Driving seat self-adaptive adjusting method, system and device based on machine vision and medium

Abstract

The invention discloses a driver seat self-adaptive adjusting method, system and device based on machine vision and a medium, and the method comprises the steps: when a vehicle is started and a brake pedal is stepped down, obtaining stress distribution time sequence data and displacement time sequence data of the brake pedal and facial expression time sequence data of a driver; inputting the stress distribution time sequence data, the displacement time sequence data and the facial expression time sequence data into a pre-trained driver seat posture deviation recognition model to obtain a target driver seat posture deviation; and according to the target driver seat posture deviation, posture adjustment is conducted on the driver seat of the vehicle. The deviation between the current posture and the optimal posture of the driver seat can be automatically recognized according to the stress distribution and displacement of the brake pedal and the facial expression of the user when the user starts the vehicle, so that the posture of the driver seat is automatically adjusted, manual operation of the user is not needed, the user can go out quickly, the driving experience of the user is improved, and the user experience is improved. The method can be applied to the technical field of vehicle control.

Description

Technical Field

[0001] The present invention relates to the field of vehicle control technology, and in particular to a method, system, device and medium for adaptively adjusting a driver's seat based on machine vision. Background Art

[0002] In current cars, due to the different heights and body shapes of different drivers, users may need to readjust the driver's seat posture before starting to drive after entering the driver's seat, which affects the user's driving experience.

[0003] In the existing technology, although some cars have a seat memory function that can save the best seat posture set by the user, the seat memory function can only store a limited number of preset seat postures (such as 3). When faced with scenarios where there are many people with driving rights and drivers change frequently (such as shared cars and car rental scenarios), the seat memory function cannot effectively play a role. After entering the driving seat, the user still needs to readjust the seat posture to meet his or her driving habits, which brings inconvenience to users who are in urgent need of travel. Summary of the Invention

[0004] The purpose of the present invention is to solve one of the technical problems existing in the prior art to at least a certain extent.

[0005] To this end, one purpose of an embodiment of the present invention is to provide a method for adaptively adjusting the driver's seat based on machine vision. The method can automatically identify the deviation between the current posture of the driver's seat and the optimal posture according to the force distribution and displacement of the brake pedal and the user's facial expression when the user starts the vehicle, thereby automatically adjusting the posture of the driver's seat without the need for manual operation by the user, facilitating quick travel for the user and improving the user's driving experience.

[0006] Another object of an embodiment of the present invention is to provide a driver's seat adaptive adjustment system based on machine vision.

[0007] In order to achieve the above technical objectives, the technical solutions adopted by the embodiments of the present invention include:

[0008] In a first aspect, an embodiment of the present invention provides a method for adaptively adjusting a driver's seat based on machine vision, comprising the following steps:

[0009] When the vehicle is started and the brake pedal is depressed, obtaining time series data of force distribution and displacement of the brake pedal and time series data of the driver's facial expression;

[0010] Inputting the force distribution time series data, the displacement time series data, and the facial expression time series data into a pre-trained driver's seat posture deviation recognition model to obtain a target driver's seat posture deviation;

[0011] The posture of the driver's seat of the vehicle is adjusted according to the target driver's seat posture deviation.

[0012] Furthermore, in one embodiment of the present invention, the acquiring of the target passenger's image information, the acquiring of the brake pedal's force distribution time series data, the displacement time series data, and the driver's facial expression time series data, specifically includes:

[0013] Obtaining force changes at various positions of the brake pedal through a pressure sensor array provided on the brake pedal to obtain the force distribution time series data;

[0014] Acquiring displacement time series data of the brake pedal through a displacement sensor provided on the brake pedal;

[0015] Cockpit image information is acquired by a camera device arranged in the cockpit, face detection is performed on the cockpit image information to obtain a facial expression image, and the facial expression time series data is obtained according to a plurality of consecutive frames of the facial expression image.

[0016] Furthermore, in one embodiment of the present invention, the driver's seat posture deviation recognition model is trained by the following steps:

[0017] Obtain the initial driver's seat posture parameters of the test vehicle;

[0018] Obtaining a time series sample of force distribution and displacement of the brake pedal and a time series sample of facial expression of the tester when the tester starts the test vehicle in a test scenario;

[0019] Obtaining optimal driver's seat posture parameters after the tester adjusts the driver's seat of the test vehicle to the most comfortable state;

[0020] generating training samples based on the force distribution time series samples, the displacement time series samples, and the facial expression time series samples, determining sample labels based on the posture deviation between the optimal driver's seat posture parameter and the initial driver's seat posture parameter, and then constructing a training data set based on the training samples and the corresponding sample labels;

[0021] The training data set is input into a pre-built CNN-LSTM hybrid neural network for training to obtain the trained driver's seat posture deviation recognition model.

[0022] Furthermore, in one embodiment of the present invention, the CNN-LSTM hybrid neural network includes an input layer, a CNN convolutional layer, a feature fusion layer, an LSTM layer, an attention layer and an output layer, the input layer is used to input the training sample, the CNN convolutional layer is used to extract features from the training sample to obtain local time series features, the feature fusion layer is used to perform feature fusion on the local time series features to obtain fused time series features, the LSTM layer is used to generate a hidden state sequence based on the fused time series features, the attention layer is used to dynamically assign weights to each dimension of the hidden state sequence based on a multi-head self-attention mechanism, and the output layer is used to map the hidden state sequence after dynamic weight assignment to a driver's seat posture deviation recognition result.

[0023] Furthermore, in one embodiment of the present invention, the training data set is input into a pre-built CNN-LSTM hybrid neural network for training to obtain the trained driver's seat posture deviation recognition model, which specifically includes:

[0024] Input the training sample through the input layer;

[0025] Performing feature extraction on the force distribution time series samples, the displacement time series samples, and the facial expression time series samples through the CNN convolutional layer to obtain force distribution time series features, displacement time series features, and facial expression time series features;

[0026] Performing feature fusion on the force distribution time series feature, the displacement time series feature, and the facial expression time series feature through the feature fusion layer to obtain a fused time series feature;

[0027] Generate a hidden state sequence according to the fused time series features through the LSTM layer;

[0028] Dynamically weighting each dimension of the hidden state sequence based on a multi-head self-attention mechanism through the attention layer;

[0029] Mapping the hidden state sequence after dynamic weight allocation to a driver's seat posture deviation recognition result through the output layer;

[0030] Determine a loss value according to the driver's seat posture deviation recognition result and the corresponding sample label;

[0031] The parameters of the CNN-LSTM hybrid neural network are updated according to the loss value through the back propagation algorithm to obtain the trained driver's seat posture deviation recognition model.

[0032] Furthermore, in one embodiment of the present invention, the adjusting the posture of the driver's seat of the vehicle according to the target driver's seat posture deviation specifically includes:

[0033] Obtaining current driver's seat posture parameters of the vehicle;

[0034] determining a target driver's seat posture parameter according to the current driver's seat posture parameter and the target driver's seat posture deviation;

[0035] The driver's seat of the vehicle is posture-adjusted according to the target driver's seat posture parameter.

[0036] Furthermore, in one embodiment of the present invention, the driver's seat adaptive adjustment method further includes the following steps:

[0037] When the gravity sensor arranged inside the driver's seat of the vehicle senses that there is someone on the driver's seat, the pressure sensor array, the displacement sensor and the camera device are activated.

[0038] In a second aspect, an embodiment of the present invention provides a driver's seat adaptive adjustment system based on machine vision, comprising:

[0039] A data acquisition module, configured to acquire time-series data on the force distribution and displacement of the brake pedal and time-series data on the driver's facial expression when the vehicle is started and the brake pedal is depressed;

[0040] a posture deviation recognition module, configured to input the force distribution time series data, the displacement time series data, and the facial expression time series data into a pre-trained driver's seat posture deviation recognition model to obtain a target driver's seat posture deviation;

[0041] A posture adjustment module is used to adjust the posture of the driver's seat of the vehicle according to the target driver's seat posture deviation.

[0042] In a third aspect, an embodiment of the present invention provides a driver's seat adaptive adjustment device based on machine vision, comprising:

[0043] at least one processor;

[0044] at least one memory for storing at least one program;

[0045] When the at least one program is executed by the at least one processor, the at least one processor implements the above-mentioned method for adaptive adjustment of the driver's seat based on machine vision.

[0046] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium storing a program executable by a processor, wherein the program executable by the processor is used to execute the above-mentioned method for adaptive adjustment of the driver's seat based on machine vision when executed by the processor.

[0047] The advantages and benefits of the present invention will be described in part in the following description and will become apparent from the following description or learned through practice of the present invention:

[0048] When the vehicle is started and the brake pedal is depressed, the embodiment of the present invention obtains the brake pedal's force distribution time series data, displacement time series data, and the driver's facial expression time series data, inputs the force distribution time series data, displacement time series data, and facial expression time series data into a pre-trained driver's seat posture deviation recognition model, obtains the target driver's seat posture deviation, and adjusts the vehicle's driver's seat posture according to the target driver's seat posture deviation. When the user starts the vehicle, the embodiment of the present invention can automatically identify the deviation between the current driver's seat posture and the optimal posture based on the brake pedal's force distribution, displacement, and the user's facial expression, thereby automatically adjusting the driver's seat posture without the need for manual operation by the user, facilitating quick travel for the user and improving the user's driving experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following introduction is made to the drawings required for use in the embodiments of the present invention. It should be understood that the drawings introduced below are only for the convenience of clearly describing some embodiments of the technical solutions of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative work.

[0050] Figure 1 A flowchart of a method for adaptively adjusting a driver's seat based on machine vision provided by an embodiment of the present invention;

[0051] Figure 2 A schematic diagram of an implementation scenario of a method for adaptively adjusting a driver's seat based on machine vision provided by an embodiment of the present invention;

[0052] Figure 3 A schematic diagram of the structure of a CNN-LSTM hybrid neural network provided in an embodiment of the present invention;

[0053] Figure 4 A structural block diagram of a driver's seat adaptive adjustment system based on machine vision provided by an embodiment of the present invention;

[0054] Figure 5 This is a structural block diagram of a driver's seat adaptive adjustment device based on machine vision provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0055] The embodiments of the present invention are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention and are not to be construed as limiting the present invention. The step numbers in the following embodiments are provided for ease of explanation only and do not limit the order of the steps. The order of execution of the steps in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0056] In the description of the present invention, "a plurality" means two or more. The terms "first" and "second" are used solely to distinguish technical features and are not to be construed as indicating or implying relative importance, or as implicitly indicating the number of the indicated technical features, or as implicitly indicating the order of the indicated technical features. Furthermore, unless otherwise defined, all technical and scientific terms used herein have the same meanings as commonly understood by those skilled in the art.

[0057] Reference Figure 1 The embodiment of the present invention provides a method for adaptively adjusting a driver's seat based on machine vision, which specifically includes the following steps:

[0058] S101, when the vehicle is started and the brake pedal is depressed, obtaining brake pedal force distribution time series data, displacement time series data, and driver's facial expression time series data;

[0059] S102, inputting the force distribution time series data, the displacement time series data, and the facial expression time series data into a pre-trained driver's seat posture deviation recognition model to obtain a target driver's seat posture deviation;

[0060] S103: Adjust the posture of the driver's seat of the vehicle according to the target driver's seat posture deviation.

[0061] Specifically, the user's stepping area, stepping direction and stepping force when stepping on the brake pedal in different driving seat postures are different, and the displacement of the brake pedal is also different. When the user feels that the action of stepping on the brake pedal does not conform to his or her own habits (for example, the driver's seat is too far back, so the user needs to straighten his or her legs to fully step on the brake pedal), there will be certain changes in micro-expressions. The embodiment of the present invention uses a neural network model to fit the force distribution, displacement of the brake pedal and the deviation of the driver's facial expression and the posture of the driver's seat (the amount to be adjusted) when the vehicle is started and the brake pedal is stepped on. Then, when the user starts the vehicle, it can automatically identify the deviation between the current posture of the driver's seat and the optimal posture based on the force distribution, displacement of the brake pedal and the user's facial expression, thereby automatically adjusting the posture of the driver's seat without the need for manual operation by the user, facilitating quick travel for the user and improving the user's driving experience.

[0062] As an optional embodiment, obtaining the brake pedal force distribution time series data, displacement time series data, and driver's facial expression time series data specifically includes:

[0063] S1011. Obtain force changes at various positions of the brake pedal using a pressure sensor array provided on the brake pedal to obtain time series data of force distribution;

[0064] S1012. Acquire displacement time series data of the brake pedal through a displacement sensor provided on the brake pedal;

[0065] S1013. Acquire cockpit image information through a camera device installed in the cockpit, perform face detection on the cockpit image information to obtain a facial expression image, and obtain facial expression time series data based on multiple consecutive frames of facial expression images.

[0066] like Figure 2 The figure shows a schematic diagram of an implementation scenario of a machine vision-based adaptive driver's seat adjustment method provided by an embodiment of the present invention. In this embodiment of the present invention, a pressure sensor array is provided on the side of the brake pedal facing the driver's seat, a displacement sensor is provided inside the brake pedal, and a camera device is provided in the cockpit (e.g., above the instrument panel). When the vehicle is powered on and the brake pedal is depressed, the pressure sensor array obtains the force changes at various positions of the brake pedal. The force distribution at each moment is sorted to obtain force distribution time series data. The displacement of the brake pedal at each moment is obtained using the displacement sensor to obtain displacement time series data. The camera device obtains cockpit image information, performs face detection on the cockpit image information to obtain a facial expression image, and then obtains facial expression time series data based on multiple consecutive frames of facial expression images.

[0067] As an optional embodiment, the driver's seat posture deviation recognition model is trained by the following steps:

[0068] S201, obtaining initial driver's seat posture parameters of the test vehicle;

[0069] S202, obtaining a time series sample of force distribution and displacement of the brake pedal and a time series sample of facial expression of the tester when the tester starts the test vehicle in a test scenario;

[0070] S203, obtaining optimal driver's seat posture parameters after the tester adjusts the driver's seat of the test vehicle to the most comfortable state;

[0071] S204, generating training samples based on the force distribution time series samples, the displacement time series samples, and the facial expression time series samples, and determining sample labels based on the posture deviation between the optimal driver's seat posture parameters and the initial driver's seat posture parameters, and then constructing a training data set based on the training samples and the corresponding sample labels;

[0072] S205: Input the training data set into a pre-built CNN-LSTM hybrid neural network for training to obtain a trained driver's seat posture deviation recognition model.

[0073] Specifically, first, the initial driver's seat posture parameters of the test vehicle are pre-acquired in the test scenario; then the tester starts the test vehicle and obtains the force distribution time series samples, displacement time series samples of the brake pedal when the test vehicle starts, and the tester's facial expression time series samples; the tester manually adjusts the driver's seat of the test vehicle until it is adjusted to the most comfortable state for himself, and obtains the optimal driver's seat posture parameters at this time; the force distribution time series samples, displacement time series samples and facial expression time series samples are used as training samples, and the sample labels are determined according to the posture deviation between the optimal driver's seat posture parameters and the initial driver's seat posture parameters to obtain a set of training data; multiple tests are carried out under different testers and different initial driver's seat posture parameters, and a training data set can be constructed after obtaining a sufficient amount of training data; the training data set is input into the pre-constructed CNN-LSTM hybrid neural network for training to obtain a trained driver's seat posture deviation recognition model.

[0074] Further as an optional implementation, the CNN-LSTM hybrid neural network includes an input layer, a CNN convolutional layer, a feature fusion layer, an LSTM layer, an attention layer and an output layer. The input layer is used to input training samples, the CNN convolutional layer is used to extract features from the training samples to obtain local time series features, the feature fusion layer is used to fuse the local time series features to obtain fused time series features, the LSTM layer is used to generate a hidden state sequence based on the fused time series features, the attention layer is used to dynamically assign weights to each dimension of the hidden state sequence based on a multi-head self-attention mechanism, and the output layer is used to map the hidden state sequence after dynamic weight assignment to the driver's seat posture deviation recognition result.

[0075] like Figure 3 The figure shows the structure diagram of the CNN-LSTM hybrid neural network provided by an embodiment of the present invention. It can be seen that the embodiment of the present invention sets three CNN convolution layers to extract local time series features of the force distribution time series samples, displacement time series samples and facial expression time series samples respectively. The force distribution time series features, displacement time series features and facial expression time series features obtained will enter the feature fusion layer for feature fusion, and the final fused time series features will be input into the LSTM layer for subsequent recognition.

[0076] As an optional implementation, the training data set is input into a pre-built CNN-LSTM hybrid neural network for training to obtain a trained driver's seat posture deviation recognition model, which specifically includes:

[0077] S2051. Input training samples through the input layer;

[0078] S2052. Perform feature extraction on the force distribution time series samples, the displacement time series samples, and the facial expression time series samples through a CNN convolutional layer to obtain force distribution time series features, displacement time series features, and facial expression time series features;

[0079] S2053, performing feature fusion on the force distribution time series features, the displacement time series features, and the facial expression time series features through a feature fusion layer to obtain a fused time series feature;

[0080] S2054, generating a hidden state sequence based on the fused time series features through the LSTM layer;

[0081] S2055. Dynamically assign weights to each dimension of the hidden state sequence based on the multi-head self-attention mechanism through the attention layer;

[0082] S2056. Mapping the hidden state sequence after dynamic weight allocation to the driver's seat posture deviation recognition result through the output layer;

[0083] S2057. Determine a loss value based on the driver's seat posture deviation recognition result and the corresponding sample label;

[0084] S2058. Update the parameters of the CNN-LSTM hybrid neural network through the back propagation algorithm according to the loss value to obtain a trained driver's seat posture deviation recognition model.

[0085] Specifically, the fused time series features are input into the LSTM layer to generate the corresponding hidden state sequence, and the dimensions of the hidden state sequence are dynamically weighted based on the multi-head self-attention mechanism, and then mapped into the driver's seat posture deviation recognition result; the loss value is determined according to the difference between the driver's seat posture deviation recognition result and the sample label, and the selection of the loss function is not limited in the embodiment of the present invention; the parameters of the CNN-LSTM hybrid neural network are updated according to the loss value through the back propagation algorithm, and then the next round of iterative training is entered. When the preset convergence condition is reached (the number of iterations reaches the threshold and the loss value is lower than the preset threshold), the training is stopped to obtain a trained driver's seat posture deviation recognition model.

[0086] As a further optional embodiment, the posture of the driver's seat of the vehicle is adjusted according to the target driver's seat posture deviation, which specifically includes:

[0087] S1031. Obtaining the current driver's seat posture parameters of the vehicle;

[0088] S1032, determining a target driver's seat posture parameter based on the current driver's seat posture parameter and the target driver's seat posture deviation;

[0089] S1033. Adjust the posture of the driver's seat of the vehicle according to the target driver's seat posture parameter.

[0090] Specifically, the real-time force distribution time series data, displacement time series data and facial expression time series data are input into a pre-trained driver's seat posture deviation recognition model to obtain the target driver's seat posture deviation; the current driver's seat posture parameters of the vehicle are obtained, and the target driver's seat posture parameters required by the user are determined based on the current driver's seat posture parameters and the target driver's seat posture deviation, and the vehicle's driver's seat posture can be adjusted according to the target driver's seat posture parameters.

[0091] It should be noted that there is a certain time interval between the user starting the vehicle and the user driving the vehicle. The embodiment of the present invention collects relevant data when the user starts the vehicle, and then uses the driver's seat posture deviation recognition model that has been trained and deployed on the vehicle side to identify the target driver's seat posture deviation, and quickly completes the adaptive posture adjustment of the driver's seat before the user drives the vehicle. The entire process does not require manual adjustment by the user and will not delay the user's travel.

[0092] As an optional embodiment, the driver's seat adaptive adjustment method further includes the following steps:

[0093] S100: When a gravity sensor disposed inside a driver's seat of a vehicle senses that there is someone in the driver's seat, a pressure sensor array, a displacement sensor, and a camera device are activated.

[0094] Specifically, if Figure 2 As shown, in an embodiment of the present invention, a gravity sensor is provided inside the driver's seat. When the gravity value sensed by the gravity sensor is greater than a certain threshold, it is determined that there is someone in the driver's seat. At this time, the pressure sensor array, displacement sensor and camera device are started, and relevant data are acquired when the vehicle is started and the brake pedal is stepped on, thereby avoiding the pressure sensor array, displacement sensor and camera device from being in a working state for a long time, reducing equipment loss and power consumption.

[0095] The above describes the method steps of an embodiment of the present invention. It can be appreciated that the embodiment of the present invention can automatically identify the deviation between the current driver's seat posture and the optimal posture based on the force distribution and displacement of the brake pedal and the user's facial expression when the user starts the vehicle, thereby automatically adjusting the driver's seat posture without the need for manual operation by the user, facilitating quick travel for the user and improving the user's driving experience.

[0096] Reference Figure 4 , an embodiment of the present invention provides a driver's seat adaptive adjustment system based on machine vision, comprising:

[0097] A data acquisition module is used to obtain time series data on the force distribution and displacement of the brake pedal and the driver's facial expression when the vehicle starts and the brake pedal is depressed;

[0098] A posture deviation recognition module is used to input the force distribution time series data, displacement time series data, and facial expression time series data into a pre-trained driver's seat posture deviation recognition model to obtain a target driver's seat posture deviation;

[0099] The posture adjustment module is used to adjust the posture of the vehicle's driver's seat according to the target driver's seat posture deviation.

[0100] The contents of the above method embodiments are all applicable to the present system embodiments. The functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0101] Reference Figure 5 , an embodiment of the present invention provides a driver's seat adaptive adjustment device based on machine vision, comprising:

[0102] at least one processor;

[0103] at least one memory for storing at least one program;

[0104] When the at least one program is executed by the at least one processor, the at least one processor implements the machine vision-based adaptive adjustment method for the driver's seat.

[0105] The contents of the above method embodiments are all applicable to the present device embodiments. The functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0106] An embodiment of the present invention also provides a computer-readable storage medium, which stores a program executable by a processor. When the program is executed by the processor, it is used to execute the above-mentioned method for adaptive adjustment of the driver's seat based on machine vision.

[0107] A computer-readable storage medium according to an embodiment of the present invention can execute a driver's seat adaptive adjustment method based on machine vision provided by an embodiment of the method according to the present invention, can execute any combination of implementation steps of the embodiment of the method, and has the corresponding functions and beneficial effects of the method.

[0108] The embodiment of the present invention also discloses a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device can read the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs Figure 1 The method shown.

[0109] In some optional embodiments, the function / operation mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the function / operation involved, the two boxes shown in succession can actually be executed substantially simultaneously or the above-mentioned boxes can sometimes be executed in reverse order. In addition, the embodiment presented and described in the flow chart of the present invention is provided in an exemplary manner for the purpose of providing a more comprehensive understanding of the technology. The disclosed method is not limited to the operation and logic flow presented herein. Optional embodiments are contemplated in which the order of the various operations is changed and the sub-operations described as a part of a larger operation are performed independently.

[0110] In addition, although the present invention is described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the above-mentioned functions and / or features can be integrated into a single physical device and / or software module, or one or more functions and / or features can be implemented in separate physical devices or software modules. It is also understood that a detailed discussion of the actual implementation of each module is not necessary for understanding the present invention. More specifically, given the properties, functions, and internal relationships of the various functional modules in the devices disclosed herein, the actual implementation of the module will be understood within the routine skills of an engineer. Therefore, a person skilled in the art can implement the present invention set forth in the claims using ordinary skills without undue experimentation. It is also understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present invention, which is determined by the full scope of the appended claims and their equivalents.

[0111] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the above methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program code.

[0112] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0113] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable media on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0114] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0115] In the above description of this specification, reference to the terms "one embodiment / example," "another embodiment / example," or "certain embodiments / examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0116] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the claims and their equivalents.

[0117] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the above embodiments. Those skilled in the art can make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A driver's seat adaptive adjustment method based on machine vision, characterized in that: The following steps are involved: When the vehicle is started and the brake pedal is depressed, obtaining time series data of force distribution and displacement of the brake pedal and time series data of the driver's facial expression; Inputting the force distribution time series data, the displacement time series data, and the facial expression time series data into a pre-trained driver's seat posture deviation recognition model to obtain a target driver's seat posture deviation; The posture of the driver's seat of the vehicle is adjusted according to the target driver's seat posture deviation.

2. The method for adaptively adjusting a driver's seat based on machine vision according to claim 1, characterized in that: The obtaining of the brake pedal force distribution time series data, displacement time series data, and driver's facial expression time series data specifically includes: Obtaining force changes at various positions of the brake pedal through a pressure sensor array provided on the brake pedal to obtain the force distribution time series data; Acquiring displacement time series data of the brake pedal through a displacement sensor provided on the brake pedal; Cockpit image information is acquired by a camera device arranged in the cockpit, face detection is performed on the cockpit image information to obtain a facial expression image, and the facial expression time series data is obtained according to a plurality of consecutive frames of the facial expression image.

3. The method for adaptively adjusting a driver's seat based on machine vision according to claim 1, characterized in that: The driver's seat posture deviation recognition model is trained by the following steps: Obtain the initial driver's seat posture parameters of the test vehicle; Obtaining a time series sample of force distribution and displacement of the brake pedal and a time series sample of facial expression of the tester when the tester starts the test vehicle in a test scenario; Obtaining optimal driver's seat posture parameters after the tester adjusts the driver's seat of the test vehicle to the most comfortable state; generating training samples based on the force distribution time series samples, the displacement time series samples, and the facial expression time series samples, determining sample labels based on the posture deviation between the optimal driver's seat posture parameter and the initial driver's seat posture parameter, and then constructing a training data set based on the training samples and the corresponding sample labels; The training data set is input into a pre-built CNN-LSTM hybrid neural network for training to obtain the trained driver's seat posture deviation recognition model.

4. The method for adaptively adjusting a driver's seat based on machine vision according to claim 3, characterized in that: The CNN-LSTM hybrid neural network includes an input layer, a CNN convolutional layer, a feature fusion layer, an LSTM layer, an attention layer and an output layer. The input layer is used to input the training sample, the CNN convolutional layer is used to extract features from the training sample to obtain local time series features, the feature fusion layer is used to fuse the local time series features to obtain fused time series features, the LSTM layer is used to generate a hidden state sequence based on the fused time series features, the attention layer is used to dynamically assign weights to each dimension of the hidden state sequence based on a multi-head self-attention mechanism, and the output layer is used to map the hidden state sequence after dynamic weight assignment to a driver's seat posture deviation recognition result.

5. The method for adaptively adjusting a driver's seat based on machine vision according to claim 4, characterized in that: The training data set is input into a pre-built CNN-LSTM hybrid neural network for training to obtain the trained driver's seat posture deviation recognition model, which specifically includes: Input the training sample through the input layer; Performing feature extraction on the force distribution time series samples, the displacement time series samples, and the facial expression time series samples through the CNN convolutional layer to obtain force distribution time series features, displacement time series features, and facial expression time series features; Performing feature fusion on the force distribution time series feature, the displacement time series feature, and the facial expression time series feature through the feature fusion layer to obtain a fused time series feature; Generate a hidden state sequence according to the fused time series features through the LSTM layer; Dynamically weighting each dimension of the hidden state sequence based on a multi-head self-attention mechanism through the attention layer; Mapping the hidden state sequence after dynamic weight allocation to a driver's seat posture deviation recognition result through the output layer; Determine a loss value according to the driver's seat posture deviation recognition result and the corresponding sample label; The parameters of the CNN-LSTM hybrid neural network are updated according to the loss value through the back propagation algorithm to obtain the trained driver's seat posture deviation recognition model.

6. The method for adaptively adjusting a driver's seat based on machine vision according to claim 1, characterized in that: The step of adjusting the posture of the driver's seat of the vehicle according to the target driver's seat posture deviation specifically includes: Obtaining current driver's seat posture parameters of the vehicle; determining a target driver's seat posture parameter according to the current driver's seat posture parameter and the target driver's seat posture deviation; The driver's seat of the vehicle is posture-adjusted according to the target driver's seat posture parameter.

7. The method for adaptively adjusting a driver's seat based on machine vision according to claim 2, characterized in that: The driver's seat adaptive adjustment method further comprises the following steps: When the gravity sensor arranged inside the driver's seat of the vehicle senses that there is someone on the driver's seat, the pressure sensor array, the displacement sensor and the camera device are activated.

8. A driver's seat adaptive adjustment system based on machine vision, characterized in that: include: A data acquisition module, configured to acquire time-series data on the force distribution and displacement of the brake pedal and time-series data on the driver's facial expression when the vehicle is started and the brake pedal is depressed; a posture deviation recognition module, configured to input the force distribution time series data, the displacement time series data, and the facial expression time series data into a pre-trained driver's seat posture deviation recognition model to obtain a target driver's seat posture deviation; A posture adjustment module is used to adjust the posture of the driver's seat of the vehicle according to the target driver's seat posture deviation.

9. A driver's seat adaptive adjustment device based on machine vision, characterized in that: include: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the driver's seat adaptive adjustment method based on machine vision as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a program executable by a processor, characterized in that: The program executable by the processor is used to execute the driver's seat adaptive adjustment method based on machine vision as described in any one of claims 1 to 7 when executed by the processor.

Images