[0206] The three-dimensional face size information determination submodule is used to determine the driver's three-dimensional face size information based on the position of the face feature points in the world coordinate system.

[0207] Optionally, the car seat adjustment device provided in this embodiment of the invention may further include: a rearview mirror angle parameter determination module and a rearview mirror angle adjustment module.

[0208] The rearview mirror angle parameter determination module is used to determine the rearview mirror angle parameters based on the driver's body shape information and the seat position parameters.

[0209] The first rearview mirror angle adjustment module is used to adjust the angle of the car rearview mirror based on the rearview mirror angle parameters.

[0210] Optionally, the car seat adjustment device provided in this embodiment of the invention may further include: an identity information acquisition module, an identity information query module, a second seat position adjustment module, and a second rearview mirror angle adjustment module.

[0211] The identity information acquisition module is used to acquire the driver's identity information based on the facial image, as the target identity information.

[0212] The identity information query module is used to query whether the target identity information exists in the driver information database.

[0213] The driver information database includes several identity information and corresponding seat position parameters and rearview mirror angle parameters for each of the identity information.

[0214] The second seat position adjustment module is used to adjust the position of the car seat based on the seat position parameters corresponding to the target identity information in the driver information database when the target identity information exists in the driver information database.

[0215] The second rearview mirror angle adjustment module is used to adjust the angle of the car rearview mirror based on the rearview mirror angle parameters corresponding to the target identity information in the driver information database.

[0216] The driver information determination module 602 is specifically used to determine the driver's three-dimensional face size information, age, and gender based on the driver's face image when the target identity information does not exist in the driver information database.

[0217] Optionally, the car seat adjustment device provided in this embodiment of the invention may further include a parameter storage module.

[0218] The parameter storage module is used to store the target identity information, the determined seat position parameters, and the determined rearview mirror angle parameters into the driver information database when the target identity information does not exist in the driver information database.

[0219] Optionally, the car seat adjustment device provided in this embodiment of the invention may further include: a coordinate system transformation relationship determination module.

[0220] The coordinate system transformation relationship determination module is used to determine the transformation relationship between the world coordinate system and the pixel coordinate system.

[0221] Optionally, the world coordinate system transformation relationship determination module includes: a checkerboard image sequence acquisition submodule and a coordinate system transformation relationship determination submodule.

[0222] The checkerboard image sequence acquisition submodule is used to acquire checkerboard image sequences.

[0223] The chessboard image sequence consists of several chessboard images corresponding to different seat positions. The chessboard images are images captured by the image acquisition device from a chessboard calibration plate placed horizontally on the horizontal surface of the car seat when the car seat is in the corresponding seat position.

[0224] The coordinate system transformation relationship determination submodule is used to determine the transformation relationship between the world coordinate system and the pixel coordinate system based on the chessboard image sequence.

[0225] Optionally, the coordinate system transformation relationship determination submodule, when determining the transformation relationship between the world coordinate system and the pixel coordinate system based on the checkerboard image sequence, is specifically used for:

[0226] Corner point recognition is performed on each chessboard image in the chessboard image sequence to obtain the corner point recognition results corresponding to each chessboard image in the chessboard image sequence. The corner point recognition results include the positions of several corner points in the corresponding chessboard image in the pixel coordinate system.

[0227] Based on the positions of several corner points on the chessboard calibration board in the world coordinate system, and the corner point recognition results corresponding to each chessboard image in the chessboard image sequence, the transformation relationship between the world coordinate system and the pixel coordinate system is determined.

[0228] Optionally, the coordinate system transformation relationship determination submodule, based on the positions of several corner points on the chessboard calibration board in the world coordinate system and the corner point recognition results corresponding to each chessboard image in the chessboard image sequence, determines the transformation relationship between the world coordinate system and the pixel coordinate system, specifically for:

[0229] Based on the positions of several corner points on the chessboard calibration board in the world coordinate system, and the corner point recognition results corresponding to the two chessboard images in the chessboard image sequence, the initial intrinsic and extrinsic parameters of the image acquisition device are determined, wherein the extrinsic parameter is the relative positional relationship between the image acquisition device and the car seat.

[0230] Construct an objective function, wherein the objective function represents the sum of the absolute values ​​of the position differences corresponding to each corner point in other chessboard images in the chessboard image sequence, wherein the position difference is the difference between the actual position of the corresponding corner point in the world coordinate system and the predicted position in the world coordinate system, and the predicted position of a corner point in the world coordinate system is based on the position of the corner point in the pixel coordinate system, as well as the current intrinsic parameters, extrinsic parameters and distortion coefficients of the image acquisition device.

[0231] By minimizing the objective function, the intrinsic parameters, extrinsic parameters, and distortion coefficients of the image acquisition device are updated until the value of the objective function is less than a preset error threshold.

[0232] Based on the final intrinsic parameters, extrinsic parameters, and distortion coefficients, the transformation relationship between the world coordinate system and the pixel coordinate system is determined.

[0233] The car seat adjustment device provided in this embodiment of the invention can automatically adjust the car seat to a position suitable for the driver's body shape, and can also automatically adjust the angle of the car rearview mirror to a suitable angle for the driver, eliminating the need for manual adjustment by the driver, reducing the driver's preparation time before driving, and providing a better user experience. Furthermore, the car seat adjustment device provided in this embodiment of the invention does not require pre-entry of driver information, making it well-suited for situations where drivers frequently change, such as car rental (where the driver in the driver's seat changes frequently) and taxis (where the passenger in the front passenger seat changes frequently).

[0234] Sixth Embodiment

[0235] This invention also provides a car seat adjustment device; please refer to [link / reference]. Figure 7 The diagram shows the structure of the car seat adjustment device, which may include: at least one processor 701, at least one communication interface 702, at least one memory 703 and at least one communication bus 704;

[0236] In this embodiment of the invention, the number of processor 701, communication interface 702, memory 703 and communication bus 704 is at least one, and processor 701, communication interface 702 and memory 703 communicate with each other through communication bus 704.

[0237] The processor 701 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention.

[0238] The memory 703 may include high-speed RAM, or it may also include non-volatile memory, such as at least one disk storage device;

[0239] The memory stores a program, which the processor can call. The program is used for:

[0240] When the driver sits in the car seat, the driver's facial image is captured;

[0241] Based on the facial image, determine the driver's three-dimensional facial dimensions, age, and gender;

[0242] From the human body proportion template library, obtain a human body proportion template that matches the driver's age and gender as the target human body proportion template, wherein the target human body proportion template contains standard three-dimensional face size information and standard body size information.

[0243] Based on the three-dimensional face size information and the target human body proportion template, the driver's body size information is predicted as the driver's body shape information;

[0244] Based on the body shape information, determine the seat position parameters;

[0245] The position of the car seat is adjusted based on the seat position parameters.

[0246] Optionally, the refined and extended functions of the program can be found in the description above.

[0247] Seventh Embodiment

[0248] This invention also provides a readable storage medium that stores a program suitable for execution by a processor, the program being used for:

[0249] When the driver sits in the car seat, the driver's facial image is captured;

[0250] Based on the facial image, determine the driver's three-dimensional facial dimensions, age, and gender;

[0251] From the human body proportion template library, obtain a human body proportion template that matches the driver's age and gender as the target human body proportion template, wherein the target human body proportion template contains standard three-dimensional face size information and standard body size information.

[0252] Based on the three-dimensional face size information and the target human body proportion template, the driver's body size information is predicted as the driver's body shape information;

[0253] Based on the body shape information, determine the seat position parameters;

[0254] The position of the car seat is adjusted based on the seat position parameters.

[0255] Optionally, the refined and extended functions of the program can be found in the description above.

[0256] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0257] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0258] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not 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

1. A method for adjusting a car seat, characterized in that, include: When the driver sits in the car seat, the driver's facial image is captured; Based on the facial image, determine the driver's three-dimensional facial dimensions, age, and gender; From the human body proportion template library, obtain a human body proportion template that matches the driver's age and gender as the target human body proportion template, wherein the target human body proportion template contains standard three-dimensional face size information and standard body size information. Based on the three-dimensional face size information and the target human body proportion template, the driver's body size information is predicted as the driver's body shape information; Based on the body shape information, determine the seat position parameters; Adjust the position of the car seat based on the seat position parameters; The step of predicting the driver's body size information based on the three-dimensional face size information and the target human body proportion template includes: The size ratio is determined based on the three-dimensional face size information and the standard three-dimensional face size information in the target human body proportion template. The driver's body size information is then determined based on the size ratio and the standard body size information in the target human body proportion template.

2. The car seat adjustment method according to claim 1, characterized in that, Based on the facial image, the driver's three-dimensional facial dimensions are determined, including: Facial feature detection is performed on the face image to obtain the position information of facial feature points in the pixel coordinate system; Based on the position information of the facial feature points in the pixel coordinate system and the predetermined transformation relationship between the world coordinate system and the pixel coordinate system, the position information of the facial feature points in the world coordinate system is determined. Based on the position information of the facial feature points in the world coordinate system, the three-dimensional facial size information of the driver is determined.

3. The method for adjusting a car seat according to any one of claims 1 to 2, characterized in that, Also includes: Based on the body shape information and the seat position parameters, determine the rearview mirror angle parameters; Adjust the angle of the car's rearview mirrors based on the aforementioned rearview mirror angle parameters.

4. The car seat adjustment method according to claim 3, characterized in that, Before determining the driver's three-dimensional facial dimensions, age, and gender based on the facial image, the method further includes: Based on the facial image, the driver's identity information is obtained as the target identity information; The system queries whether the target identity information exists in the driver information database, wherein the driver information database includes several identity information and seat position parameters and rearview mirror angle parameters corresponding to the several identity information respectively; If the target identity information exists in the driver information database, the position of the car seat is adjusted based on the seat position parameters corresponding to the target identity information in the driver information database, and the angle of the car rearview mirror is adjusted based on the rearview mirror angle parameters corresponding to the target identity information in the driver information database. If the target identity information is not found in the driver information database, the process of determining the driver's three-dimensional facial dimensions, age, and gender based on the facial image is performed.

5. The method for adjusting a car seat according to claim 4, characterized in that, Also includes: If the target identity information is not found in the driver information database, the target identity information, along with the determined seat position parameters and the determined rearview mirror angle parameters, are stored in the driver information database.

6. The method for adjusting a car seat according to claim 2, characterized in that, Determining the transformation relationship between the world coordinate system and the pixel coordinate system includes: A checkerboard image sequence is obtained, wherein the checkerboard image sequence consists of several checkerboard images corresponding to different seat positions, and the checkerboard images are images acquired by an image acquisition device from a checkerboard calibration plate placed horizontally on the horizontal surface of the car seat when the car seat is in the corresponding seat position. Based on the chessboard image sequence, the transformation relationship between the world coordinate system and the pixel coordinate system is determined.

7. The method for adjusting a car seat according to claim 6, characterized in that, Determining the transformation relationship between the world coordinate system and the pixel coordinate system based on the chessboard image sequence includes: Corner point recognition is performed on each chessboard image in the chessboard image sequence to obtain the corner point recognition results corresponding to each chessboard image in the chessboard image sequence. The corner point recognition results include the positions of several corner points in the corresponding chessboard image in the pixel coordinate system. Based on the positions of several corner points on the chessboard calibration board in the world coordinate system, and the corner point recognition results corresponding to each chessboard image in the chessboard image sequence, the transformation relationship between the world coordinate system and the pixel coordinate system is determined.

8. The method for adjusting a car seat according to claim 7, characterized in that, The process of determining the transformation relationship between the world coordinate system and the pixel coordinate system based on the positions of several corner points on the checkerboard calibration board in the world coordinate system and the corner point recognition results corresponding to each checkerboard image in the checkerboard image sequence includes: Based on the positions of several corner points on the chessboard calibration board in the world coordinate system, and the corner point recognition results corresponding to the two chessboard images in the chessboard image sequence, the initial intrinsic and extrinsic parameters of the image acquisition device are determined, wherein the extrinsic parameter is the relative positional relationship between the image acquisition device and the car seat. Construct an objective function, wherein the objective function represents the sum of the absolute values ​​of the position differences corresponding to each corner point in other chessboard images in the chessboard image sequence, wherein the position difference is the difference between the actual position of the corresponding corner point in the world coordinate system and the predicted position in the world coordinate system, and the predicted position of a corner point in the world coordinate system is based on the position of the corner point in the pixel coordinate system, as well as the current intrinsic parameters, extrinsic parameters and distortion coefficients of the image acquisition device. By minimizing the objective function, the intrinsic parameters, extrinsic parameters, and distortion coefficients of the image acquisition device are updated until the value of the objective function is less than a preset error threshold. Based on the final intrinsic parameters, extrinsic parameters, and distortion coefficients, the transformation relationship between the world coordinate system and the pixel coordinate system is determined.

9. A car seat adjustment device, characterized in that, include: Image acquisition module, driver information determination module, human body proportion template determination module, body shape information prediction module, seat position parameter determination module, and seat position adjustment module; The image acquisition module is used to acquire the driver's facial image when the driver sits in the car seat; The driver information determination module is used to determine the driver's three-dimensional facial size information, as well as the driver's age and gender, based on the facial image. The human body proportion template determination module is used to obtain a human body proportion template that matches the driver's age and gender from the human body proportion template library as a target human body proportion template, wherein the target human body proportion template contains standard three-dimensional face size information and standard body size information. The body shape information prediction module is used to predict the driver's body size information based on the three-dimensional face size information and the target human body proportion template, as the driver's body shape information; The seat position parameter determination module is used to determine the seat position parameters based on the body shape information; The seat position adjustment module is used to adjust the position of the car seat based on the seat position parameters; Specifically, when the body size prediction module predicts the driver's body size information based on the three-dimensional face size information and the target human body proportion template, it is used for: The size ratio is determined based on the three-dimensional face size information and the standard three-dimensional face size information in the target human body proportion template. The driver's body size information is then determined based on the size ratio and the standard body size information in the target human body proportion template.

10. The automobile seat adjustment device according to claim 9, characterized in that, Also includes: Rearview mirror angle parameter determination module and rearview mirror angle adjustment module; The rearview mirror angle parameter determination module is used to determine the rearview mirror angle parameters based on the body shape information and the seat position parameters; The rearview mirror angle adjustment module is used to adjust the angle of the car rearview mirror based on the rearview mirror angle parameters.

Images