Vehicle-mounted air conditioner adjusting method, system and device based on machine vision and medium

Abstract

The invention discloses a vehicle-mounted air conditioner adjusting method, system and device based on machine vision and a medium. The method comprises the steps that cabin position information, body surface temperature distribution information and human body image information of a target passenger are obtained; determining physical feature information and sitting posture information of the target passenger according to the human body image information; inputting the body surface temperature distribution information, the physical feature information and the sitting posture information into a pre-trained air outlet control parameter prediction model to obtain target air outlet control parameters; and a target air outlet at the corresponding position is started according to the cockpit position information, and the air outlet pitch angle, the air outlet spread angle, the air outlet amount and the air outlet temperature of the target air outlet are controlled according to the target air outlet control parameters. The method improves the adjustment precision of the vehicle-mounted air conditioner and the riding experience of the user, and can be widely 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 vehicle air conditioning adjustment method, system, device and medium based on machine vision. Background Art

[0002] With the rapid development of technology and the increasing intelligence of automobiles, the comfort of the in-car environment has become a key factor affecting the driving experience. As a key device for improving in-car comfort, in-car air conditioning consists of a compressor, condenser, throttling element, evaporator, fan, and necessary control components. It is primarily used to regulate the temperature and humidity in the vehicle, providing a comfortable riding environment for passengers.

[0003] Traditional automotive air-conditioning systems mainly rely on manual adjustment of the direction and air volume of the air outlet. This method is not only time-consuming and labor-intensive, but also unable to precisely meet the personalized needs of each passenger. Especially in multi-passenger scenarios, manual adjustment cannot take into account the comfort needs of all passengers.

[0004] In recent years, existing intelligent technologies can preset plans to achieve modes such as blowing towards people, blowing away people, and sweeping the air. However, the preset plans are relatively rigid and have the disadvantage of being unable to identify passengers of different body shapes and ages. They are also unable to achieve intelligent and precise control of the air direction, air volume, and air temperature, affecting the user's riding experience. Summary of the Invention

[0005] 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.

[0006] To this end, an object of an embodiment of the present invention is to provide a vehicle air conditioning adjustment method based on machine vision, which improves the accuracy of vehicle air conditioning adjustment and the user's riding experience.

[0007] Another object of an embodiment of the present invention is to provide a vehicle air conditioning control system based on machine vision.

[0008] In order to achieve the above technical objectives, the technical solutions adopted by the embodiments of the present invention include:

[0009] In a first aspect, an embodiment of the present invention provides a vehicle air conditioning adjustment method based on machine vision, comprising the following steps:

[0010] Obtain the target occupant's cabin position information, body surface temperature distribution information, and human body image information;

[0011] determining the physical feature information and sitting posture information of the target occupant based on the human body image information;

[0012] Inputting the body surface temperature distribution information, the body feature information, and the sitting posture information into a pre-trained air outlet control parameter prediction model to obtain target air outlet control parameters;

[0013] The target air outlet at the corresponding position is activated according to the cabin position information, and the air outlet pitch angle, air outlet diffusion angle, air outlet volume and air outlet temperature of the target air outlet are controlled according to the target air outlet control parameters.

[0014] Furthermore, in one embodiment of the present invention, obtaining the cabin position information, body surface temperature distribution information, and human body image information of the target occupant specifically includes:

[0015] Acquiring infrared image information of the target cockpit by using an infrared camera disposed in the target cockpit;

[0016] determining the cabin position information and the body surface temperature distribution information of a plurality of the target occupants based on the infrared image information;

[0017] A visible light camera at a corresponding position is activated according to the cockpit position information, and the human body image information of the target occupant is acquired through the visible light camera.

[0018] Furthermore, in one embodiment of the present invention, determining the physical feature information and sitting posture information of the target occupant based on the human body image information specifically includes:

[0019] performing face detection on the human body image information to obtain facial image information, and identifying the gender information and age information of the target occupant based on the facial image information;

[0020] Performing human body key point detection on the human body image information to obtain human body key point information, and determining the body shape information and the sitting posture information of the target occupant based on the human body key point information;

[0021] The physical feature information is determined according to the gender information, the age information and the body shape information.

[0022] Furthermore, in one embodiment of the present invention, the air outlet control parameter prediction model is trained by the following steps:

[0023] Obtaining body surface temperature distribution samples, physical characteristics samples, and sitting posture samples of the test person in the test vehicle;

[0024] Adjusting the air outlet pitch angle, air outlet diffusion angle, air outlet volume, and air outlet temperature of the air outlet corresponding to the test person, and recording the test person's subjective feelings to obtain the optimal air outlet control parameters that make the test person most comfortable;

[0025] Using the body surface temperature distribution sample, the body feature sample, and the sitting posture sample as first training samples, and determining corresponding air outlet control parameter labels according to the optimal air outlet control parameters;

[0026] Inputting the first training sample into a pre-built convolutional neural network to obtain a predicted value of the air outlet control parameter;

[0027] determining a first loss value according to the air outlet control parameter prediction value and the air outlet control parameter label;

[0028] The parameters of the convolutional neural network are updated according to the first loss value to obtain the trained air outlet control parameter prediction model.

[0029] Furthermore, in one embodiment of the present invention, the vehicle air conditioning adjustment method further includes the following steps:

[0030] Obtaining time series data of the surface temperature distribution and sitting posture of the target occupant during the current period;

[0031] Inputting the body surface temperature distribution time series data and the sitting posture time series data into a pre-trained air outlet adjustment parameter prediction model to obtain a target air outlet adjustment parameter;

[0032] The air outlet pitch angle, air outlet diffusion angle, air outlet volume and air outlet temperature of the target air outlet are adjusted according to the target air outlet adjustment parameters.

[0033] Furthermore, in one embodiment of the present invention, the air outlet adjustment parameter prediction model is trained by the following steps:

[0034] Adjusting the air outlet pitch angle, air outlet diffusion angle, air outlet volume, and air outlet temperature of the air outlet corresponding to the test person in the test vehicle, and recording the test person's subjective feelings to obtain the initial air outlet control parameters that make the test person most comfortable;

[0035] Obtaining a time series sample of the body surface temperature distribution and a time series sample of the sitting posture of the test person within a preset time period;

[0036] After the preset period, the air outlet pitch angle, air outlet diffusion angle, air outlet volume, and air outlet temperature of the air outlet corresponding to the test person are adjusted, and the test person's subjective feeling is recorded to obtain the current air outlet control parameters that make the test person most comfortable;

[0037] Determining an optimal air outlet adjustment parameter according to a parameter change between the current air outlet control parameter and the initial air outlet control parameter;

[0038] Using the body surface temperature distribution time series samples and the sitting posture time series samples as second training samples, and determining corresponding air outlet adjustment parameter labels according to the optimal air outlet adjustment parameters;

[0039] Inputting the second training sample into a pre-built CNN-LSTM hybrid neural network to obtain a predicted value of the air outlet adjustment parameter;

[0040] determining a second loss value according to the air outlet adjustment parameter prediction value and the air outlet adjustment parameter label;

[0041] The parameters of the CNN-LSTM hybrid neural network are updated according to the second loss value to obtain the trained air outlet adjustment parameter prediction model.

[0042] Furthermore, in one embodiment of the present invention, the CNN-LSTM hybrid neural network includes an input layer, a CNN convolution layer, a feature fusion layer, an LSTM layer, an attention layer, and an output layer. The second training sample is input into the pre-built CNN-LSTM hybrid neural network to obtain the predicted value of the air outlet adjustment parameter, which specifically includes:

[0043] Inputting the second training sample through the input layer;

[0044] Performing feature extraction on the body surface temperature distribution time series samples and the sitting posture time series samples through the CNN convolutional layer to obtain body surface temperature distribution time series features and sitting posture time series features;

[0045] Performing feature fusion on the body surface temperature distribution time series feature and the sitting posture time series feature through the feature fusion layer to obtain a fused time series feature;

[0046] Generate a hidden state sequence according to the fused time series features through the LSTM layer;

[0047] Dynamically weighting each dimension of the hidden state sequence based on a multi-head self-attention mechanism through the attention layer;

[0048] The hidden state sequence after dynamic weight allocation is mapped to the predicted value of the air flow adjustment parameter through the output layer.

[0049] In a second aspect, an embodiment of the present invention provides a vehicle air conditioning adjustment system based on machine vision, comprising:

[0050] An information acquisition module is used to obtain the cabin position information, body surface temperature distribution information and human body image information of the target occupant;

[0051] An image recognition module, configured to determine the physical features and sitting posture information of the target occupant based on the human body image information;

[0052] an air flow control parameter prediction module, configured to input the body surface temperature distribution information, the body feature information, and the sitting posture information into a pre-trained air flow control parameter prediction model to obtain target air flow control parameters;

[0053] The air outlet control module is used to start the target air outlet at the corresponding position according to the cabin position information, and control the air outlet pitch angle, air outlet diffusion angle, air outlet volume and air outlet temperature of the target air outlet according to the target air outlet control parameters.

[0054] In a third aspect, an embodiment of the present invention provides a vehicle air conditioning control device based on machine vision, comprising:

[0055] at least one processor;

[0056] at least one memory for storing at least one program;

[0057] When the at least one program is executed by the at least one processor, the at least one processor implements the above-mentioned vehicle air conditioning adjustment method based on machine vision.

[0058] 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 vehicle air conditioning adjustment method based on machine vision when executed by the processor.

[0059] 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:

[0060] The embodiment of the present invention obtains the target occupant's cabin position information, body surface temperature distribution information, and human image information, determines the target occupant's physical characteristics and sitting posture information based on the human image information, inputs the body surface temperature distribution information, physical characteristics information, and sitting posture information into a pre-trained air outlet control parameter prediction model to obtain target air outlet control parameters, activates the target air outlet at the corresponding position based on the cabin position information, and controls the target air outlet pitch angle, air outlet diffusion angle, air volume, and air outlet temperature based on the target air outlet control parameters. The embodiment of the present invention predicts the optimal air outlet control parameters for the air-conditioning outlet at the corresponding position based on the occupant's body surface temperature distribution, physical characteristics, and sitting posture, thereby automatically controlling and adjusting the air-conditioning outlet without the need for manual operation by the user, thereby improving the accuracy of vehicle air conditioning adjustment and the user's riding experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] 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.

[0062] Figure 1 A flowchart of a vehicle air conditioning adjustment method based on machine vision provided by an embodiment of the present invention;

[0063] Figure 2 Another flowchart of a vehicle air conditioning adjustment method based on machine vision provided by an embodiment of the present invention;

[0064] Figure 3 A structural block diagram of a vehicle air conditioning control system based on machine vision provided by an embodiment of the present invention;

[0065] Figure 4 This is a structural block diagram of a vehicle air conditioning control device based on machine vision provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0066] The embodiments of the present invention are described in detail below, examples of which 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.

[0067] 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.

[0068] 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:

[0069] S101, obtaining cockpit position information, body surface temperature distribution information, and human body image information of the target occupant;

[0070] S102, determining the physical characteristics and sitting posture information of the target occupant based on the human body image information;

[0071] S103, inputting the body surface temperature distribution information, the body feature information, and the sitting posture information into a pre-trained air outlet control parameter prediction model to obtain target air outlet control parameters;

[0072] S104: Activate the target air outlet at the corresponding position according to the cockpit position information, and control the air outlet pitch angle, air outlet diffusion angle, air outlet volume, and air outlet temperature of the target air outlet according to the target air outlet control parameters.

[0073] Specifically, users with different physical characteristics have different air outlet requirements for air conditioning vents, and users with different body surface temperature distributions or different sitting postures also have different air outlet requirements for air conditioning vents. The embodiment of the present invention uses a neural network model to fit the user's body surface temperature distribution, body shape characteristics and sitting posture with their actual air outlet requirements, and then predicts the optimal air outlet control parameters of the air conditioning outlet at the corresponding position based on the occupant's body surface temperature distribution, body shape characteristics and sitting posture, so that the air conditioning outlet can be automatically controlled and adjusted without the need for manual operation by the user, thereby improving the accuracy of the vehicle air conditioning adjustment and the user's riding experience.

[0074] As an optional implementation, obtaining the cabin position information, body surface temperature distribution information, and human body image information of the target occupant specifically includes:

[0075] S1011. Acquire infrared image information of the target cockpit through an infrared camera disposed in the target cockpit;

[0076] S1012. Determine cabin position information and body surface temperature distribution information of several target occupants based on the infrared image information;

[0077] S1013: Activate a visible light camera at a corresponding position according to the cockpit position information, and obtain human body image information of the target occupant through the visible light camera.

[0078] As a further optional implementation, determining the target occupant's physical features and sitting posture information based on the human body image information specifically includes:

[0079] S1021. Performing face detection on the human body image information to obtain facial image information, and identifying the gender and age information of the target occupant based on the facial image information;

[0080] S1022: Detecting key points of the human body image to obtain key point information of the human body, and determining the body shape and sitting posture information of the target occupant based on the key point information of the human body;

[0081] S1023. Determine physical feature information based on gender information, age information, and body shape information.

[0082] Specifically, this embodiment of the present invention uses an infrared camera installed in the vehicle to capture the occupant's cabin position and body surface temperature distribution in real time. Simultaneously, based on the cabin position, a visible light camera is activated at a corresponding location to capture the occupant's body image. Using image processing, face detection, and body key point detection, the system identifies the occupant's physical characteristics (gender, age, body type, etc.) and sitting posture. Furthermore, this embodiment of the present invention monitors changes in the occupant's body surface temperature distribution and sitting posture in real time.

[0083] As an optional implementation, the air output control parameter prediction model is trained by the following steps:

[0084] S201, obtaining a body surface temperature distribution sample, a body feature sample, and a sitting posture sample of a test person in a test vehicle;

[0085] S202, adjusting the air outlet pitch angle, air outlet diffusion angle, air outlet volume, and air outlet temperature of the air outlet corresponding to the test person, and recording the test person's subjective feelings to obtain the optimal air outlet control parameters that make the test person most comfortable;

[0086] S203: Using the body surface temperature distribution sample, the body feature sample, and the sitting posture sample as the first training sample, and determining the corresponding air outlet control parameter label according to the optimal air outlet control parameter;

[0087] S204: Input the first training sample into a pre-built convolutional neural network to obtain a predicted value of an air outlet control parameter;

[0088] S205: Determine a first loss value based on the predicted value of the air outlet control parameter and the air outlet control parameter label;

[0089] S206: Update the parameters of the convolutional neural network according to the first loss value to obtain a trained air outlet control parameter prediction model.

[0090] Specifically, first, the surface temperature distribution samples, physical feature samples and sitting posture samples of the test person in the test vehicle are obtained; then, the air outlet pitch angle, air outlet diffusion angle, air outlet volume and air outlet temperature of the air outlet corresponding to the test person are adjusted, and the subjective feelings of the test person are recorded to obtain the optimal air outlet control parameters that make the test person most comfortable; the surface temperature distribution samples, physical feature samples and sitting posture samples are used as the first training samples, and the corresponding air outlet control parameter labels are determined according to the optimal air outlet control parameters to obtain a set of training data; multiple tests are carried out under different test persons, different surface temperature distribution samples and different sitting posture samples, and a first training data set can be constructed after obtaining a sufficient amount of training data; the first training data set is input into a pre-constructed convolutional neural network for training to obtain a trained air outlet control parameter prediction model.

[0091] The system of the embodiment of the present invention can predict the target air outlet control parameters of the corresponding air-conditioning outlet according to the occupant's physical characteristics, sitting posture information and real-time monitored body surface temperature distribution by training the air outlet control parameter prediction model, including the air outlet pitch angle, air outlet diffusion angle, air volume and air outlet temperature, to ensure that the air outlet direction and air volume can accurately meet the occupant's comfort needs; then, the target air outlet at the corresponding position is started according to the determined cabin position information, and the air outlet pitch angle, air outlet diffusion angle, air volume and air outlet temperature of the target air outlet is controlled according to the target air outlet control parameters, so that the air-conditioning outlet at the corresponding position can be controlled and adjusted as soon as the user gets on the vehicle.

[0092] It should be noted that the embodiment of the present invention can be used in a vehicle use scenario with a single passenger, and can also be used in a vehicle use scenario with multiple passengers. When the number of passengers is not less than 2, the optimal air outlet control parameters are predicted for each passenger, and then the air-conditioning outlets at the corresponding positions of each passenger are controlled and adjusted accordingly.

[0093] Reference Figure 2 As an optional embodiment, the vehicle air conditioning adjustment method further includes the following steps:

[0094] S105, obtaining the target occupant's body surface temperature distribution time series data and sitting posture time series data during the current period;

[0095] S106, inputting the body surface temperature distribution time series data and the sitting posture time series data into a pre-trained air flow adjustment parameter prediction model to obtain a target air flow adjustment parameter;

[0096] S107 , adjusting the air outlet pitch angle, air outlet diffusion angle, air outlet volume, and air outlet temperature of the target air outlet according to the target air outlet adjustment parameters.

[0097] Specifically, during the ride, infrared images and body images of the passengers are continuously collected, and the occupants' body surface temperature distribution time series data and sitting posture time series data are determined through continuous multiple frames of infrared images and body images, and then input into a pre-trained air outlet adjustment parameter prediction model to predict the amplitude of dynamic adjustment of the air outlet pitch angle, air outlet diffusion angle, air outlet volume and air outlet temperature. Based on the original predicted optimal air outlet control parameters, the parameters of the air-conditioning outlet are further adjusted dynamically in real time to ensure the comfort of the passengers.

[0098] In some optional embodiments, the air outlet temperature and air volume can be adjusted in a timely manner according to factors such as the operating time of the vehicle's air-conditioning system, the current time point, and the vehicle's driving status to ensure that passengers can enjoy the best comfort experience throughout the journey.

[0099] In some optional embodiments, the occupants' personalized air-conditioning preferences can also be recorded, and personalized learning can be performed based on the occupants' historical data to generate thermal comfort adjustment strategies for specific occupants. When the same occupant is identified again, the system will optimize the air-conditioning operation control parameters in combination with the personalized thermal comfort adjustment strategy to provide a more accurate and personalized comfort experience.

[0100] As an optional implementation, the air flow adjustment parameter prediction model is trained by the following steps:

[0101] S301, adjusting the air outlet pitch angle, air outlet diffusion angle, air outlet volume, and air outlet temperature of the air outlet corresponding to the test person in the test vehicle, and recording the test person's subjective feelings to obtain initial air outlet control parameters that are most comfortable for the test person;

[0102] S302, obtaining a time series sample of the test person's body surface temperature distribution and a time series sample of the sitting posture within a preset time period;

[0103] S303, after a preset period of time, adjusting the air outlet pitch angle, air outlet diffusion angle, air outlet volume, and air outlet temperature of the air outlet corresponding to the test person, and recording the test person's subjective feelings to obtain the current air outlet control parameters that are most comfortable for the test person;

[0104] S304: determining an optimal air outlet adjustment parameter based on a parameter change between the current air outlet control parameter and the initial air outlet control parameter;

[0105] S305: Using the body surface temperature distribution time series samples and the sitting posture time series samples as second training samples, and determining corresponding air flow adjustment parameter labels according to the optimal air flow adjustment parameters;

[0106] S306: Input the second training sample into a pre-built CNN-LSTM hybrid neural network to obtain a predicted value of the air outlet adjustment parameter;

[0107] S307: Determine a second loss value according to the air outlet adjustment parameter prediction value and the air outlet adjustment parameter label;

[0108] S308. Update the parameters of the CNN-LSTM hybrid neural network according to the second loss value to obtain a trained air outlet adjustment parameter prediction model.

[0109] Specifically, first, the air outlet pitch angle, air outlet diffusion angle, air volume and air outlet temperature of the air outlet at the corresponding position of the test person in the test vehicle are adjusted, and the test person's subjective feelings are recorded to obtain the initial air outlet control parameters that make the test person most comfortable; then, the air-conditioning air outlet operation state is kept unchanged within a preset period of time, and the surface temperature distribution time series sample and sitting posture time series sample of the test person within the preset period of time are obtained; and then, after the preset period of time, the air outlet pitch angle, air outlet diffusion angle, air volume and air outlet temperature of the air outlet at the corresponding position of the test person are adjusted, and the test person's subjective feelings are recorded to obtain the current air outlet control parameters that make the test person most comfortable ; Use the surface temperature distribution time series samples and the sitting posture time series samples as the second training samples, determine the optimal air outlet adjustment parameters according to the parameter changes of the current air outlet control parameters and the initial air outlet control parameters, and use the optimal air outlet adjustment parameters as the corresponding air outlet adjustment parameter labels to obtain a set of training data; conduct multiple tests under different test personnel, different surface temperature distribution time series samples, and different sitting posture time series samples, and construct a second training data set after obtaining a sufficient amount of training data; input the second training data set into the pre-constructed CNN-LSTM hybrid neural network for training to obtain a trained air outlet adjustment parameter prediction model.

[0110] As an optional implementation, the CNN-LSTM hybrid neural network includes an input layer, a CNN convolution layer, a feature fusion layer, an LSTM layer, an attention layer, and an output layer. The second training sample is input into the pre-built CNN-LSTM hybrid neural network to obtain a predicted value of the air outlet adjustment parameter, which specifically includes:

[0111] S3061. Input a second training sample through the input layer;

[0112] S3062. Perform feature extraction on the body surface temperature distribution time series samples and the sitting posture time series samples through a CNN convolutional layer to obtain body surface temperature distribution time series features and sitting posture time series features;

[0113] S3063. Perform feature fusion on the body surface temperature distribution time series features and the sitting posture time series features through a feature fusion layer to obtain a fused time series feature.

[0114] S3064, generate a hidden state sequence based on the fused time series features through the LSTM layer;

[0115] S3065. Dynamically assign weights to each dimension of the hidden state sequence based on the multi-head self-attention mechanism through the attention layer;

[0116] S3066. Map the hidden state sequence after dynamic weight allocation to the predicted value of the air flow adjustment parameter through the output layer.

[0117] Specifically, the embodiment of the present invention sets up two CNN convolution layers to extract local time series features of the body surface temperature distribution time series samples and the sitting posture time series samples respectively. The obtained body surface temperature distribution time series features and sitting posture time series features 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; after the fused time series features are input into the LSTM layer, the corresponding hidden state sequence is generated, and the various dimensions of the hidden state sequence are dynamically weighted based on the multi-head self-attention mechanism, and then mapped to the wind adjustment parameter prediction value; 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 wind adjustment parameter prediction model.

[0118] The above describes the method steps of the embodiment of the present invention. It is understandable that the embodiment of the present invention predicts the optimal air outlet control parameters of the air-conditioning outlet at the corresponding position based on the occupant's body surface temperature distribution, physical characteristics and sitting posture, so that the air-conditioning outlet can be automatically controlled and adjusted without the need for manual operation by the user, thereby improving the accuracy of the vehicle air-conditioning adjustment and the user's riding experience; in addition, the optimal air outlet adjustment parameters of the air-conditioning outlet at the corresponding position are predicted based on the occupant's body surface temperature distribution time series data and sitting posture time series data, so that the air-conditioning outlet can be continuously adjusted during the user's ride, further improving the accuracy of the vehicle air-conditioning adjustment and the user's riding experience.

[0119] Reference Figure 3 , an embodiment of the present invention provides a vehicle air conditioning adjustment system based on machine vision, comprising:

[0120] An information acquisition module is used to obtain the cabin position information, body surface temperature distribution information and human body image information of the target occupant;

[0121] An image recognition module is used to determine the target occupant's physical characteristics and sitting posture information based on the human body image information;

[0122] The air flow control parameter prediction module is used to input body surface temperature distribution information, body feature information, and sitting posture information into a pre-trained air flow control parameter prediction model to obtain target air flow control parameters;

[0123] The air outlet control module is used to activate the target air outlet at the corresponding position according to the cabin position information, and control the air outlet pitch angle, air outlet diffusion angle, air outlet volume and air outlet temperature of the target air outlet according to the target air outlet control parameters.

[0124] 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.

[0125] Reference Figure 4 , an embodiment of the present invention provides a vehicle air conditioning adjustment device based on machine vision, comprising:

[0126] at least one processor;

[0127] at least one memory for storing at least one program;

[0128] When the at least one program is executed by the at least one processor, the at least one processor implements the vehicle air conditioning adjustment method based on machine vision.

[0129] 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.

[0130] An embodiment of the present invention further provides a computer-readable storage medium storing a program executable by a processor. When the program is executed by the processor, it is used to execute the above-mentioned vehicle air conditioning adjustment method based on machine vision.

[0131] A computer-readable storage medium according to an embodiment of the present invention can execute a vehicle air conditioning adjustment method based on machine vision provided by an embodiment of the method of 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.

[0132] 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.

[0133] 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.

[0134] 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.

[0135] 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.

[0136] 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.

[0137] 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.

[0138] 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.

[0139] 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.

[0140] 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.

[0141] 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 vehicle air conditioning adjustment method based on machine vision, characterized in that: The following steps are involved: Obtain the target occupant's cabin position information, body surface temperature distribution information, and human body image information; determining the physical feature information and sitting posture information of the target occupant based on the human body image information; Inputting the body surface temperature distribution information, the body feature information, and the sitting posture information into a pre-trained air outlet control parameter prediction model to obtain target air outlet control parameters; The target air outlet at the corresponding position is activated according to the cabin position information, and the air outlet pitch angle, air outlet diffusion angle, air outlet volume and air outlet temperature of the target air outlet are controlled according to the target air outlet control parameters.

2. The vehicle air conditioning adjustment method based on machine vision according to claim 1, characterized in that: The step of obtaining the cabin position information, body surface temperature distribution information, and human body image information of the target occupant specifically includes: Acquiring infrared image information of the target cockpit by using an infrared camera disposed in the target cockpit; determining the cabin position information and the body surface temperature distribution information of a plurality of the target occupants based on the infrared image information; A visible light camera at a corresponding position is activated according to the cockpit position information, and the human body image information of the target occupant is acquired through the visible light camera.

3. The vehicle air conditioning adjustment method based on machine vision according to claim 1, characterized in that: The determining of the physical feature information and sitting posture information of the target occupant based on the human body image information specifically includes: performing face detection on the human body image information to obtain facial image information, and identifying the gender information and age information of the target occupant based on the facial image information; Performing human body key point detection on the human body image information to obtain human body key point information, and determining the body shape information and the sitting posture information of the target occupant based on the human body key point information; The physical feature information is determined according to the gender information, the age information and the body shape information.

4. The vehicle air conditioning adjustment method based on machine vision according to claim 1, characterized in that: The air output control parameter prediction model is trained by the following steps: Obtaining body surface temperature distribution samples, physical characteristics samples, and sitting posture samples of the test person in the test vehicle; Adjusting the air outlet pitch angle, air outlet diffusion angle, air outlet volume, and air outlet temperature of the air outlet corresponding to the test person, and recording the test person's subjective feelings to obtain the optimal air outlet control parameters that make the test person most comfortable; Using the body surface temperature distribution sample, the body feature sample, and the sitting posture sample as first training samples, and determining corresponding air outlet control parameter labels according to the optimal air outlet control parameters; Inputting the first training sample into a pre-built convolutional neural network to obtain a predicted value of the air outlet control parameter; determining a first loss value according to the air outlet control parameter prediction value and the air outlet control parameter label; The parameters of the convolutional neural network are updated according to the first loss value to obtain the trained air outlet control parameter prediction model.

5. A vehicle air conditioning adjustment method based on machine vision according to any one of claims 1 to 4, characterized in that: The vehicle air conditioning adjustment method further comprises the following steps: Obtaining time series data of the surface temperature distribution and sitting posture of the target occupant during the current period; Inputting the body surface temperature distribution time series data and the sitting posture time series data into a pre-trained air outlet adjustment parameter prediction model to obtain a target air outlet adjustment parameter; The air outlet pitch angle, air outlet diffusion angle, air outlet volume and air outlet temperature of the target air outlet are adjusted according to the target air outlet adjustment parameters.

6. The vehicle air conditioning adjustment method based on machine vision according to claim 5, characterized in that: The air flow adjustment parameter prediction model is trained by the following steps: Adjusting the air outlet pitch angle, air outlet diffusion angle, air outlet volume, and air outlet temperature of the air outlet corresponding to the test person in the test vehicle, and recording the test person's subjective feelings to obtain the initial air outlet control parameters that make the test person most comfortable; Obtaining a time series sample of the body surface temperature distribution and a time series sample of the sitting posture of the test person within a preset time period; After the preset period, the air outlet pitch angle, air outlet diffusion angle, air outlet volume, and air outlet temperature of the air outlet corresponding to the test person are adjusted, and the test person's subjective feeling is recorded to obtain the current air outlet control parameters that make the test person most comfortable; Determining an optimal air outlet adjustment parameter according to a parameter change between the current air outlet control parameter and the initial air outlet control parameter; Using the body surface temperature distribution time series samples and the sitting posture time series samples as second training samples, and determining corresponding air outlet adjustment parameter labels according to the optimal air outlet adjustment parameters; Inputting the second training sample into a pre-built CNN-LSTM hybrid neural network to obtain a predicted value of the air outlet adjustment parameter; determining a second loss value according to the air outlet adjustment parameter prediction value and the air outlet adjustment parameter label; The parameters of the CNN-LSTM hybrid neural network are updated according to the second loss value to obtain the trained air outlet adjustment parameter prediction model.

7. The vehicle air conditioning adjustment method based on machine vision according to claim 6, characterized in that: The CNN-LSTM hybrid neural network includes an input layer, a CNN convolution layer, a feature fusion layer, an LSTM layer, an attention layer, and an output layer. The second training sample is input into the pre-built CNN-LSTM hybrid neural network to obtain the predicted value of the air outlet adjustment parameter, which specifically includes: Inputting the second training sample through the input layer; Performing feature extraction on the body surface temperature distribution time series samples and the sitting posture time series samples through the CNN convolutional layer to obtain body surface temperature distribution time series features and sitting posture time series features; Performing feature fusion on the body surface temperature distribution time series feature and the sitting posture 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; The hidden state sequence after dynamic weight allocation is mapped to the predicted value of the air flow adjustment parameter through the output layer.

8. A vehicle air conditioning control system based on machine vision, characterized in that: include: An information acquisition module is used to obtain the cabin position information, body surface temperature distribution information and human body image information of the target occupant; An image recognition module, configured to determine the physical features and sitting posture information of the target occupant based on the human body image information; an air flow control parameter prediction module, configured to input the body surface temperature distribution information, the body feature information, and the sitting posture information into a pre-trained air flow control parameter prediction model to obtain target air flow control parameters; The air outlet control module is used to start the target air outlet at the corresponding position according to the cabin position information, and control the air outlet pitch angle, air outlet diffusion angle, air outlet volume and air outlet temperature of the target air outlet according to the target air outlet control parameters.

9. A vehicle air conditioning control 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 vehicle air conditioning 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 processor-executable program is used to execute the vehicle air conditioning adjustment method based on machine vision as claimed in any one of claims 1 to 7 when executed by the processor.

Images