Air supply control methods, devices, fan equipment and readable storage media

Abstract

This application relates to an airflow control method, apparatus, fan device, and readable storage medium, comprising: acquiring an infrared image of the area covered by the airflow of an electric fan device; performing feature segmentation on the infrared image according to a preset image processing model to obtain a target area and a background area, wherein the target area includes a human body; acquiring the location information of the target area to obtain human body location information; and controlling the electric fan device to deliver airflow towards the area where the human body is located according to the human body location information. This application uses an infrared camera on an electric fan to acquire infrared images and uses a preset image processing model to perform image analysis on the infrared images, which can accurately acquire the human body location information within the airflow coverage area and enable the fan device to accurately deliver airflow towards the human body.

Description

Technical Field

[0001] This application relates to the field of intelligent electrical appliance technology, and in particular to an air supply control method, device, fan equipment, and readable storage medium. Background Technology

[0002] With the development of technology, smart homes are gradually becoming a part of modern life. Among them, smart home appliances, as an important component of smart homes, directly affect the user experience due to their level of intelligence.

[0003] While some smart electric fans on the market now have certain adaptive adjustment functions, most rely on temperature sensors or simple motion detection devices, which cannot accurately locate the human body and thus affect the user experience. Summary of the Invention

[0004] Therefore, it is necessary to provide an air supply control method, device, fan equipment, and readable storage medium that can accurately locate the human body position and perform air supply processing according to the human body position, in order to address the above-mentioned technical problems.

[0005] In a first aspect, this application provides an air supply control method, including:

[0006] Acquire infrared images of the area covered by the airflow from the electric fan;

[0007] The infrared image is segmented according to a preset image processing model to obtain a target region and a background region, wherein the target region is the region including the human body;

[0008] Obtain the location information of the target area to obtain the human body location information;

[0009] The electric fan is controlled to direct airflow toward the area where the person is located, based on the person's location information.

[0010] In one embodiment, the step of segmenting the infrared image according to a preset image processing model to obtain a target region and a background region includes:

[0011] The infrared image is segmented into target and background regions based on the target convolutional neural network model, wherein the target convolutional neural network model performs convolution processing on the infrared image using a cosine similarity algorithm.

[0012] In one embodiment, processing the infrared image according to the target convolutional neural network model to obtain human body location information includes:

[0013] The infrared image is segmented using a threshold to obtain the target image and the background image;

[0014] The target image and the background image are processed sequentially using the target convolutional kernel, activation layer and max pooling layer to obtain the target feature vector and the background feature vector;

[0015] The target feature vector and the background feature vector are input into a target binary classification problem classifier to obtain the target region and the background region;

[0016] The human body location information is obtained based on the location information of the target area.

[0017] In one embodiment, thresholding the infrared image to obtain a target image and a background image includes:

[0018] Obtain at least one feature threshold, wherein the feature threshold is related to the target human body;

[0019] The grayscale data in the infrared image is segmented according to the feature threshold to obtain a target image and a background image, wherein the target image includes a target human body.

[0020] In one embodiment, the step of sequentially processing the target image and the background image according to the target convolutional kernel, activation layer, and max pooling layer to obtain target feature vector and background feature vector includes:

[0021] The cosine similarity algorithm is used as the target convolution kernel to perform convolution processing on the target image and the background image;

[0022] The ReLU function is used as the activation function of the activation layer to activate the convolutional image;

[0023] The activated image is then subjected to feature extraction using the max pooling layer to obtain target features and background features.

[0024] The target feature and the background feature are flattened according to a preset vector specification to obtain the target feature vector and the background feature vector.

[0025] In one embodiment, the target binary classification problem classifier is a sigmoid classifier.

[0026] In one embodiment, controlling the electric fan to direct airflow toward the area where the person is located according to the human body location information includes:

[0027] Obtain the location information of all human bodies within the area covered by the airflow;

[0028] When the area covered by the blowing air includes the location information of at least two human bodies, the target head-shaking path is planned based on the human body location information.

[0029] The electric fan's steering mechanism is controlled according to the target oscillation path to deliver oscillating air to all human bodies within the airflow coverage area.

[0030] Secondly, this application also provides an air supply control device, comprising:

[0031] The image acquisition module is used to acquire infrared images of the area covered by the airflow from the electric fan.

[0032] The feature segmentation module is used to segment the infrared image according to a preset image processing model to obtain a target region and a background region, wherein the target region is a region including the human body;

[0033] The location acquisition module is used to acquire the location information of the target area to obtain the human body location information;

[0034] The air supply control module is used to control the electric fan to deliver air towards the area where the human body is located, according to the human body location information.

[0035] Thirdly, this application also provides a fan device, including a memory and a processor. The memory stores a computer program, and the fan device further includes an infrared camera connected to the processor. The infrared camera is used to acquire infrared images, and the processor executes the computer program to implement the steps of the air supply control method described in the first aspect.

[0036] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the air supply control method described in the first aspect.

[0037] In summary, this application proposes an airflow control method, device, fan equipment, and readable storage medium, comprising: acquiring an infrared image of the area covered by the airflow of an electric fan equipment; performing feature segmentation on the infrared image according to a preset image processing model to obtain a target area and a background area, wherein the target area includes a human body; acquiring the location information of the target area to obtain human body location information; and controlling the electric fan equipment to deliver airflow towards the area where the human body is located according to the human body location information. This application acquires infrared images by using an infrared camera on an electric fan and performs image analysis on the infrared images using a preset image processing model, which can accurately acquire the human body location information within the airflow coverage area and enable the fan equipment to accurately deliver airflow towards the human body. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of the fan device in one embodiment;

[0039] Figure 2This is a flowchart illustrating the air supply control method in one embodiment;

[0040] Figure 3 This is a flowchart illustrating the steps of obtaining human body location information according to a preset image processing model in one embodiment.

[0041] Figure 4 This is a schematic diagram of the steps for thresholding an image in one embodiment;

[0042] Figure 5 This is a flowchart illustrating the steps of obtaining target feature vectors and background feature vectors based on a target convolutional neural network model in one embodiment.

[0043] Figure 6 This is a flowchart illustrating the air supply control method in another embodiment;

[0044] Figure 7 This is a schematic diagram illustrating an application scenario of the air supply control method in one embodiment;

[0045] Figure 8 This is a structural block diagram of the air supply control device in one embodiment;

[0046] Figure 9 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0048] In one embodiment, such as Figure 1 As shown, a fan device is provided, in which an infrared camera is installed on the body of the fan device. The infrared camera can be used to acquire infrared images of the area covered by the air blown by the fan device.

[0049] In this embodiment, the fan device also includes components such as a controller, a fan head, a steering mechanism, a body, and a base. The steering mechanism enables remote control of the fan head in the horizontal direction and vertical oscillation control. In one feasible embodiment, the steering mechanism can achieve 360° oscillation control of the fan head by releasing the limiting component. The airflow control method in this embodiment uses infrared image recognition to accurately determine the position of a person within the fan device's airflow coverage area, thereby achieving flexible oscillation airflow within the coverage area and greatly improving the oscillation airflow experience of the fan device.

[0050] The controller in this embodiment includes a memory and a processor. The memory stores a computer program, and the processor is connected to an infrared camera. The processor executes the steps of the air supply control method in the following embodiments.

[0051] In one embodiment, such as Figure 2 As shown, an air supply control method is provided, which is applied to... Figure 1 Taking the fan device in the example, the following steps are included:

[0052] S201, acquire an infrared image of the area covered by the airflow from the electric fan.

[0053] S202, the infrared image is segmented according to a preset image processing model to obtain the target region and the background region, wherein the target region is the region including the human body.

[0054] S203, Obtain the location information of the target area to obtain the human body location information.

[0055] S204, control the electric fan equipment to blow air towards the area where the human body is located according to the human body's location information.

[0056] In this embodiment, infrared images of the area covered by the airflow can be acquired using an infrared camera mounted on the electric fan. Alternatively, infrared signals within the airflow coverage area can be collected using an infrared sensor located near the electric fan or an infrared sensor mounted on the electric fan itself, and then converted into infrared images. This embodiment does not limit the actual method of acquiring infrared images.

[0057] In feasible embodiments, if the electric fan acquires infrared images through an infrared sensor or infrared camera located near the area where the electric fan is located, the electric fan needs to be equipped with a communication module for data communication with the device mounted on the infrared sensor or infrared camera. The specific type of communication module can be configured according to the needs of the actual application scenario.

[0058] In this embodiment, the preset image processing model can accurately extract image features and extract target features corresponding to the fan device control mode. For example, if the fan device is in a control mode that tracks a human body to deliver air, the preset image processing model will extract the human body from the infrared image, divide it into a target area including the human body and a background area excluding the human body, obtain the human body position information through the target area, and finally control the air delivery towards the area involved in the human body position information.

[0059] If the fan is in control mode that delivers air based on a target human body, the preset image processing model will extract the target human body from the infrared image and divide it into a target area that includes the target human body and a background area that does not include the target human body. The background area may contain non-target human bodies. Then, the target human body's position information is obtained through the target area, and finally, airflow control is applied to the area defined by the human body's position information.

[0060] It should be noted that, in this embodiment, the target human body refers to a human body with certain characteristic types, such as adults, children, men, women, and the elderly. Non-target human bodies refer to human bodies other than the target human body.

[0061] In summary, this embodiment provides an airflow control method that uses an infrared camera to capture images of the human body within the airflow coverage area of ​​a fan device, achieving precise location of the body and then controlling the fan device to track the body and implement oscillating airflow control. The airflow control method proposed in this embodiment not only greatly improves the accuracy of human body location acquisition but also enables a more intelligent and flexible oscillating airflow control method, ensuring that the fan's airflow direction always changes with the body's position, providing users with a superior airflow experience.

[0062] In one embodiment, the infrared image is segmented according to a preset image processing model to obtain a target region and a background region, including:

[0063] The infrared image is segmented based on the target convolutional neural network model to obtain the target region and the background region. The target convolutional neural network model uses the cosine similarity algorithm to perform convolution processing on the infrared image.

[0064] In this embodiment, the preset image processing model is an optimized Convolutional Neural Network (CNN) model. During infrared image processing, the CNN model can extract features from the infrared image through multiple layers of convolution, pooling, and fully connected layers for subsequent image analysis and recognition tasks. This embodiment uses a CNN model to collect human position information.

[0065] In practical applications, the exact location of people near the fan will constantly change. This embodiment introduces a cosine similarity algorithm into the convolutional layer of the CNN model to compare the similarity of feature vectors in the infrared image, thereby better separating human feature vectors and background feature vectors, achieving better target human feature extraction, and assisting the fan device to more accurately determine the location of the human body.

[0066] In one embodiment, such as Figure 3As shown, infrared images are processed using a target convolutional neural network model to obtain human body location information, including:

[0067] S301, perform threshold segmentation on the infrared image to obtain the target image and background image.

[0068] S302, the target image and background image are processed sequentially according to the target convolutional kernel, activation layer and max pooling layer to obtain target feature vector and background feature vector.

[0069] S303: Input the target feature vector and background feature vector into the target binary classification problem classifier to obtain the target region and background region.

[0070] S304, obtain human body location information based on the location information of the target area.

[0071] In this embodiment, the threshold segmentation algorithm for infrared images can be selected according to the needs of the actual application scenario. By sequentially processing the target image and background image using the optimized target convolutional kernel, activation layer, and max pooling layer, more accurate target feature vectors and background feature vectors can be extracted.

[0072] In this embodiment, the target feature vector can be a feature vector corresponding to a human body. In a feasible embodiment, the target feature vector can be a feature vector corresponding to different types of human bodies. For example, if the human body to be identified is an adult, then the target feature vector is the feature vector corresponding to an adult. If the human body to be identified is a child, then the target feature vector is the feature vector corresponding to a child.

[0073] Background feature vectors can be feature vectors corresponding to the external environment excluding the human body, or they can be the environment excluding the target human body and non-target human bodies. It should be noted that the feature vector content included in the target feature vector and background feature vector can be flexibly configured according to the needs of the actual application scenario.

[0074] In this embodiment, the target binary classification classifier can componentize the feature vector and distinguish between the target region and the background region. It should be noted that the target region can be a region of the human body, including the target type. The target region can be directly distinguished by the binary classification classifier, and then the human body's location information can be obtained based on the location information of the target region. The target binary classification classifier can be selected according to the needs of the actual application scenario. In one embodiment, the target binary classification classifier is a sigmoid classifier. The calculation of the sigmoid function is relatively simple and can handle large-scale datasets. Furthermore, the probability values ​​output by the sigmoid classifier are easy to understand, and the model's parameters (weights) can intuitively explain the influence of each feature on the classification result.

[0075] In one embodiment, such as Figure 4 As shown, threshold segmentation is performed on the infrared image to obtain the target image and background image, including:

[0076] S401, Obtain at least one feature threshold, wherein the feature threshold is related to the target human body.

[0077] S402, perform image segmentation on the grayscale data in the infrared image according to the feature threshold to obtain the target image and the background image, wherein the target image includes the target human body.

[0078] In this embodiment, the target human body is a specified type of human body, such as an adult, child, man, woman, or elderly person. The setting of the feature threshold is directly related to the target human body. This embodiment can achieve precise tracking of the position of a specified type of human body by setting the feature threshold corresponding to the target human body, thereby enabling the fan device to track and blow air.

[0079] It should be noted that the number of feature thresholds can be one or more, and the number of feature thresholds can be determined according to the accuracy required for the threshold segmentation algorithm in the actual application scenario.

[0080] In one embodiment, such as Figure 5 As shown, the target image and background image are processed sequentially using the target convolutional kernel, activation layer, and max pooling layer to obtain target feature vectors and background feature vectors, including:

[0081] S501 uses the cosine similarity algorithm as the target convolution kernel to perform convolution processing on the target image and the background image.

[0082] S502 uses the ReLU function as the activation function of the activation layer to perform activation processing on the convolutional image.

[0083] S503 extracts features from the activated image using a max pooling layer to obtain target and background features.

[0084] S504, flatten the target features and background features according to the preset vector specifications to obtain the target feature vector and background feature vector.

[0085] In this embodiment, the cosine similarity algorithm is used as the target convolution kernel. The specific implementation method for convolution processing the target image and the background image can be referred to the specific implementation method in the foregoing embodiments, and will not be repeated here.

[0086] Using the Rectified Linear Unit (ReLU) function as the activation function of the activation layer can effectively increase the nonlinearity of the model. The sparse activation characteristics of the ReLU function help reduce the risk of overfitting and improve the model's generalization ability.

[0087] In this embodiment, to obtain the target feature vector and the background feature vector, it is necessary to classify the feature vectors using a target binary classification problem classifier. Therefore, it is also necessary to flatten the target feature and the background feature according to the preset vector specifications to obtain the target feature vector and the background feature vector.

[0088] In this embodiment, the preset vector size can be 1*1. It should be noted that the preset vector size can be determined based on the type of classifier for the objective binary classification problem in the actual application scenario.

[0089] In one embodiment, such as Figure 6 As shown, controlling the electric fan to direct airflow towards the area where the person is located, based on the person's location information, includes:

[0090] S601, obtains the location information of all human bodies within the air blowing coverage area.

[0091] S602, when the area covered by the blowing air includes the location information of at least two human bodies, plans the target head-shaking path based on the human body location information.

[0092] S603 controls the fan's steering mechanism according to the target oscillation path to oscillate and deliver airflow to all people within the airflow coverage area.

[0093] In this embodiment, when the fan device executes the air delivery program, it can achieve flexible oscillation path planning to ensure that when the fan oscillates and delivers air, it can deliver air to all human bodies or target human bodies within the air delivery coverage area.

[0094] In this embodiment, the limiting component of the steering mechanism is a switchable limiting component. When the limiting component is opened, the steering mechanism can rotate 360° in all directions. When the limiting component is closed, the steering mechanism can only rotate in a limited direction, such as the horizontal direction.

[0095] like Figure 7 As shown, the location information of all human bodies within the airflow coverage area includes the location information of user A, user B, user C, and user D. In this embodiment, the fan device can plan the fan oscillation path including the location information of user A, user B, user C, and user D to ensure that the fan device can blow air onto user A, user B, user C, and user D respectively during one oscillation airflow process. One oscillation airflow process is the airflow process where the fan head starts rotating from its current position and returns to its current position.

[0096] Assuming users A and C are the target human bodies, the fan device in this embodiment can plan a fan oscillation path that includes the human body position information of users A and C. At this time, user B is not within the airflow coverage area. It should be noted that user C can be included in the fan oscillation path containing the human body position information of users A and C, or user C can be excluded from the fan oscillation path containing the human body position information of users A and C through adjustment of the steering mechanism.

[0097] In summary, this embodiment provides an airflow control method. The fan device can identify the location information of a person's image based on infrared images, and then adjust the oscillation angle to blow air towards the person's location, achieving airflow following the person's movement. This improves the user experience, ensuring the fan always blows air directly at the user, enhancing comfort. It also enhances the product's intelligence level by optimizing image processing algorithms, reducing the system's computational burden, and improving real-time performance.

[0098] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0099] Based on the same inventive concept, this application also provides an air supply control device for implementing the air supply control method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more air supply control device embodiments provided below can be found in the limitations of the air supply control method described above, and will not be repeated here.

[0100] In one embodiment, such as Figure 8 As shown, an air supply control device 800 is provided, including: an image acquisition module 810, a feature segmentation module 820, a position acquisition module 830, and an air supply control module 840, wherein:

[0101] Image acquisition module 810 is used to acquire infrared images of the area covered by the airflow of the electric fan device;

[0102] The feature segmentation module 820 is used to segment the infrared image according to a preset image processing model to obtain the target region and the background region, wherein the target region is the region including the human body;

[0103] The location acquisition module 830 is used to acquire the location information of the target area and obtain the human body location information.

[0104] The air supply control module 840 is used to control the electric fan to deliver air to the area where the person is located based on the person's location information.

[0105] In one embodiment, the feature segmentation module 820 is specifically used to segment the infrared image according to the target convolutional neural network model to obtain the target region and the background region. The target convolutional neural network model uses the cosine similarity algorithm to perform convolution processing on the infrared image.

[0106] In one embodiment, the feature segmentation module 820 is specifically used to perform threshold segmentation on the infrared image to obtain a target image and a background image; process the target image and the background image sequentially according to the target convolution kernel, activation layer and max pooling layer to obtain target feature vector and background feature vector; input the target feature vector and the background feature vector into a target binary classification problem classifier to obtain the target region and the background region; and obtain human body position information based on the position information of the target region.

[0107] In one embodiment, the feature segmentation module 820 is specifically used to obtain at least one feature threshold, wherein the feature threshold is related to the target human body; and to perform image segmentation on the grayscale data in the infrared image according to the feature threshold to obtain a target image and a background image, wherein the target image includes the target human body.

[0108] In one embodiment, the feature segmentation module 820 is specifically used to perform convolution processing on the target image and the background image by using the cosine similarity algorithm as the target convolution kernel; to perform activation processing on the convolved image by using the ReLU function as the activation function of the activation layer; to extract features from the activated image through the max pooling layer to obtain target features and background features; and to flatten the target features and background features according to the preset vector specifications to obtain target feature vector and background feature vector.

[0109] In one embodiment, the air supply control module 840 is specifically used to acquire the location information of all human bodies within the air supply coverage area; when the air supply coverage area includes the location information of at least two human bodies, it plans a target oscillation path based on the human body location information; and controls the steering mechanism of the electric fan according to the target oscillation path to deliver oscillating air to all human bodies within the air supply coverage area.

[0110] In summary, this embodiment provides an airflow control device. The fan can identify the location information of a person's image based on infrared images, and then adjust its oscillation angle to blow air towards the person's location, achieving airflow following the person's movement. This improves the user experience, ensuring the fan always blows air directly at the user, enhancing comfort. It also enhances the product's intelligence level by optimizing image processing algorithms, reducing the system's computational burden, and improving real-time performance.

[0111] Each module in the aforementioned air supply control device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0112] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 9 As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements an airflow control method. The display unit is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0113] Those skilled in the art will understand that Figure 9 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0114] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0115] Acquire infrared images of the area covered by the airflow from the electric fan;

[0116] The infrared image is segmented according to a preset image processing model to obtain a target region and a background region, wherein the target region is the region including the human body;

[0117] Obtain the location information of the target area to obtain the human body location information;

[0118] The electric fan is controlled to direct airflow toward the area where the person is located, based on the person's location information.

[0119] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0120] Acquire infrared images of the area covered by the airflow from the electric fan;

[0121] The infrared image is segmented according to a preset image processing model to obtain a target region and a background region, wherein the target region is the region including the human body;

[0122] Obtain the location information of the target area to obtain the human body location information;

[0123] The electric fan is controlled to direct airflow toward the area where the person is located, based on the person's location information.

[0124] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0125] Acquire infrared images of the area covered by the airflow from the electric fan;

[0126] The infrared image is segmented according to a preset image processing model to obtain a target region and a background region, wherein the target region is the region including the human body;

[0127] Obtain the location information of the target area to obtain the human body location information;

[0128] The electric fan is controlled to direct airflow toward the area where the person is located, based on the person's location information.

[0129] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0130] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0131] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0132] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An air supply control method, characterized in that, include: Acquire infrared images of the area covered by the airflow from the electric fan; The infrared image is segmented based on the target convolutional neural network model to obtain the target region and the background region. The target region includes the human body. The target convolutional neural network model uses the cosine similarity algorithm to perform convolution processing on the infrared image. Obtain the location information of the target area, and obtain the human body location information based on the location information of the target area; The electric fan is controlled to direct airflow toward the area where the person is located, based on the person's location information. The step of segmenting the infrared image based on the target convolutional neural network model to obtain the target region and the background region includes: Obtain at least one feature threshold, wherein the feature threshold is related to the target human body; The grayscale data in the infrared image is segmented according to the feature threshold to obtain a target image and a background image, wherein the target image includes a target human body; The target image and the background image are processed sequentially using the target convolutional kernel, activation layer and max pooling layer to obtain the target feature vector and the background feature vector; The target feature vector and the background feature vector are input into a target binary classification problem classifier to obtain the target region and the background region.

2. The method according to claim 1, characterized in that, The step of processing the target image and the background image sequentially according to the target convolutional kernel, activation layer, and max pooling layer to obtain target feature vector and background feature vector includes: The cosine similarity algorithm is used as the target convolution kernel to perform convolution processing on the target image and the background image; The ReLU function is used as the activation function of the activation layer to activate the convolutional image; The activated image is then subjected to feature extraction using the max pooling layer to obtain target features and background features. The target feature and the background feature are flattened according to a preset vector specification to obtain the target feature vector and the background feature vector.

3. The method according to claim 1, characterized in that, The classifier for the target binary classification problem is the sigmoid classifier.

4. The method according to claim 1, characterized in that, The step of controlling the electric fan to direct airflow toward the area where the person is located according to the human body location information includes: Obtain the location information of all human bodies within the area covered by the airflow; When the area covered by the blowing air includes the location information of at least two human bodies, the target head-shaking path is planned based on the human body location information. The electric fan's steering mechanism is controlled according to the target oscillation path to deliver oscillating air to all human bodies within the airflow coverage area.

5. An air supply control device, characterized in that, include: The image acquisition module is used to acquire infrared images of the area covered by the airflow from the electric fan. The feature segmentation module is used to segment the infrared image according to the target convolutional neural network model to obtain the target region and the background region. The target region is the region including the human body. The target convolutional neural network model uses the cosine similarity algorithm to perform convolution processing on the infrared image. The location acquisition module is used to acquire the location information of the target area and obtain the human body location information based on the location information of the target area. The air supply control module is used to control the electric fan to deliver air towards the area where the human body is located according to the human body location information. The feature segmentation module is further configured to obtain at least one feature threshold, wherein the feature threshold is related to the target human body; The grayscale data in the infrared image is segmented according to the feature threshold to obtain a target image and a background image, wherein the target image includes a target human body; The target image and the background image are processed sequentially using the target convolutional kernel, activation layer and max pooling layer to obtain the target feature vector and the background feature vector; The target feature vector and the background feature vector are input into a target binary classification problem classifier to obtain the target region and the background region.

6. A fan device, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, The fan device further includes an infrared camera connected to the processor. The infrared camera is used to acquire infrared images. When the processor executes the computer program, it implements the steps of the air supply control method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the air supply control method according to any one of claims 1 to 4.

Images