An active noise reduction control method for a range hood based on expression recognition, a range hood, and a storage medium

Abstract

The embodiment of the application discloses a kind of range hood active noise reduction control methods based on expression identification, range hood and storage medium, it is related to range hood technical field, wherein method includes: the face image information of user in the preset area is obtained by the camera module;According to the expression identification of the face image information, obtain the emotional information of user;According to the emotional information, judge the auditory comfort of user;If the result of judgment in step S3 is uncomfortable, then the noise reduction parameter of the active noise reduction module is increased by the first preset step, until the result of judgment in the step S3 is comfortable.The method of the present application realizes the dynamic matching of noise reduction effect and user subjective feeling, and automatically adapts the optimal noise reduction parameter according to individual differences, overcoming the defect that open-loop control cannot adapt to working condition changes.

Description

Technical Field

[0001] This invention relates to the field of range hood technology, and in particular to an active noise reduction control method for a range hood based on facial expression recognition, a range hood, and a storage medium. Background Technology

[0002] As an indispensable appliance in modern kitchens, the core function of a range hood is to promptly extract cooking fumes and exhaust gases to maintain clean kitchen air. To meet users' demands for high-intensity, high-load smoke extraction, existing range hoods generally optimize extraction efficiency by increasing fan airflow and pressure parameters. However, this design directly results in high-intensity, wide-bandgap operating noise from the fan system during high-speed operation. Prolonged exposure to this noise not only severely impacts users' cooking experience and mood but may also pose potential health risks. Therefore, effectively reducing the operating noise of range hoods has become a key technical indicator for evaluating product competitiveness.

[0003] Currently, noise reduction technologies for range hoods are mainly divided into two categories: passive noise reduction and active noise reduction. Passive noise reduction technology primarily achieves noise blocking and absorption by arranging physical structures such as sound-absorbing cotton, damping plates, and sound insulation structures inside the air duct, but its effect on suppressing low-frequency noise is generally poor. Active noise control (ANC) technology has received widespread attention because it can effectively suppress low-frequency noise.

[0004] However, existing active noise reduction solutions for range hoods have significant shortcomings. The control logic, parameter-based noise reduction, and effect judgment of current active noise reduction systems all heavily rely on objective physical acoustic parameters such as sound pressure level and noise frequency collected by acoustic sensors like microphones. Noise reduction control focuses solely on lowering the physical decibel value of noise, and most employ open-loop control modes, which have significant limitations. On one hand, objective acoustic parameters cannot adequately match the user's subjective auditory experience. The noise reduction system cannot capture the user's true feelings and emotional feedback regarding the noise reduction effect in real time. Even if the noise level meets the standard after reduction, certain frequency bands of noise may still cause user dissatisfaction, which the noise reduction system cannot detect and adjust accordingly. Therefore, it cannot make the noise reduction effect match the user's actual experience, lacking a closed-loop noise reduction mechanism centered on the user's subjective feeling. On the other hand, different users have significant individual differences in their sensitivity and tolerance to noise. Uniform physical noise reduction parameters cannot adapt to the personalized noise reduction needs of different users, resulting in a discrepancy between the noise reduction effect and the user's actual comfort experience. Summary of the Invention

[0005] In view of this, the present invention provides an active noise reduction control method for a range hood based on facial expression recognition, a range hood and a storage medium, aiming to solve the problems in the background art mentioned above.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A method for active noise reduction control of a range hood based on facial expression recognition, wherein the range hood includes a camera module and an active noise reduction module, and the method includes the following steps: S1. Obtain facial image information of users within a preset area through the camera module; S2. Perform facial expression recognition based on the facial image information to obtain the user's emotional information; S3. Determine the user's auditory comfort level based on the emotional information; S4. If the judgment result in step S3 is uncomfortable, then increase the noise reduction parameters of the active noise reduction module by the first preset step size until the judgment result in step S3 is comfortable.

[0007] Optionally, increasing the noise reduction parameters of the active noise reduction module by a first preset step size includes the following steps: The number of times the judgment result in step S3 is "uncomfortable" within a preset time period is counted. If the number of times exceeds the first preset threshold, the noise reduction parameters of the active noise reduction module are increased by the second preset step size; otherwise, the noise reduction parameters of the active noise reduction module are increased by the first preset step size, wherein the second preset step size is greater than the first preset step size.

[0008] Optionally, increasing the noise reduction parameters of the active noise reduction module by a first preset step size includes the following steps: If the judgment result in step S3 is discomfort, then the duration of the discomfort result is calculated. If the duration exceeds the second preset threshold, then the noise reduction parameter of the active noise reduction module is increased by the third preset step size. Otherwise, the noise reduction parameter of the active noise reduction module is increased by the first preset step size. The third preset step size is greater than the first preset step size.

[0009] Optionally, the following step may be included between step S1 and step S2: Facial recognition is performed based on the facial image information to obtain the user's identity; Determine whether the user's identity belongs to a pre-stored user; If not, proceed directly to the next steps and collect the operating data of the active noise cancellation module in real time. Bind and store the operating data with the user identity, wherein the operating data includes at least the personalized noise cancellation parameters of the current user. If so, the personalized noise reduction parameters bound to the user identity are obtained, and the active noise reduction module is controlled to operate under the personalized noise reduction parameters bound to the user identity.

[0010] Optionally, before the step of performing facial recognition based on the facial image information to obtain the user's identity, the following steps are further included: In response to the power-on information, the current gear of the range hood is obtained and the noise reduction parameter corresponding to the current gear is obtained from the pre-stored gear-noise reduction parameter mapping table; The active noise cancellation module is controlled to operate under the noise cancellation parameters corresponding to the current gear.

[0011] Optionally, if the judgment result in step S3 is uncomfortable, then adjusting the noise reduction parameters of the active noise reduction module by a first preset step size until the judgment result in step S3 is comfortable includes the following steps: If the judgment result in step S3 is discomfort, then determine whether there is a gear shift; If there is a gear change, it is determined whether there is a personalized noise reduction parameter associated with the changed gear. If there is, the active noise reduction module is controlled to run under the personalized noise reduction parameter. If there is no personalized noise reduction parameter, the active noise reduction module is controlled to run under the noise reduction parameter corresponding to the changed gear. The noise reduction parameter corresponding to the changed gear is obtained from the pre-stored gear-noise reduction parameter mapping table. Proceed to step S1. If there is no gear shift, the noise reduction parameters of the active noise reduction module are increased by a first preset step size until the judgment result in step S3 is comfortable.

[0012] Optionally, before responding to the power-on information, obtaining the current setting of the range hood and retrieving the noise reduction parameter corresponding to the current setting from the pre-stored setting-noise reduction parameter mapping table, the method further includes: Control the range hood to operate at each speed setting and collect noise signals during the operation of each speed setting; The noise reduction parameters corresponding to each gear level are determined based on the noise signals of each gear level. Establish and pre-store the correspondence between location and noise reduction parameters.

[0013] Optionally, the following steps may be included before step S1: The camera module periodically detects whether there are users within a preset area; If so, proceed to step S1; If no user is detected within a preset number of consecutive detection cycles, the current noise reduction parameters of the active noise reduction module are cached, and the current noise reduction parameters of the active noise reduction module are reduced according to a preset amount. If a user is detected again, the current noise reduction parameters are retrieved and the active noise reduction module is controlled to restore to the current noise reduction parameters.

[0014] Furthermore, to achieve the above objectives, the present invention also provides a range hood, characterized in that the range hood includes a camera module and an active noise reduction module, the active noise reduction module includes a speaker, the speaker is used to emit noise reduction waves to cancel noise, and the range hood applies the active noise reduction control method for range hoods based on facial expression recognition described in any of the preceding claims.

[0015] Furthermore, to achieve the above objectives, the present invention also provides a storage medium containing computer-executable instructions, characterized in that the computer-executable instructions, when executed by a computer processor, are used to execute any of the aforementioned active noise reduction control methods for range hoods based on facial expression recognition.

[0016] The present invention has at least the following beneficial effects: (1) In step S1 of this invention, the user's facial image information is collected in real time through the camera module. In step S2, the user's emotional information is obtained. In step S3, the user's auditory comfort is judged based on the emotional information. In step S4, the noise reduction parameter adjustment is only started when the judgment result is "uncomfortable". After each adjustment, the system returns to step S1 to re-collect images, re-identify expressions, and re-judge comfort until the user's auditory comfort feedback changes from "uncomfortable" to "comfortable". Thus, the active noise reduction control method of this invention no longer aims to reduce the decibel value as the end point, but rather aims to restore the user's emotional comfort as the ultimate goal, achieving a dynamic match between the noise reduction effect and the user's actual feeling.

[0017] (2) This invention achieves automatic adaptation to the personalized noise reduction needs of different users by gradually increasing the noise reduction parameters with a preset step size and monitoring user emotional feedback in real time. This solves the technical problem that different users have individual differences in noise sensitivity and tolerance, and that uniform physical noise reduction parameters cannot adapt to diverse needs. In step S4 of this invention, the noise reduction parameters are gradually increased with a first preset step size, and the auditory comfort is reassessed after each adjustment until the user's auditory comfort is comfortable. This adjustment method ensures that the noise reduction parameters of each user eventually converge to their individual auditory comfort level. Users with high sensitivity will continue to adjust until the noise reduction effect is strong enough, while users with low sensitivity can stop adjusting at an earlier stage. Thus, this invention achieves automatic personalized customization of noise reduction intensity, overcoming the limitation of the existing technology where uniform parameters cannot take into account individual differences.

[0018] (3) The closed-loop control logic of this invention overcomes the technical defects of existing active noise reduction schemes, which mostly use open-loop control and cannot cope with changes in operating conditions. Specifically, existing open-loop schemes output noise reduction waves according to preset parameters, and the noise reduction effect decreases once the noise characteristics change. However, the loop structure consisting of steps S1 to S4 in this invention enables the system to continuously track changes in user emotions and adjust the noise reduction strategy in a timely manner, overcoming the inherent defect of open-loop control being unable to cope with changes in operating conditions.

[0019] (4) This invention can be achieved simply by adding a camera module to the existing range hood and optimizing the control algorithm. There is no need to modify the structure of the air duct, fan, etc. The modification cost is low and it is suitable for large-scale industrial promotion. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] To gain a more complete understanding of this application and its beneficial effects, the following description will be provided in conjunction with the accompanying drawings. In the following description, the same reference numerals denote the same parts.

[0022] Figure 1 This is a flowchart illustrating an active noise reduction control method for a range hood based on facial expression recognition, provided in an embodiment of the present invention. Figure 2 This is a flowchart illustrating an active noise reduction control method for a range hood based on facial expression recognition, provided in an embodiment of the present invention. Figure 3 This is a flowchart illustrating an active noise reduction control method for a range hood based on facial expression recognition, provided in an embodiment of the present invention. Figure 4 This is a flowchart illustrating an active noise reduction control method for a range hood based on facial expression recognition, provided in an embodiment of the present invention. Figure 5 This is a flowchart illustrating an active noise reduction control method for a range hood based on facial expression recognition, provided in an embodiment of the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0024] Furthermore, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0025] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0026] Example 1: This embodiment discloses an active noise reduction control method for range hoods based on facial expression recognition. This method judges the user's auditory comfort level in the current noise environment by recognizing the user's facial expressions in real time, and dynamically adjusts the noise reduction parameters of the active noise reduction module accordingly, forming a closed-loop control system from perception to judgment, then to adjustment, and then back to perception, so as to achieve an adaptive and personalized active noise reduction effect.

[0027] The following is a detailed description of an active noise reduction control method for a range hood based on facial expression recognition, as disclosed in this embodiment. Figure 1 As shown, Figure 1 This is a flowchart illustrating an active noise reduction control method for a range hood based on facial expression recognition, as described in this embodiment. The range hood hardware system adapted in this embodiment includes at least a camera module and an active noise reduction module. The active noise reduction control method for a range hood based on facial expression recognition in this embodiment includes the following steps: Step S1: Obtain facial image information of users within a preset area through the camera module.

[0028] In this embodiment, the camera module acquires facial image information of the user within a preset area. The camera module can be an infrared camera or other imaging device, primarily used for real-time, continuous acquisition of facial image information of the user within the preset area, providing raw image data support for subsequent steps such as expression recognition and obtaining emotional information. The camera module can be installed on the range hood body, for example, on the outer surface of the smoke collection hood. It is important to note that the installation position of the camera module in this embodiment should cover the cooking operation area in front of the range hood; that is, when the user is cooking, the camera module should be directly or approximately directly facing the user's face to acquire clear and identifiable facial image information. The preset area in this embodiment refers to the spatial range within which the camera module can effectively acquire facial images. In one embodiment, the preset area is a three-dimensional space area 0.3 meters to 2.0 meters away from the range hood and 0.8 meters to 2 meters high, covering the area where the user is typically located while cooking. The specific range of the preset area can be preset according to the field of view, installation height, and installation angle of the camera module; this invention does not limit this. In this embodiment, "user" refers to the person currently using the range hood for cooking. "Facial image information" in this embodiment refers to image data containing the user's facial features captured by the camera module. These facial features include at least image data of key facial feature points such as the overall facial contour, eyebrows, eyes, corners of the mouth, and facial muscle state. This facial image information can be a single frame or a sequence of multiple consecutive images.

[0029] Understandably, this step involves continuously collecting facial image information of users within a preset area in real time using the camera module, providing raw image data for subsequent expression recognition.

[0030] Step S2: Perform facial expression recognition based on the facial image information to obtain the user's emotional information.

[0031] In this step, an expression recognition algorithm is used to process and analyze the facial image information obtained in step S1 to identify the user's facial expressions, specifically identifying typical facial features such as frowning, baring teeth, downturned corners of the mouth, facial stiffness, relaxed expression, and relaxed corners of the mouth. Those skilled in the art will understand that facial expression recognition technology is relatively mature, and commonly used expression recognition models include, but are not limited to, classification models based on Convolutional Neural Networks (CNNs) and temporal models based on Recurrent Neural Networks (RNNs), which will not be elaborated upon in this embodiment.

[0032] After recognizing a user's facial expression, the system obtains the user's emotional information based on the recognized facial expression. In one embodiment, the emotional information includes, but is not limited to, the following categories: comfort, pleasure, neutrality, irritability, dissatisfaction, pain, disgust, etc.

[0033] Step S3: Determine the user's auditory comfort level based on the emotional information.

[0034] In this step, the user's auditory comfort is determined based on the emotional information, that is, the emotional information identified in step S2 is mapped to the user's auditory comfort level, i.e., whether the user feels comfortable with the noise environment generated by the range hood. Auditory comfort here refers to the user's subjective feeling and tolerance to the current noise environment, describing whether the user feels comfortable and calm in the current noise environment, or conversely, feels irritable and uncomfortable. In this embodiment, auditory comfort is a binary state quantity, including two states: "comfortable" and "uncomfortable." In other embodiments, it can also be a continuous or discrete quantity with multiple levels.

[0035] In one embodiment, auditory comfort is determined based on matching emotional information with a pre-defined lookup table. Specifically, an emotion-comfort mapping table is pre-established, mapping different categories of emotional information to "comfortable" or "uncomfortable." For example: The emotional information mentioned here specifically refers to the user's emotional state that is strongly correlated with the experience of kitchen noise. Therefore, there is no need for complex multi-dimensional emotional classification. In a preferred embodiment, the emotional information is divided into only two categories: positive emotions and negative emotions. Negative emotions correspond to the user's negative experience of irritability, impatience, and depression caused by excessive noise from the range hood. Positive emotions correspond to the user's relaxed state of being satisfied with the current noise reduction effect and having no auditory discomfort. When the emotional information is negative, it indicates that the user's auditory comfort is uncomfortable. When the emotional information is positive, it indicates that the user's auditory comfort is comfortable.

[0036] When the emotional information identified in step S2 is "irritable" or "dissatisfied", the auditory comfort level is judged as "uncomfortable"; when the emotional information is "comfortable" or "pleasant", the auditory comfort level is judged as "comfortable".

[0037] Understandably, this step, based on the emotional information identified in step S2, determines the user's auditory comfort level in the current noisy environment. This is the core basis for deciding whether to trigger noise reduction adjustment and when to stop it in subsequent steps. Furthermore, since the camera module continuously and in real-time acquires user facial image information within a preset area, the user's emotional information obtained through facial expression recognition is also real-time. Therefore, the user's auditory comfort level in this step is also real-time.

[0038] Step S4: If the judgment result in step S3 is uncomfortable, then increase the noise reduction parameters of the active noise reduction module by the first preset step size until the judgment result in step S3 is comfortable.

[0039] In this step, when step S3 determines that the user's auditory comfort level is "uncomfortable," it indicates that the current noise reduction effect of the active noise cancellation module is insufficient and needs to be enhanced. Specifically, the noise reduction effect is enhanced by gradually increasing the noise reduction parameters of the active noise cancellation module with a first preset step size, and the user's auditory comfort level is reassessed after each adjustment until the user's auditory comfort level becomes "comfortable."

[0040] In this embodiment, the first preset step size refers to the adjustment increment each time the noise reduction parameter is adjusted, and it is a pre-set fixed value. The specific value of the first preset step size is determined in advance based on factors such as the model of the range hood and the performance of the active noise reduction module, and this embodiment does not impose any restrictions on it. In one embodiment, the first preset step size is the adjustment step size of the amplitude component of the noise reduction wave, and its value is 1% to 5% of the maximum allowable amplitude of the noise reduction wave. The purpose of setting the step size is to avoid sudden changes in the noise reduction effect caused by abrupt changes in the noise reduction parameter, and to achieve smooth noise reduction adjustment. The noise reduction parameter mentioned here refers to the adjustable parameter or parameter set used to control the working state of the active noise reduction module. In this embodiment, the noise reduction parameter may include, but is not limited to, the following parameters: (1) Amplitude component: A parameter used to control the amplitude of the noise reduction wave. The larger the amplitude, the stronger the noise reduction wave's ability to cancel out the noise wave.

[0041] (2) Phase component: A parameter that controls the phase relationship between the noise reduction wave and the noise sound wave, used to ensure that the phase of the noise reduction wave is opposite to or tends to be opposite to the phase of the noise sound wave. In one embodiment, the noise reduction parameter is the amplitude component of the noise reduction wave, and step S4 enhances the cancellation effect of the noise reduction wave on the noise sound wave by adjusting the amplitude component.

[0042] In this embodiment, increasing the noise reduction parameter refers to enhancing the ability of the noise-canceling wave output by the active noise cancellation module to cancel out noise waves, thereby reducing the residual noise sound pressure level at the user's location. Furthermore, "increasing" is an incremental operation relative to the current noise reduction parameter, with each adjustment increment being the first preset step size. Preferably, increasing the noise reduction parameter involves increasing the amplitude of the noise-canceling wave.

[0043] In this embodiment, after each adjustment of the noise reduction parameter by a first preset step size, the adjustment is not stopped immediately. Instead, steps S1 to S3 are continuously executed in a loop to re-obtain the user's emotional information based on the latest acquired facial image information and determine auditory comfort. If the determination result is still "uncomfortable," the noise reduction parameter is further adjusted by the first preset step size; if the determination result is "comfortable," the adjustment is stopped, and the current noise reduction parameter remains unchanged until the user's auditory comfort is identified as "uncomfortable" the next time. It should be noted that when adjusting the noise reduction parameter, the adjustment stops after reaching the maximum allowable amplitude and runs with the current noise reduction parameter. This maximum allowable amplitude can be the maximum allowable value of ANC.

[0044] To better understand this invention, the overall working principle of this invention is described below in conjunction with specific application scenarios: When a user turns on the range hood to cook, the noise generated may cause discomfort. At this time, a camera module continuously captures images of the user, obtaining facial information through these images. When a user is irritated by the noise, their facial expression typically shows features such as frowning and downturned corners of the mouth. This invention extracts these features from the acquired facial image information to identify negative emotions such as "irritability" or "dissatisfaction." Based on the identified negative emotions, this invention determines the user's auditory comfort level as "uncomfortable." Therefore, the active noise cancellation module gradually increases its noise cancellation parameters (e.g., increasing the amplitude of the noise cancellation wave) by a first preset step size, gradually enhancing the cancellation effect of the noise wave on the noise sound wave, and gradually reducing the noise at the user's location. As the noise decreases, the user's irritability usually gradually subsides, and their facial expression gradually changes from "frowning" and "dissatisfaction" to positive features such as "natural" and "calm." When a positive emotion is detected in the user, the auditory comfort level is determined to be "comfortable". The noise reduction parameters are then stopped from being increased further and are maintained at the current level, thus achieving a complete closed-loop control process.

[0045] The beneficial effects of this embodiment will be described in detail below.

[0046] (1) This invention solves the technical deficiency of existing active noise cancellation systems that rely solely on objective acoustic physical parameters and cannot perceive the user's subjective auditory experience by constructing a closed-loop feedback mechanism and using the user's real-time emotional feedback as the basis and endpoint for noise reduction control. Specifically, in existing noise reduction schemes that use physical parameters such as sound pressure level as control targets, although the decibel value after noise reduction meets the standard, the user may still be dissatisfied with the noise reduction effect, but the system cannot know and make adjustments. In this invention, step S1 collects the user's facial image information in real time through the camera module, step S2 obtains the user's emotional information, step S3 judges the user's auditory comfort based on the emotional information, and step S4 only starts adjusting the noise reduction parameters when the judgment result is "uncomfortable". After each adjustment, it returns to step S1 to re-collect images, re-identify expressions, and re-judge comfort until the user's auditory comfort feedback changes from "uncomfortable" to "comfortable". Thus, the active noise reduction control method of this invention no longer aims to reduce the decibel value as the endpoint, but rather aims to restore the user's emotional comfort as the ultimate goal, realizing a dynamic match between the noise reduction effect and the user's real experience.

[0047] (2) This invention achieves automatic adaptation to the personalized noise reduction needs of different users by gradually increasing the noise reduction parameters with a preset step size and monitoring user emotional feedback in real time. This solves the technical problem that different users have individual differences in noise sensitivity and tolerance, and that uniform physical noise reduction parameters cannot adapt to diverse needs. Specifically, different users have significant individual differences in their perception of the same noise environment. For example, noise of the same decibel value may be unbearable for one user, but has no effect on another. In step S4 of this invention, the noise reduction parameters are gradually increased with a first preset step size, and the auditory comfort is reassessed after each adjustment until the user's auditory comfort is comfortable. This adjustment method ensures that the noise reduction parameters of each user eventually converge to their individual auditory comfort level. Users with high sensitivity will continue to adjust until the noise reduction effect is strong enough, while users with low sensitivity can stop adjusting at an earlier stage. Thus, this invention achieves automatic personalized customization of noise reduction intensity, overcoming the limitation of uniform parameters in the prior art that cannot take into account individual differences.

[0048] (3) The closed-loop control logic of this invention overcomes the technical defects of existing active noise reduction schemes, which mostly use open-loop control and cannot cope with changes in operating conditions. Specifically, existing open-loop schemes output noise reduction waves according to preset parameters, and the noise reduction effect decreases once the noise characteristics or user position changes. However, the loop structure consisting of steps S1 to S4 in this invention enables the system to continuously track changes in user emotions and adjust the noise reduction strategy in a timely manner, overcoming the inherent defect of open-loop control being unable to cope with changes in operating conditions.

[0049] (4) This invention can be achieved simply by adding a camera module to the existing range hood and optimizing the control algorithm. There is no need to modify the structure of the air duct, fan, etc. The modification cost is low and it is suitable for large-scale industrial promotion.

[0050] Example 2: refer to Figure 2 This embodiment is a further description of the first embodiment described above.

[0051] The active noise reduction control method for range hoods based on facial expression recognition described in this embodiment includes the following steps: Increasing the noise reduction parameters of the active noise reduction module by a first preset step size includes the following steps: The number of times the judgment result in step S3 is "uncomfortable" within a preset time period is counted. If the number of times exceeds the first preset threshold, the noise reduction parameters of the active noise reduction module are increased by the second preset step size; otherwise, the noise reduction parameters of the active noise reduction module are increased by the first preset step size, wherein the second preset step size is greater than the first preset step size.

[0052] The difference between this embodiment and Embodiment 1 lies in the specific implementation of step S4, "adjusting the noise reduction parameters of the active noise reduction module by a first preset step size." In this embodiment, the adjustment is not fixed at the first preset step size each time, but rather different adjustment step sizes are dynamically selected based on the number of times the user experiences discomfort within a preset time period.

[0053] Specifically, the system first counts the number of times the judgment result of "uncomfortable" occurs in step S3 within a preset time period. In this embodiment, after the active noise cancellation module starts adjusting, the system records an discomfort event each time step S3 is executed and a "discomfortable" judgment result is obtained. The preset time is a pre-set time window used to count the total number of times the "discomfortable" judgment result occurs within this time window. The preset time can be preset according to the actual application scenario. In one embodiment, the preset time ranges from 5 seconds to 30 seconds.

[0054] Subsequently, the statistically obtained number of discomfort episodes is compared with a first preset threshold. It is determined whether the number of episodes exceeds the first preset threshold, which is a pre-set value used to judge whether the user's discomfort level has reached a point requiring faster adjustment. Based on the comparison results, a step size for adjusting the noise reduction parameters is selected. When the number of discomfort episodes exceeds the first preset threshold, it indicates that the user is experiencing frequent negative emotions in the current noisy environment, and the need for noise reduction is more urgent. In this case, using a larger second preset step size allows the noise reduction parameters to adapt to the user's auditory comfort more quickly, shortening the total duration of the user's discomfort and improving the response speed and user experience of noise reduction adjustment. Conversely, when the number of discomfort episodes does not exceed the first preset threshold, it indicates that the user's negative emotions are not frequent. In this case, continuing to use a smaller first preset step size avoids overshooting of the noise reduction effect (i.e., the noise reduction amplitude exceeds the noise amplitude, generating new noise) due to an excessively large step size, ensuring the smoothness of the adjustment process.

[0055] In one embodiment, the first preset step size is 2% of the maximum allowable amplitude of the noise reduction wave, and the second preset step size is 5% of the maximum allowable amplitude of the noise reduction wave.

[0056] The adjustment process in this embodiment also adopts a closed-loop feedback mechanism. After each adjustment is completed with the first preset step size or the second preset step size, the process returns to step S1 to re-collect facial image information, re-identify facial expressions, re-judge auditory comfort, and execute the step size selection logic of this embodiment again (i.e., re-count the number of uncomfortable times within the preset time and select the step size based on whether it exceeds the first preset threshold) until the judgment result of step S3 is comfortable and the adjustment stops.

[0057] This embodiment dynamically selects the adjustment step size by statistically analyzing the frequency of user discomfort within a preset time period. When the user frequently experiences discomfort, a large step size is used for rapid adjustment; when the user experiences occasional discomfort, a small step size is used. This achieves an adaptive match between the adjustment speed and the urgency of the user's needs. While ensuring smooth adjustment, this solution effectively shortens the duration of user discomfort, further improving the efficiency of noise reduction adjustment and the user experience.

[0058] Example 3: refer to Figure 3 This embodiment is a further description of the first embodiment described above.

[0059] The active noise reduction control method for range hoods based on facial expression recognition described in this embodiment includes the following steps: Increasing the noise reduction parameters of the active noise reduction module by a first preset step size includes the following steps: If the judgment result in step S3 is discomfort, then the duration of the discomfort result is calculated. If the duration exceeds the second preset threshold, then the noise reduction parameter of the active noise reduction module is increased by the third preset step size. Otherwise, the noise reduction parameter of the active noise reduction module is increased by the first preset step size. The third preset step size is greater than the first preset step size.

[0060] The difference between this embodiment and Embodiment 1 lies in the specific implementation method of increasing the noise reduction parameters of the active noise reduction module with a first preset step size in step S4. In this embodiment, the adjustment is not fixed at the first preset step size each time, but rather different adjustment step sizes are dynamically selected according to the duration of the user's discomfort.

[0061] Specifically, when the judgment result in step S3 is discomfort, the duration of the discomfort result is calculated.

[0062] In this embodiment, after the active noise cancellation module starts adjusting, it begins timing the duration of the uncomfortable state whenever step S3 obtains a "discomfort" judgment result. The duration refers to the length of time the user is continuously in the "discomfort" state.

[0063] In one embodiment, the duration is calculated as follows: the timer starts from the moment when step S3 first determines "uncomfortable" and continues until the result of step S3 changes to "comfortable".

[0064] Subsequently, the calculated duration of discomfort is compared with a second preset threshold to determine whether the duration exceeds the second preset threshold. The second preset threshold is a pre-set time value used to determine whether the user's level of discomfort has reached the level of "needing to speed up the adjustment".

[0065] The second preset threshold can be preset according to the actual application scenario. In one embodiment, the value of the second preset threshold is in the range of 2 seconds to 10 seconds, preferably 7 seconds. That is, when the user is in an "uncomfortable" state for more than 7 seconds, it indicates that the user's negative emotions towards the current noisy environment persist and have not been relieved, and the noise reduction effect needs to be enhanced at a faster speed.

[0066] When the duration of discomfort exceeds the second preset threshold, it indicates that the user is in a prolonged negative emotional state and has a strong need for noise cancellation. In this case, adjusting with a larger third preset step size allows the noise cancellation parameters to adapt to the user's auditory comfort more quickly, shortening the total duration of discomfort and improving the response speed of noise cancellation adjustment.

[0067] Conversely, when the duration of discomfort does not exceed the second preset threshold, it indicates that the user's discomfort is still in a brief and occasional state. At this time, continue to use a smaller first preset step size for adjustment, which can gradually improve the noise reduction effect and avoid the noise reduction effect overshoot caused by too large a step size (i.e., the noise reduction amplitude exceeds the noise sound wave amplitude and generates new noise), ensuring the smoothness of the adjustment process.

[0068] In one embodiment, the first preset step size is 2% of the maximum allowable amplitude of the noise-reduced wave, and the third preset step size is 6% of the maximum allowable amplitude of the noise-reduced wave. Compared to the second preset step size in Embodiment 2, the third preset step size in this embodiment can be set to the same or a larger value, because the cumulative duration better reflects the user's true and stable discomfort state.

[0069] The adjustment process in this embodiment also adopts a closed-loop feedback mechanism. After each adjustment is completed with the first preset step size or the third preset step size, the process returns to step S1, re-collects facial image information, re-identifies facial expressions, re-judges auditory comfort, and executes the step size selection logic of this embodiment again (i.e., recalculates the duration of discomfort and selects the step size based on whether it exceeds the second preset threshold), until the judgment result of step S3 is "comfortable" and the adjustment stops.

[0070] This embodiment dynamically selects the adjustment step size by monitoring the duration of the user's discomfort. When the user continuously exhibits discomfort, a large step size is used for rapid adjustment; when the user only experiences brief discomfort, a small step size is used for adjustment. In actual range hood applications, occasional facial movements by the user (such as rubbing their eyes or yawning) may lead to a false "discomfort" judgment in a single expression recognition. However, this false judgment usually does not last for a long time. Judging by the duration can filter out such occasional noise.

[0071] It is understood that Embodiment 2 selects the step size based on the "number of times of discomfort within a preset time," which is suitable for detecting the frequency of user discomfort; Embodiment 3 selects the step size based on the "duration of discomfort," which is suitable for detecting the duration of user discomfort. Both can be used independently or in combination. In one combined embodiment, a larger step size is used for adjustment when the number of times of discomfort exceeds a first preset threshold or the duration of discomfort exceeds a second preset threshold; in another combined embodiment, a larger step size is used only when both conditions are met simultaneously. Those skilled in the art can choose a suitable combination method according to actual application needs, and this invention does not limit this.

[0072] Example 4: refer to Figure 4This embodiment is a further description of Embodiment 1 above. Its purpose is to provide a personalized noise reduction parameter calling and storage mechanism based on user identity recognition, enabling the range hood to provide customized noise reduction methods for different users, and continuously learn and update the user's personalized preferences during use.

[0073] The active noise reduction control method for range hoods based on facial expression recognition described in this embodiment further includes the following steps between step S1 and step S2: Facial recognition is performed based on the facial image information to obtain the user's identity; If not, proceed directly to the next steps and collect the operating data of the active noise reduction module in real time. Determine the personalized noise reduction parameters of the current user based on the operating data, bind the personalized noise reduction parameters of the current user with the user identity and store them. If so, the personalized noise reduction parameters bound to the user identity are obtained, and the active noise reduction module is controlled to operate under the personalized noise reduction parameters bound to the user identity.

[0074] In this embodiment, a face recognition algorithm is used to perform face recognition on the face image information obtained in step S1 in order to obtain the user's identity. The aforementioned face recognition algorithm can use existing technologies, which will not be described in detail here.

[0075] After obtaining the user's identity, it is determined whether the user belongs to a pre-stored user. In one embodiment, the obtained user identity is compared with the user database already stored in the range hood system to determine whether the user is a pre-stored user. The pre-stored users mentioned here can be users stored in the local storage of the range hood or in a cloud server connected to it. When a corresponding record can be found in the user database, the user is determined to be a pre-stored user; otherwise, the user is determined to be a non-pre-stored user (i.e., a new user).

[0076] It should be noted that the steps of obtaining user identity and determining whether the user identity belongs to a pre-stored user are part of the initialization phase. This initialization phase is only executed when the range hood is started and no user identity is logged in. That is, when the range hood is in an unlogged-in state, user identity is obtained and logged in through a single face recognition. After the initialization phase is completed, the range hood is already in the logged-in state of that user identity. In the subsequent iterative adjustment process, the system directly runs with the logged-in user identity, without needing to perform face recognition to obtain user identity again. In other words, in the iterative cycle of steps S1 to S4, each time it returns, it only needs to re-obtain facial image information and perform expression recognition, without repeating the face recognition login process, thereby reducing unnecessary computational overhead and improving the system's response speed.

[0077] When the system determines that the current user is not a pre-stored user (i.e., a new user), it means that the system has not yet stored the user's personalized noise reduction parameters. At this point, the range hood system enters learning mode. First, the facial expression recognition closed-loop adjustment process from steps S2 to S4 is executed directly. The system can start with the factory default noise reduction parameters and gradually increase the noise reduction parameters according to the user's auditory comfort until the user's auditory comfort changes from "uncomfortable" to "comfortable", thus completing a complete noise reduction adjustment process for the new user.

[0078] Secondly, during and after the adjustment process, the system collects real-time operational data from the active noise cancellation module. This operational data includes, but is not limited to: noise cancellation parameters matched to the current user's auditory comfort level, the step size recorded during the adjustment process, the total time from the start of adjustment to reaching a comfortable state, the user's identity information, and the current range hood's setting. Then, based on the collected operational data, the system determines the user's personalized noise cancellation parameters.

[0079] Finally, the user's personalized noise reduction parameters are bound to the user's identity and stored in the range hood's local storage or a cloud server connected to it. After this storage, the user becomes a "pre-stored user" the next time they use the device, and the system can directly call up the stored personalized noise reduction parameters without having to readjust them from the default settings.

[0080] Understandably, even if a user is not a pre-stored user, the system will gradually store personalized noise reduction parameters bound to the current user as it runs. Therefore, when the system detects that the current user's hearing comfort is uncomfortable again, it can directly call the personalized noise reduction parameters to run the system. Even when the system is running with the stored personalized noise reduction parameters, it will continue to monitor the user's emotions, judge the user's hearing comfort, and update the personalized noise reduction parameters according to the user's hearing comfort.

[0081] When the system determines that the user is a pre-registered user, it means that the current user has previously used this range hood, and the system has stored personalized noise reduction parameters for them. At this time, the system retrieves the personalized noise reduction parameters bound to the user's identity from local storage or the cloud server, and controls the active noise reduction module to run directly under these personalized noise reduction parameters.

[0082] In this embodiment, the personalized noise reduction parameters can be the noise reduction parameters that the user has finally converged and stored during the historical use of the range hood through the active noise reduction control method for range hoods based on facial expression recognition described in any embodiment of the present invention. These personalized noise reduction parameters represent the user's individual auditory comfort level, that is, the noise reduction parameter configuration that enables the user's emotional feedback to reach a "comfortable" state under the user's own noise sensitivity and tolerance conditions.

[0083] Understandably, by directly obtaining pre-stored personalized noise reduction parameters, the range hood can quickly enter a noise reduction state suitable for the current user after being turned on, without having to go through a gradual adjustment process, thus shortening the user's waiting time.

[0084] In one embodiment, after calling the pre-stored personalized noise reduction parameters, the range hood system does not operate entirely under those parameters. Instead, it uses these personalized noise reduction parameters as an initial basis to continue executing the facial expression recognition closed-loop adjustment process from steps S1 to S4. That is, starting with the pre-stored personalized noise reduction parameters, the system adjusts based on the user's real-time facial expression to adapt to the user's changing noise reduction needs in different cooking scenarios (e.g., the noise is higher when stir-frying).

[0085] Understandably, regardless of whether the user is a pre-registered user, the system will not stop monitoring the user's emotions due to the existence of personalized noise reduction parameters. While running with personalized noise reduction parameters, the system continuously acquires the user's facial image information, recognizes expressions, and judges auditory comfort through steps S1 to S3, and dynamically updates and optimizes the stored personalized noise reduction parameters based on the user's real-time auditory comfort.

[0086] Example 5: This embodiment is a further description of the aforementioned embodiment four.

[0087] The active noise reduction control method for range hoods described in this embodiment further includes the following steps before the step of performing facial recognition based on the facial image information to obtain the user's identity: In response to the power-on information, the current gear of the range hood is obtained and the noise reduction parameter corresponding to the current gear is obtained from the pre-stored gear-noise reduction parameter mapping table; The active noise cancellation module is controlled to operate under the noise cancellation parameters corresponding to the current gear.

[0088] In this embodiment, when the range hood receives a power-on message (such as when the user presses the power button, turns on the range hood via voice command, or triggers the power-on via smart linkage), the system responds immediately and first obtains the current operating level of the range hood.

[0089] In this embodiment, the range hood typically has multiple operating speeds, including but not limited to low speed, medium speed, high speed, and stir-fry speed. Different speeds correspond to different fan speeds and air volumes, resulting in differences in noise spectrum characteristics and sound pressure levels.

[0090] After obtaining the current gear, the noise reduction parameters corresponding to that gear are queried from the pre-stored gear-noise reduction parameter mapping table.

[0091] After obtaining the noise reduction parameters corresponding to the current gear, the active noise reduction module is controlled to output noise reduction waves according to the noise reduction parameters to initially cancel the noise generated by the range hood.

[0092] It is understandable that the noise reduction parameters corresponding to the gear level in this step are average noise reduction parameters that can achieve a relatively good level of auditory comfort for most users.

[0093] This embodiment enables the active noise cancellation module to quickly enter the appropriate noise cancellation state based on the current level before the user generates emotional feedback, thereby shortening the noise cancellation response time at the initial startup and improving the user's initial user experience.

[0094] After the above steps are completed, the system proceeds to step S1, which uses the camera module to acquire the user's facial image information within a preset area, and then sequentially executes steps S2 (expression recognition), S3 (judging auditory comfort), and S4 (noise reduction parameter adjustment) to achieve noise reduction adjustment based on the user's auditory comfort.

[0095] Example 6: refer to Figure 5 This embodiment is a further description of the aforementioned Embodiment 5.

[0096] In practical applications, range hoods typically have multiple operating speeds (such as low, medium, high, and high-speed settings). Different speeds correspond to different fan speeds and airflow rates, resulting in significant differences in noise levels. For example, high-speed or high-speed settings produce higher noise levels due to the high fan speed and large airflow; conversely, low-speed settings produce lower noise levels due to the low fan speed and small airflow. When a user changes the range hood's operating speed or the range hood adaptively adjusts its settings, the noise source characteristics fundamentally change. Continuing to use the original noise reduction parameters for incremental adjustments may not allow for a quick adaptation to the new noise environment and could even lead to incorrect adjustments.

[0097] To address the aforementioned problems, this embodiment describes an active noise reduction control method for range hoods based on facial expression recognition, comprising: If the judgment result in step S3 is uncomfortable, then the noise reduction parameters of the active noise reduction module are increased by a first preset step size until the judgment result in step S3 is comfortable, which includes the following steps: If the judgment result in step S3 is "uncomfortable", then determine whether there is a gear shift.

[0098] If there is a gear change, it is determined whether there is a personalized noise reduction parameter associated with the changed gear. If there is, the active noise reduction module is controlled to run under the personalized noise reduction parameter. If there is no personalized noise reduction parameter, the active noise reduction module is controlled to run under the noise reduction parameter corresponding to the changed gear. The noise reduction parameter corresponding to the changed gear is obtained from the pre-stored gear-noise reduction parameter mapping table. Proceed to step S1. If there is no gear shift, the noise reduction parameters of the active noise reduction module are increased by a first preset step size until the judgment result in step S3 is comfortable.

[0099] In this embodiment, when the system detects that the user's auditory comfort level is "uncomfortable," it first determines whether the range hood has changed its speed setting, rather than immediately initiating a step-by-step adjustment of the noise reduction parameters. The speed setting change refers to the range hood switching from one operating speed setting to another. This could be an increase in speed, such as switching from a low speed setting to a high speed setting, or a decrease in speed, such as switching from a medium speed setting to a low speed setting.

[0100] When a gear shift is detected, the system further queries whether there are personalized noise reduction parameters associated with the current user's identity and the shifted gear.

[0101] In this embodiment, for each pre-stored user, the system stores corresponding personalized noise reduction parameters based on the user's noise reduction preferences at different levels.

[0102] In one embodiment, the storage format of the personalized noise reduction parameters is: (user ID, gear position identifier, personalized noise reduction parameters (including amplitude components, phase components, etc.)). When the system needs to query the personalized noise reduction parameters of a user in the current gear position, it can obtain the corresponding personalized noise reduction parameters by jointly retrieving the user ID and gear position identifier.

[0103] As mentioned earlier, even if a user is determined not to be a pre-stored user, the system gradually stores personalized noise reduction parameters associated with the current user as it operates. Therefore, even if the system determines the user is not a pre-stored user during facial recognition, it can still further query whether personalized noise reduction parameters are associated with the current user's identity and the changed gear. Conversely, if the user is determined to be a pre-stored user, the system can directly query whether personalized noise reduction parameters are associated with the current user's identity and the changed gear.

[0104] When a personalized noise cancellation parameter associated with the current user's identity and the changed noise level is found, the system directly retrieves that parameter and controls the active noise cancellation module to operate under that parameter. This allows the system to quickly switch to the most suitable noise cancellation state for the user at that specific noise level after changing gears, eliminating the need for repeated step-by-step adjustments, shortening the user's waiting time after gear changes, and improving the continuity of the user experience.

[0105] It should be noted that after running with the personalized noise reduction parameters, the system does not stop judging the user's auditory comfort. This is to confirm whether the current noise reduction parameters can indeed adapt to the user's actual feeling at the current level. If the auditory comfort is judged to be "comfortable", the current parameters are maintained; if it is still "uncomfortable", the subsequent logic of this embodiment continues to be executed, including adjusting with preset steps.

[0106] If no personalized noise reduction parameters are found that correspond to the current user's identity and the changed gear, it indicates that the user has not used that gear before, or that the closed-loop convergence of facial expression recognition has not been completed when using that gear. In this case, the system retrieves the noise reduction parameters corresponding to the changed gear from the pre-stored gear-noise reduction parameter mapping table and controls the active noise reduction module to operate under those noise reduction parameters.

[0107] After the active noise cancellation module operates under the noise cancellation parameters corresponding to the changed gear, the system also reacquires the user's facial image information and executes steps S2 and S3 in sequence to monitor the user's auditory comfort in the new gear in real time, and decides whether further step adjustment is needed based on the comfort judgment result.

[0108] In one embodiment, if the system finally converges to a comfortable state after a complete closed-loop adjustment process (steps S1 to S4) at that gear, the noise reduction parameters that finally stabilize at that gear are used as the user's personalized noise reduction parameters at that gear, and stored in association with the user's identity and gear identifier, so that they can be directly called in subsequent use.

[0109] Understandably, when the range hood's speed decreases, the noise level also decreases accordingly. If the noise reduction intensity remains the same as before the speed decrease, overcompensation might occur because the amplitude of the reduced noise wave exceeds the amplitude of the current noise wave. This means the reduced noise wave itself becomes a new noise source, degrading the user's auditory experience. In this case, the user's comfort level is determined to be uncomfortable. Since the noise reduction parameters of the active noise cancellation module cannot be increased in this situation, when a speed decrease is detected, the noise reduction intensity is switched to a level matching the current speed. Alternatively, if a personalized noise reduction parameter exists that has been converged through historical closed-loop adjustments for the current user at the current speed, then that parameter is used; otherwise, the noise reduction parameter corresponding to the current speed is used, thus avoiding secondary noise pollution caused by a mismatch in noise reduction intensity.

[0110] In this embodiment, when it is determined that there is no gear change, it indicates that the range hood is currently operating at a stable gear level and the noise source characteristics have not fundamentally changed. At this time, the noise reduction parameters of the active noise reduction module are gradually increased by a first preset step size, and after each adjustment, the user's auditory comfort is reassessed based on the re-acquired facial image information and by executing steps S2 and S3 until the user is comfortable.

[0111] This embodiment introduces gear shift detection. When the user's auditory comfort level is uncomfortable, it prioritizes determining whether a gear shift has occurred and quickly matches the corresponding noise reduction parameters based on the shifted gear (prioritizing the user's personalized noise reduction parameters for that gear; otherwise, it uses the noise reduction parameters corresponding to the gear). This avoids response lag or incorrect adjustment direction caused by blindly using the original parameters for step-by-step adjustments after a gear shift. Simultaneously, for downshift scenarios, by switching the noise reduction intensity to a level matching the current gear, it effectively prevents overcompensation caused by the noise reduction wave amplitude exceeding the noise wave amplitude, thus preventing the noise reduction wave itself from becoming a new noise source.

[0112] Example 7: This embodiment is a further description of Embodiment 5 described above.

[0113] The active noise reduction control method for range hoods based on facial expression recognition described in this embodiment further includes the following steps before responding to power-on information, obtaining the current speed setting of the range hood, and retrieving the noise reduction parameter corresponding to the current speed setting from a pre-stored speed setting-noise reduction parameter mapping table: Control the range hood to operate at each speed setting and collect noise signals during the operation of each speed setting; The noise reduction parameters corresponding to each gear level are determined based on the noise signals of each gear level. Establish and pre-store the correspondence between location and noise reduction parameters.

[0114] This embodiment provides a method for pre-establishing a speed-noise reduction parameter mapping table. First, the range hood is controlled to operate at various speed settings. In one embodiment, the range hood is controlled to start from the lowest speed and sequentially increase to the highest speed. At each speed setting, after the range hood has stabilized (e.g., after running at that speed setting for 30 seconds), noise signals during operation at that speed setting are collected.

[0115] The noise signal can be acquired using an acoustic sensor (such as a microphone). The acquired noise signal includes at least the noise sound pressure level and noise spectrum information for each speed setting. In one embodiment, the noise signal acquisition duration for each speed setting is 10 to 30 seconds to ensure that the acquired data is statistically representative.

[0116] Based on the collected noise signals for each gear level, the noise reduction parameters corresponding to each gear level are determined. The noise reduction parameters are the control parameters used by the active noise reduction module to generate the noise reduction wave, including but not limited to the amplitude component and phase component of the noise reduction wave.

[0117] In one embodiment, the noise reduction parameters corresponding to each gear level are determined using the following method: (1) Perform spectrum analysis on the collected noise signal, convert the time domain signal into the frequency domain signal through Fast Fourier Transform (FFT), obtain the spectrum distribution characteristics of the noise at this level, and identify the main noise frequency components and their corresponding amplitudes.

[0118] (2) Based on the basic principle of active noise reduction (i.e., to achieve destructive interference by generating sound waves with the same amplitude and opposite phase as the noise sound waves), determine the ideal noise reduction wave parameters corresponding to each main frequency component: the amplitude component is equal to the noise amplitude, and the phase component is opposite to the noise phase (i.e., the phase difference is 180°).

[0119] (3) Correct the ideal noise reduction parameters to obtain the optimal noise reduction parameters that can actually be executed at each level.

[0120] Next, the correspondence between the noise reduction level and the noise reduction parameter is established and pre-stored. In one embodiment, the noise reduction parameters corresponding to each determined level are used to establish a level-noise reduction parameter mapping table according to the correspondence between the level and the parameter, and the table is pre-stored in the non-volatile memory of the range hood.

[0121] In one embodiment, the gear-noise reduction parameter mapping table can be stored using the following data structure: After pre-storage, when the range hood is turned on during actual use, the system can quickly query and obtain the corresponding noise reduction parameters from the mapping table by reading the current setting, thus realizing the rapid initialization of the active noise reduction module.

[0122] It is understandable that the aforementioned process of establishing the correspondence between the gear level and the noise reduction parameters can be established before leaving the factory, can be manually triggered by the user during use, can be automatically executed when the range hood is first powered on, or can be executed periodically after the range hood has been running for a period of time to compensate for changes in acoustic characteristics caused by factors such as equipment aging, and to ensure the accuracy of the correspondence between each gear level and the basic noise reduction parameters.

[0123] This embodiment collects and analyzes noise signals at each speed setting, establishes and stores the correspondence between the speed setting and noise reduction parameters, and provides a data foundation for rapid noise reduction initialization after the range hood is turned on.

[0124] Example 8: This embodiment is a further description of the aforementioned Embodiment 1.

[0125] The active noise reduction control method for range hoods based on facial expression recognition described in this embodiment further includes the following steps before step S1: The camera module periodically detects whether there are users within a preset area; If so, proceed to step S1; If no user is detected within a preset number of consecutive detection cycles, the current noise reduction parameters of the active noise reduction module are cached, and the current noise reduction parameters of the active noise reduction module are reduced according to a preset amount. If a user is detected again, the current noise reduction parameters are retrieved and the active noise reduction module is controlled to restore to the current noise reduction parameters.

[0126] In actual use of a range hood, users are not always in the cooking area. They may temporarily leave the kitchen to handle other tasks. In situations where the user is not present, the range hood often continues to operate, and the ANC continues to output noise-reducing waves at the original noise reduction parameters. However, since the user is no longer in the preset area, this noise reduction wave has no practical meaning for the user, and instead causes unnecessary energy consumption and continuous wear and tear on the speakers.

[0127] In response, this embodiment provides a method to reduce the noise reduction intensity of ANC when the user temporarily leaves the cooking area, thereby reducing unnecessary noise waves.

[0128] In this embodiment, a camera module continuously monitors a preset area for the presence of a user at a preset detection period. The preset area is the same as the preset area described in step S1, i.e., the spatial range within which the camera module can effectively capture facial images. The detection period is a preset time interval, for example, detecting once every 5 seconds.

[0129] In one embodiment, the camera module acquires images of a preset area according to a detection cycle, and then uses a human detection algorithm to determine whether a human silhouette or face exists in the image. If a human body or face is detected, it is determined that a user is present; if not detected, it is determined that no user is present. In another embodiment, a human infrared sensor can be used for auxiliary detection to improve the accuracy of the detection.

[0130] When a user is detected within a preset area during a certain detection cycle, it indicates that the user is currently located within the cooking area. At this point, the closed-loop adjustment process of the active noise reduction control method based on facial expression recognition is initiated or continues. The process then proceeds directly to step S1, where the user's facial image information is acquired via the camera module, and steps S2 through S4 are executed sequentially.

[0131] If no user is detected within a preset number of consecutive detection cycles, it indicates that the user has temporarily left the cooking area. The number of "preset number of consecutive detection cycles" can be preset according to the actual application scenario, such as 3 consecutive detection cycles, 5 consecutive detection cycles, etc. The purpose of setting multiple consecutive detection cycles is to effectively filter out occasional detection omissions caused by the user briefly leaving the camera's field of view (such as when the user bends down to pick up an item, turns around and leaves, etc.).

[0132] Once it is determined that the user has left, the current noise reduction parameters of the active noise cancellation module are first cached in non-volatile memory or RAM so that the noise reduction state can be quickly restored to before the user left when they return. Then, the current noise reduction parameters of the active noise cancellation module are reduced according to a preset amount, that is, the noise reduction intensity is decreased.

[0133] In one embodiment, the preset amount is 50% of the maximum value of the noise reduction parameter, that is, the noise reduction intensity is reduced to half of the level before departure. In another embodiment, the preset amount is 100% of the maximum value of the noise reduction parameter, that is, the system directly turns off ANC and completely stops outputting noise-reduced waves.

[0134] After the system has reduced noise reduction parameters, the camera module continues to monitor the preset area for users at the original detection cycle. When a user is detected again in a certain detection cycle, it indicates that the user has returned to the cooking area.

[0135] At this point, the current noise reduction parameters (i.e., the noise reduction parameters before the user left) are retrieved from the cache, and the ANC is controlled to restore the operation to the current noise reduction parameters, so that the user can immediately enjoy the same noise reduction effect as before leaving after returning, without having to readjust from scratch.

[0136] After restoring the noise reduction parameters to the state before departure, the system proceeds to step S1, which uses the camera module to acquire the user's facial image information, and then executes steps S2 to S4 in sequence to restart the closed-loop noise reduction adjustment process based on facial expression recognition.

[0137] This embodiment achieves intelligent energy-saving control of the active noise cancellation module by periodically detecting the user's presence through a camera module. When the user leaves the cooking area, the system automatically caches the current noise cancellation parameters and reduces the noise cancellation intensity, effectively reducing ANC output power and energy consumption. When the user returns, the system quickly restores the noise cancellation state before leaving, so the user does not need to readjust to the change in noise cancellation effect. The entire detection, caching, noise reduction, and recovery process is completed automatically without manual user intervention.

[0138] Example 9: The present invention also provides a range hood, which includes a camera module and an active noise cancellation module. The active noise cancellation module includes a speaker, which is used to emit noise-canceling waves to cancel noise. The range hood applies the active noise cancellation control method for range hoods based on facial expression recognition described in any of the foregoing embodiments. The beneficial effects of the active noise cancellation control method for range hoods based on facial expression recognition described in any of the foregoing embodiments are not elaborated here.

[0139] Example 10: According to embodiments of the present invention, a computer-readable storage medium is provided, comprising various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk. The computer-readable storage medium stores computer-executable instructions. When these computer-executable instructions are executed by a computer processor, the processor executes the active noise reduction control method for a range hood based on facial expression recognition described in any of the foregoing embodiments.

[0140] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a general-purpose hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the related technology, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to enable a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments.

[0141] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; under the concept of the present invention, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the present invention as described above, which are not provided in detail for the sake of brevity; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for active noise reduction control of a range hood based on facial expression recognition, characterized in that, The range hood includes a camera module and an active noise reduction module, and the method includes the following steps: S1. Obtain facial image information of users within a preset area through the camera module; S2. Perform facial expression recognition based on the facial image information to obtain the user's emotional information; S3. Determine the user's auditory comfort level based on the emotional information; S4. If the judgment result in step S3 is uncomfortable, then increase the noise reduction parameters of the active noise reduction module by the first preset step size until the judgment result in step S3 is comfortable.

2. The active noise reduction control method for range hoods based on facial expression recognition as described in claim 1, characterized in that, The step of increasing the noise reduction parameters of the active noise reduction module by a first preset step size includes the following steps: The number of times the judgment result in step S3 is "uncomfortable" within a preset time period is counted. If the number of times exceeds the first preset threshold, the noise reduction parameters of the active noise reduction module are increased by the second preset step size; otherwise, the noise reduction parameters of the active noise reduction module are increased by the first preset step size, wherein the second preset step size is greater than the first preset step size.

3. The active noise reduction control method for range hoods based on facial expression recognition as described in claim 1, characterized in that, The step of increasing the noise reduction parameters of the active noise reduction module by a first preset step size includes the following steps: If the judgment result in step S3 is discomfort, then the duration of the discomfort result is calculated. If the duration exceeds the second preset threshold, then the noise reduction parameter of the active noise reduction module is increased by the third preset step size. Otherwise, the noise reduction parameter of the active noise reduction module is increased by the first preset step size. The third preset step size is greater than the first preset step size.

4. The active noise reduction control method for range hoods based on facial expression recognition as described in claim 1, characterized in that, Between step S1 and step S2, the following steps are also included: Facial recognition is performed based on the facial image information to obtain the user's identity; Determine whether the user's identity belongs to a pre-stored user; If not, proceed directly to the next steps and collect the operating data of the active noise cancellation module in real time. Bind and store the operating data with the user identity, wherein the operating data includes at least the personalized noise cancellation parameters of the current user. If so, the personalized noise reduction parameters bound to the user identity are obtained, and the active noise reduction module is controlled to operate under the personalized noise reduction parameters bound to the user identity.

5. The active noise reduction control method for range hoods based on facial expression recognition as described in claim 4, characterized in that, Before the step of performing facial recognition based on the facial image information to obtain the user's identity, the following steps are also included: In response to the power-on information, the current gear of the range hood is obtained and the noise reduction parameter corresponding to the current gear is obtained from the pre-stored gear-noise reduction parameter mapping table; The active noise cancellation module is controlled to operate under the noise cancellation parameters corresponding to the current gear.

6. The active noise reduction control method for range hoods based on facial expression recognition as described in claim 5, characterized in that, If the judgment result in step S3 is uncomfortable, then the noise reduction parameters of the active noise reduction module are increased by a first preset step size until the judgment result in step S3 is comfortable, which includes the following steps: If the judgment result in step S3 is discomfort, then determine whether there is a gear shift; If there is a gear change, it is determined whether there is a personalized noise reduction parameter associated with the changed gear. If yes, the active noise reduction module is controlled to run under the personalized noise reduction parameter. If no, the active noise reduction module is controlled to run under the noise reduction parameter corresponding to the changed gear. The noise reduction parameter corresponding to the changed gear is obtained from the pre-stored gear-noise reduction parameter mapping table. Proceed to step S1. If there is no gear shift, the noise reduction parameters of the active noise reduction module are increased by a first preset step size until the judgment result in step S3 is comfortable.

7. The active noise reduction control method for range hoods based on facial expression recognition as described in claim 5, characterized in that, Before responding to the power-on information, obtaining the current speed setting of the range hood and retrieving the noise reduction parameter corresponding to the current speed setting from the pre-stored speed setting-noise reduction parameter mapping table, the process further includes: Control the range hood to operate at each speed setting and collect noise signals during the operation of each speed setting; The noise reduction parameters corresponding to each gear level are determined based on the noise signals of each gear level. Establish and pre-store the correspondence between location and noise reduction parameters.

8. The active noise reduction control method for range hoods based on facial expression recognition as described in claim 1, characterized in that, The following steps are included before step S1: The camera module periodically detects whether there are users within a preset area; If so, proceed to step S1; If no user is detected within a preset number of consecutive detection cycles, the current noise reduction parameters of the active noise reduction module are cached, and the current noise reduction parameters of the active noise reduction module are reduced according to a preset amount. If a user is detected again, the current noise reduction parameters are retrieved and the active noise reduction module is controlled to restore to the current noise reduction parameters.

9. A range hood, characterized in that, The range hood includes a camera module and an active noise reduction module. The active noise reduction module includes a speaker, which is used to emit noise reduction waves to cancel out noise. The range hood applies the active noise reduction control method for range hoods based on facial expression recognition as described in any one of claims 1-8.

10. A storage medium containing computer-executable instructions, characterized in that, The computer-executable instructions, when executed by a computer processor, are used to perform the active noise reduction control method for a range hood based on facial expression recognition as described in any one of claims 1-8.

Images