Range hood and control method and apparatus therefor, and storage medium

Abstract

A range hood and a control method and apparatus therefor, and a storage medium. The range hood comprises: a vision detection sensor, which is arranged above a cooktop. The method comprises: acquiring image data generated by a vision detection sensor; on the basis of the image data, determining smoke data; and on the basis of the smoke data, controlling a range hood.

Description

Range hood and control method, device and storage medium thereof CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to Chinese patent applications No. 202311872664.2 and No. 202311871411.3 filed on December 29, 2023, the entire contents of which are incorporated herein by reference. Technical Field

[0002] The present disclosure relates to the field of oil fume removal, and in particular to a range hood and a control method, device, and storage medium thereof. Background Art

[0003] A range hood (also known as a range hood) is a kitchen appliance that purifies the kitchen environment. Typically installed above the kitchen stove, it quickly extracts waste from stove combustion and harmful cooking fumes, exhausting them outdoors. It also condenses and collects the fumes, reducing pollution, purifying the air, and providing safety features like anti-poisoning and explosion-proofing.

[0004] In practice, users often forget or fail to turn on their range hoods in time. To improve the user experience, intelligent smoke detection can be implemented to automatically extract the smoke. However, in actual operation, accurate smoke detection cannot be achieved due to interference from external factors such as the use of cooking utensils.

[0005] To enhance the user experience, related technologies can use infrared light technology to intelligently detect smoke and extract it. However, infrared light detectors are typically positioned within a specific field of view or angle range. Light outside this angle range may be blocked by the edges of the lens or fail to focus on the detector, affecting detection effectiveness. This can lead to delayed operation of the range hood and smoke extraction, causing smoke to escape. Summary of the Invention

[0006] In view of this, the embodiments of the present disclosure provide a range hood and a control method, device and storage medium thereof, aiming to ensure timely operation of the range hood and smoke extraction.

[0007] In a first aspect, an embodiment of the present disclosure provides a method for controlling a range hood, wherein the range hood includes: a visual detection sensor, wherein the visual detection sensor is arranged above a stove, and the method includes: obtaining image data generated by the visual detection sensor; determining smoke data based on the image data; and controlling the range hood based on the smoke data.

[0008] In some embodiments, the range hood further comprises a fan. Determining the smoke data based on the image data comprises determining the smoke data corresponding to the image data based on color space data of the image data. Controlling the range hood based on the smoke data comprises determining a target operating gear for the fan based on the smoke data and gear thresholds for at least two operating gears; and controlling the fan to operate at the target operating gear.

[0009] In some embodiments, determining the smoke data corresponding to the image data based on the color space data of the image data includes: performing color space conversion based on the first color space data of the image data to generate second color space data of the image data; and determining the smoke data corresponding to the image data based on the second color space data.

[0010] In some embodiments, the first color space data based on the image data is subjected to color space conversion to generate the second color space data of the image data, including: obtaining a first color space type corresponding to the first color space data; determining a second color space type to be converted based on the first color space type; and performing color space conversion on the first color space data based on a conversion rule corresponding to the second color space type to generate second color space data.

[0011] In some embodiments, the color space conversion is performed on the first color space data based on the conversion rules corresponding to the second color space type to generate second color space data, including: obtaining first color pixel feature data corresponding to the first color space data; determining second color pixel feature data of the second color space data based on the first color pixel feature data and the conversion rules; and generating the second color space data based on the second color pixel feature data.

[0012] In some embodiments, determining the smoke data corresponding to the image data based on the second color space data includes: determining third color space data based on the second color space data and the smoke color interval value, the third color space data being the second color space data within the smoke color interval value; and generating the smoke data based on the third color space data.

[0013] In some embodiments, the method further includes: determining the mean and variance value of the second color pixel feature data based on the second color space data; generating the upper limit and lower limit values ​​of the smoke color interval value based on the mean and the variance value; and determining the smoke color interval value based on the upper limit and lower limit values.

[0014] In some embodiments, generating the smoke data based on the third color space data includes: performing numerical normalization processing on the third color space data; performing bit operation on the third color space data after numerical normalization processing and the first color space data to generate the smoke data representing the size of the oil smoke area.

[0015] In some embodiments, the target operating gear of the fan is determined based on the smoke data and the gear thresholds of at least two operating gears, including: determining the proportion information of the current oil smoke area to the area of ​​the image based on the size information of the image corresponding to the smoke data and the image data; and determining the target operating gear of the fan based on the proportion information and the gear threshold.

[0016] In some embodiments, the method also includes: if it is determined that the current proportion information is greater than or equal to a first gear threshold, switching the fan to operate at the operating gear corresponding to the first gear threshold, the first gear threshold being the switching threshold of the operating gear above the fan speed of the current operating gear; if it is determined that the current proportion information is less than a second gear threshold, after determining that the difference between the second gear threshold and the current proportion information is greater than or equal to a set value, switching the fan to operate at the operating gear corresponding to the second gear threshold, the second gear threshold being the switching threshold of the operating gear below the fan speed of the current operating gear.

[0017] In some embodiments, acquiring image data generated by the visual detection sensor includes acquiring current frame image data generated by the visual detection sensor. Determining smoke data based on the image data includes processing the current frame image data to obtain smoke data, wherein the smoke data includes at least two types of first smoke data, second smoke data, and third smoke data; wherein the first smoke data is used to represent smoke concentration data based on color extraction, the second smoke data is used to represent smoke concentration data based on static filtering extraction, and the third smoke data is used to represent smoke concentration data based on dynamic filtering extraction. Controlling the range hood based on the smoke data includes generating smoke detection data based on the smoke data, wherein the smoke detection data represents the current size of the smoke area.

[0018] In some embodiments, the current frame image data is processed to obtain smoke data, including at least two of the following steps: generating the first smoke data based on the current frame image data and set threshold data, the set threshold data being used to characterize the smoke color interval distribution; generating the second smoke data based on the current frame image data and background image data, the background image data characterizing the background image in a smoke-free state; generating the third smoke data based on the current frame image data and a set number of historical frame image data before the current frame image data.

[0019] In some embodiments, the range hood further includes a fan, and the method further includes: controlling the operation of the fan based on the smoke detection data and a set smoke threshold.

[0020] In some embodiments, generating smoke detection data based on the smoke data includes: binarizing each type of data in the smoke data to generate binarized data corresponding to each type of data; and generating the smoke detection data based on the binarized data.

[0021] In some embodiments, generating smoke detection data based on the smoke data includes: determining whether the third smoke data is valid smoke data; if the third smoke data is valid smoke data, generating smoke detection data based on at least two of the first smoke data, the second smoke data, and the third smoke data; if the third smoke data is not valid smoke data, generating smoke detection data based on the first smoke data and the second smoke data.

[0022] In some embodiments, the method further includes: generating a standard deviation of the contour center of gravity of the third smoke data based on the third smoke data and an edge detection algorithm, wherein the standard deviation is used to characterize the contour discreteness of the third smoke data; if the standard deviation is greater than or equal to a contour discreteness threshold, determining that the third smoke data is valid smoke data.

[0023] In some embodiments, generating the first smoke data based on the current frame image data and the set threshold data includes: performing color space conversion based on the first color space data of the current frame image data to generate the second color space data of the current frame image data; generating the first smoke data based on the second color space data and the set threshold data.

[0024] In some embodiments, generating the first smoke data based on the second color space data and the set threshold data includes: generating initial first smoke data based on the second color space data and the set threshold data; and performing morphological processing on the initial first smoke data to generate the first smoke data.

[0025] In some embodiments, generating the second smoke data based on the current frame image data and the background image data includes: performing an image difference algorithm based on the current frame image data and the background image data to generate pixel data of the difference area; generating the second smoke data based on the pixel data of the difference area.

[0026] In some embodiments, the generating of the second smoke data based on the pixel data of the difference area includes: extracting the pixel data of the difference area that meets the set threshold based on the pixel data of the difference area and a set threshold; extracting the background image data that meets the grayscale threshold based on the background image data and the grayscale threshold; generating the second smoke data based on the pixel data of the difference area that meets the set threshold and the background image data that meets the grayscale threshold.

[0027] In some embodiments, the generating of the third smoke data based on the current frame image data and a set number of historical frame image data before the current frame image data includes: for the set number of consecutive historical frame image data, performing a differential operation based on the current frame image data and each of the historical frame image data to generate pixel data of multiple difference areas; generating the third smoke data based on the pixel data of the multiple difference areas.

[0028] In some embodiments, the generating of the third smoke data based on the pixel data of the multiple difference areas includes: performing an overlay operation based on the pixel data of the multiple difference areas; determining the moving target and the contour information of the moving target based on the pixel data of the multiple difference areas after the overlay operation; and generating the third smoke data based on the contour information and the contour information threshold.

[0029] In a second aspect, to implement the method of the present embodiment, the present embodiment further provides a third control device for the range hood. The third control device comprises a third acquisition module, a third processing module, and a third generation module. The third acquisition module is configured to acquire image data generated by the visual detection sensor. The third processing module is configured to determine smoke data based on the image data. The third generation module is configured to control the range hood based on the smoke data.

[0030] In a third aspect, an embodiment of the present disclosure provides a range hood, comprising: a visual detection sensor, wherein the visual detection sensor is arranged above a stove; the range hood further comprises: a processor and a memory for storing a computer program that can be run on the processor, wherein the processor, when running the computer program, executes the steps of the method described in the first aspect above.

[0031] In a fourth aspect, an embodiment of the present disclosure provides a computer storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described in the first aspect are implemented.

[0032] The technical solution provided by the embodiments of the present disclosure provides a range hood including: a visual detection sensor, the visual detection sensor being disposed above a stovetop; and a method comprising: obtaining current frame image data generated by the visual detection sensor; processing the current frame image data to obtain smoke data, the smoke data including at least two of first smoke data, second smoke data, and third smoke data; wherein the first smoke data is used to represent smoke concentration data based on color extraction, the second smoke data is used to represent smoke concentration data based on static filtering extraction, and the third smoke data is used to represent smoke concentration data based on dynamic filtering extraction; and generating smoke detection data based on the smoke data, the smoke detection data representing the current size of the smoke area. In this manner, the first smoke data of the embodiments of the present disclosure can better highlight the color and brightness characteristics of the smoke and reduce interference from other color information; the second smoke data extracted based on static filtering can better eliminate the influence of static objects, thereby avoiding environmental interference; and the third smoke data extracted based on dynamic filtering can perform motion analysis of the smoke and reduce interference from environmental changes; and generating smoke detection data based on at least two of the first, second, and third smoke data, the smoke detection data representing the current size of the smoke area, thereby improving the accuracy of oil smoke detection and avoiding environmental interference.

[0033] The technical solution provided by the embodiments of the present disclosure provides a control method for a range hood, wherein the range hood includes: a visual detection sensor and a fan, wherein the visual detection sensor is disposed above a stove. The method includes: acquiring image data generated by the visual detection sensor; determining smoke data corresponding to the image data based on color space data of the image data; determining a target operating gear of the fan based on the smoke data and gear thresholds of at least two operating gears; and controlling the fan to operate at the target operating gear. In this way, by acquiring images and generating smoke data based on the generated image data, smoke data can be detected immediately, avoiding the problem of detection delay of infrared detection technology and improving the efficiency of oil smoke detection; and determining the operating gear of the range hood fan based on the smoke data and gear threshold, ensuring that the range hood can operate intelligently and reducing the possibility of oil smoke escape. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] FIG1-1 is a flow chart of a range hood control method according to some embodiments of the present disclosure;

[0035] FIG1-2 is a schematic diagram of a current original image captured by a camera according to an application example in some embodiments of the present disclosure;

[0036] 1-3 are schematic diagrams of a workflow of Solution 1 according to an application example in some embodiments of the present disclosure;

[0037] 1-4 are schematic diagrams of smoke images after color extraction according to application examples in some embodiments of the present disclosure;

[0038] 1-5 are schematic diagrams of a workflow of Solution 2 according to an application example in some embodiments of the present disclosure;

[0039] FIG1-6 is a schematic diagram of an image of a data frame 2 of static frame difference filtering according to an application example in some embodiments of the present disclosure;

[0040] 1-7 are schematic diagrams of a workflow of Solution 3 according to an application example in some embodiments of the present disclosure;

[0041] FIG1-8 is a schematic diagram of a smoke image of a dynamic difference frame data frame 3 according to an application example in some embodiments of the present disclosure;

[0042] 1-9 are schematic diagrams of a process for generating a result data frame according to an application example in some embodiments of the present disclosure;

[0043] Figures 1-10 are image diagrams of result data frames according to application examples in some embodiments of the present disclosure;

[0044] 1-11 are schematic structural diagrams of a first control device of a range hood according to some embodiments of the present disclosure;

[0045] 1-12 are schematic structural diagrams of range hoods according to some embodiments of the present disclosure.

[0046] FIG2-1 is a flow chart of a control method for a range hood according to some embodiments of the present disclosure;

[0047] FIG2-2 is a schematic diagram of a workflow for oil smoke detection of a range hood according to an application example in some embodiments of the present disclosure;

[0048] 2-3 are schematic diagrams of oil smoke characteristic images according to application examples in some embodiments of the present disclosure;

[0049] 2-4 are schematic diagrams of cooking utensils not generating oil smoke according to application examples of some embodiments of the present disclosure;

[0050] 2-5 are schematic diagrams of cooking utensils generating oil smoke according to application examples of some embodiments of the present disclosure;

[0051] FIG2-6 is a schematic diagram of smoke before bitwise operation according to an application example in some embodiments of the present disclosure;

[0052] FIG2-7 is a schematic diagram of smoke after bitwise operation according to an application example in some embodiments of the present disclosure;

[0053] FIG2-8 is a schematic diagram of smoke before median filtering according to an application example in some embodiments of the present disclosure;

[0054] FIG2-9 is a schematic diagram of smoke after median filtering according to an application example in some embodiments of the present disclosure; and

[0055] Figure 2-10 is a schematic structural diagram of a second control device of a range hood according to some embodiments of the present disclosure.

[0056] The correspondence between reference numerals and component names is as follows:

[0057] 1-1000, first control device; 1-1010, first acquisition module; 1-1020, first processing module; 1-1030, first generation module; 1-1040, first control module; 1-1050, first determination module; 1-1060, first conversion module; 1-1070, first extraction module;

[0058] 12-1100, range hood; 12-1101, processor; 12-1102, memory; 12-1103, user interface; 12-1104, network interface; 12-1105, bus system;

[0059] 2-1000, second control device; 2-1010, second acquisition module; 2-1020, second determination module; 2-1030, second control module; 2-1040, second conversion module; 2-1050, second generation module; 2-1060, second processing module; 2-1070, second switching module. DETAILED DESCRIPTION

[0060] The present disclosure will be described in detail below with reference to the accompanying drawings and embodiments.

[0061] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art in the art of the present disclosure. The terms used herein in the specification of the present disclosure are only for the purpose of describing specific embodiments and are not intended to limit the present disclosure.

[0062] In a first aspect, according to some embodiments of the present disclosure, a control method for a range hood is provided, wherein the range hood includes: a visual detection sensor, wherein the visual detection sensor is arranged on a stove, and the method includes: obtaining image data generated by the visual detection sensor; determining smoke data based on the image data; and controlling the range hood based on the smoke data.

[0063] A visual detection sensor is a device used to collect and process visual information and is a crucial component of a machine vision system. Visual sensors include cameras or video cameras. These sensors are installed above the cooktop and primarily monitor smoke and other gases generated during cooking. The fan is a core component of the range hood, providing suction to draw smoke, steam, and odors generated during cooking away from the cooktop area and into the external environment.

[0064] In some embodiments, the range hood further comprises a fan; determining the smoke data based on the image data comprises: determining the smoke data corresponding to the image data based on color space data of the image data; and controlling the range hood based on the smoke data comprises: determining a target operating gear of the fan based on the smoke data and gear thresholds of at least two operating gears; and controlling the fan to operate at the target operating gear.

[0065] An embodiment of the present disclosure provides a control method for a range hood, referring to FIG. 2-1 , including steps 2-110 to 2-140.

[0066] In step 2-110: image data generated by the visual detection sensor is acquired.

[0067] Image data is digital visual information captured by visual detection sensors such as cameras. Image data includes pixels, resolution, and color space type.

[0068] A pixel is the basic unit of image display and the smallest element that makes up digital images and videos. Each pixel represents a fixed point on the screen or in an image, with a specific color and brightness. Each pixel has a brightness value for one or more color channels (such as red, green, and blue). Image resolution defines the image's dimensions, typically expressed in pixel width and height, for example, 640x480, 1920x1080, etc. A color space is a mathematical model or system used to describe and represent color.

[0069] In step 2-120, smoke data corresponding to the image data is determined based on the color space data of the image data. To identify smoke data corresponding to the image data, the image's color space data must first be analyzed. Color space data refers to how colors are represented in an image. Different color spaces help highlight different types of features in the image.

[0070] Different color spaces use different coordinate systems and parameters to define color. These parameters are generally related to the human eye's perception of color. Color spaces include at least one of the following: RGB (Red, Green, Blue) color space and HSV (Hue, Saturation, Value) color space.

[0071] Based on the color space data of the image data, the characteristic information representing the smoke in the image is identified and extracted, and it is determined which parts of the image contain the smoke data, thereby determining the smoke data corresponding to the image data.

[0072] In step 2-130: determining a target operating gear of the blower based on smoke data and gear thresholds of at least two operating gears.

[0073] The fan's operating gear determines its ability to extract and exhaust smoke. Different gears correspond to different fan speeds or suction levels. Gear thresholds are pre-set based on factors such as smoke levels and fan performance. Each gear threshold represents a specific smoke handling capacity. Smoke below this threshold can be handled by a lower fan speed gear, while smoke above this threshold requires a higher fan speed gear.

[0074] In some embodiments, the fan operating gear includes at least two operating gears. Each gear corresponds to a different fan speed or suction level. The gear threshold can be adjusted based on factors such as smoke level, environmental conditions, and equipment performance.

[0075] The area of ​​the smoke data may be compared with the gear thresholds of at least two operating gears. If the area of ​​the smoke data is lower than the lowest gear threshold, the fan may be maintained at the lowest operating gear or turned off. If the area of ​​the smoke data is higher than the highest gear threshold, the fan should be set to the highest operating gear to extract and exhaust smoke at its maximum capacity.

[0076] In step 2-140: the fan is controlled to operate at the target operating gear.

[0077] Based on the smoke data and the gear thresholds for at least two operating gears, a target operating gear for the fan can be determined. A target operating gear instruction is generated and sent to the fan. Upon receiving the target operating gear instruction, the fan begins adjusting its operating parameters. In some embodiments, adjusting the fan's operating parameters may include changing motor speed, adjusting valve opening, adjusting blade angle, and other operations to achieve the target exhaust capacity.

[0078] In this way, by acquiring images and generating smoke data based on the generated image data, the smoke data can be detected in the first time, avoiding the problem of detection delay of infrared detection technology and improving the detection efficiency of oil smoke; and based on the smoke data and the gear threshold, the operating gear of the range hood fan is determined to ensure the intelligent operation of the range hood and reduce the possibility of oil smoke escape.

[0079] In some embodiments, determining smoke data corresponding to the image data based on color space data of the image data includes:

[0080] Performing color space conversion based on the first color space data of the image data to generate second color space data of the image data;

[0081] Based on the second color space data, smoke data corresponding to the image data is determined.

[0082] RGB is based on the additive color principle, creating other colors by combining the three basic colors of red, green, and blue at varying brightness levels. It is widely used in electronic displays and televisions. HSV color is described by three parameters: hue, saturation, and value, which better aligns with human color perception.

[0083] Color space conversion refers to the process of converting color values ​​in one color space to equivalent color values ​​in another color space. This is because different color spaces are suitable for different applications and devices, and the human eye's perception of color may be represented differently in different color spaces.

[0084] Assuming that the first color space data is RGB data, in order to be more consistent with human perception experience of color and facilitate color adjustment and selection, the embodiment of the present disclosure converts the RGB data into HSV data to generate second color space data of the image data.

[0085] Based on the second color space data, it is possible to determine which parts of the image data correspond to smoke and generate smoke data corresponding to the image data. The smoke data here is used to characterize the smoke condition in the current environment, including information such as smoke concentration, distribution, and movement.

[0086] In this way, by converting the image from the first color space to the second color space, which is more suitable for smoke detection, the visual characteristics of the smoke can be more effectively highlighted, and based on the second color space data, the smoke data corresponding to the image data can be determined, thereby improving the distinguishability and accuracy of the smoke data.

[0087] In some embodiments, performing color space conversion based on first color space data of image data to generate second color space data of the image data includes:

[0088] Obtaining a first color space type corresponding to the first color space data;

[0089] Determining a second color space type to be converted based on the first color space type;

[0090] Based on the conversion rule corresponding to the second color space type, the first color space data is converted into a color space to generate second color space data.

[0091] Before performing the color space conversion on the first color space, the color space type used by the current image data is obtained. In some embodiments, the determination and identification of the color space type can be achieved by using a specific color space detection algorithm.

[0092] The second color space type to be converted can be determined based on the first color space data and a preset mapping relationship. The mapping relationship here represents the correspondence between the first color space type and the second color space type. Users can preset the mapping relationship of color space types based on actual needs and device characteristics.

[0093] Based on the second color space type, a conversion rule corresponding to the second color space type may be determined. The conversion rule includes a color space conversion formula or algorithm. These conversion rules are typically generated based on color theory and mathematical models and are used to map values ​​from one color space to another color space.

[0094] The RGB color space represents color using a linear combination of three color components. Any color is related to these three components, and these three components are highly correlated. Therefore, continuously changing color is not intuitive; adjusting the image's color requires modifying these three components. Images captured in natural environments are easily affected by natural lighting, occlusion, and shadows, meaning they are sensitive to brightness. However, the three components of the RGB color space are closely related to brightness; any change in brightness will cause all three components to change accordingly, without a more intuitive way to express this.

[0095] The HSV color space is closer to people's perception of color than the RGB color space. It intuitively expresses the hue, vividness, and brightness of a color, making it easier to compare colors. The HSV color space consists of three components: Hue (hue, representing color information, i.e., its position on the spectrum), Saturation (saturation, color purity, indicating the color's proximity to the spectrum; higher saturation indicates a darker color), and Value (value, which determines the brightness of a color in the color space; higher value indicates a brighter color).

[0096] Assuming that the first color space data is RGB data, in order to be more consistent with human perception experience of color and facilitate color adjustment and selection, the embodiment of the present disclosure converts the RGB data into HSV data to generate second color space data of the image data.

[0097] In this way, by successfully converting the data in the first color space into the second color space, the smoke has more obvious features in the second color space, so that the smoke area can be better extracted and identified.

[0098] In some embodiments, performing color space conversion on the first color space data based on a conversion rule corresponding to the second color space type to generate the second color space data includes:

[0099] Obtaining first color pixel characteristic data corresponding to the first color space data;

[0100] Determining second color pixel characteristic data of the second color space data based on the first color pixel characteristic data and the conversion rule;

[0101] Second color space data is generated based on the second color pixel characteristic data.

[0102] The first color pixel feature data refers to feature data obtained by analyzing and describing the color information of each pixel in the first color space data. This feature data can reflect the distribution of colors in the image. The first color pixel feature data may be multiple. For example, in the RGB color space, the color pixel feature data includes three feature values: red (R), green (G), and blue (B).

[0103] The first color pixel characteristic data is subjected to a second conversion based on the conversion rule to determine second color pixel characteristic data of the second color space data, and the second color space data is generated based on the second color pixel characteristic data. The second color pixel characteristic data corresponds to the first color pixel characteristic data.

[0104] Assume that the color pixel feature data of the RGB color space is converted into the color pixel feature data of HSV. The color pixel feature data of HSV includes three feature values, namely, H, S, and V values.

[0105] In this way, by determining the second color pixel feature data of the second color space data based on the first color pixel feature data and the conversion rules; generating the second color space data based on the second color pixel feature data, it is more conducive to highlighting the specific characteristics of the smoke and improving the ability to distinguish smoke from other objects.

[0106] In some embodiments, determining smoke data corresponding to the image data based on the second color space data includes:

[0107] Determining third color space data based on the second color space data and the smoke color interval, where the third color space data is the second color space data within the smoke color interval;

[0108] Smoke data is generated based on the third color space data.

[0109] The smoke color interval is commonly used to characterize the typical color characteristics of smoke in this color space. This interval can be a continuous color range or a set of discrete color points or areas.

[0110] Based on the second color space data and the smoke color range, the third color space data is the portion of the second color space data that is filtered to identify color values ​​within the smoke color range. This filtered data is more likely to represent smoke, and a mask is created. Based on whether each pixel falls within the smoke color range, a binary mask is created. Typically, pixels within the smoke color range are marked as 255, while pixels outside the smoke color range are marked as 0.

[0111] The created mask is used to perform element-level multiplication (or bitwise AND operation) with the second color space data, so that the pixels belonging to the smoke color interval can be retained, and the pixels that do not belong to the smoke color interval are set to the background color (such as black or transparent), thereby generating and generating smoke data based on the third color space data, so that it can be used for subsequent smoke detection and analysis operations, which can help reduce noise and interference and improve the accuracy and efficiency of smoke detection.

[0112] In some embodiments, the method further comprises:

[0113] determining a mean value and a variance value of the second color pixel feature data based on the second color pixel feature data of the second color space data;

[0114] Generate the upper and lower limits of the smoke color range based on the mean and variance values;

[0115] Determine the smoke color range based on the upper and lower limits.

[0116] A mean and variance of the second color pixel feature data may be determined based on the second color pixel feature data. For example, assuming the second color pixel feature data includes hue (H), saturation (S), value (V), or other related features, for each feature dimension, the mean and variance for that feature dimension are calculated.

[0117] The mean can be represented by M, and the variance can be represented by S. In the H dimension, the upper threshold TH1 = M + S; the lower threshold TH2 = MS. Based on the upper and lower thresholds, the smoke color range is determined. This smoke color range is [TH1, TH2].

[0118] In this way, based on the second color pixel feature data of the second color space data, the mean and variance values ​​are determined, and the smoke color interval is generated, which improves the accuracy of the smoke color interval and is conducive to improving the accuracy and efficiency of smoke detection.

[0119] In some embodiments, generating smoke data based on the third color space data includes:

[0120] performing numerical normalization processing on the third color space data;

[0121] A bitwise operation is performed on the third color space data after the numerical normalization process and the first color space data to generate smoke data representing the size of the oil smoke area.

[0122] In some embodiments, each pixel value in the third color space data is normalized to be converted to a range of 0 to 1. This helps eliminate the effects of different pixel value ranges, making subsequent processing more consistent and accurate.

[0123] In some embodiments, a bitwise operation is performed on the normalized third color space data and the first color space data. For example, a bitwise AND operation can be performed on the pixel values ​​at corresponding positions in the normalized third color space data and the first color space data. This can preserve pixels that appear as smoke in both color spaces.

[0124] In this way, the third color space data can be numerically normalized and bitwise processed with the first color space data to generate smoke data representing the size of the oil smoke area. This method combines information from different color spaces, accurately quantifying the smoke area and improving the efficiency and accuracy of smoke image detection.

[0125] In some embodiments, determining a target operating gear of the blower based on smoke data and gear thresholds of at least two operating gears includes:

[0126] Determine the proportion of the current oil smoke area to the area of ​​the image based on the smoke data and the size information of the image corresponding to the image data;

[0127] The target operating gear of the fan is determined based on the ratio information and the gear threshold.

[0128] The image data corresponds to image size information including the image width and height. Based on the smoke data and the image size information corresponding to the image data, information about the ratio of the current oil smoke area to the image area is determined. For example, the smoke data includes pixels with non-zero values, and the number of these pixels is determined and counted. The ratio of the current oil smoke area to the image area is the ratio of the number of pixels to the area of ​​the image.

[0129] Multiple gear thresholds may be set, such as low, medium, and high gear thresholds, corresponding to different thresholds of the proportion of the oil smoke area to the area of ​​the image, thereby determining the target operating gear.

[0130] In this way, the disclosed embodiment calculates the current proportion of the oil smoke area to the image area based on smoke data and image size information. It then determines the target operating gear of the fan based on this proportion and a preset gear threshold. This allows for dynamic adjustment of the fan's operating intensity based on actual oil smoke conditions, effectively removing oil smoke and improving environmental quality and equipment energy efficiency.

[0131] In some embodiments, the method further comprises:

[0132] If it is determined that the current ratio information is greater than or equal to the first gear threshold, the fan is switched to operate at the operating gear corresponding to the first gear threshold, where the first gear threshold is a switching threshold of the operating gear above the fan speed of the current operating gear;

[0133] If it is determined that the current proportional information is less than the second gear threshold, after determining that the difference between the second gear threshold and the current proportional information is greater than or equal to the set value, the fan is switched to operate at the operating gear corresponding to the second gear threshold. The second gear threshold is the switching threshold of the operating gear below the fan speed of the current operating gear.

[0134] During operation, if the current ratio information is determined to be greater than or equal to the first gear threshold, indicating that the smoke concentration has increased and a higher gear is required for smoke removal, the fan will be switched to the operating gear corresponding to the first gear threshold. The first gear threshold is the operating gear switching threshold above the fan speed of the current operating gear. In this way, when the smoke is heavy, the range hood will automatically switch to a higher gear, ensuring that the smoke can be removed quickly and effectively, improving work efficiency.

[0135] If it is determined that the current ratio information is less than the second gear threshold, it indicates that the smoke concentration has decreased. However, the gear adjustment is not performed immediately at this time. In some embodiments, in order to avoid frequent switching of operating gears in the case of repeated fluctuations in smoke, after determining that the difference between the second gear threshold and the current ratio information is greater than or equal to a set value, the fan is switched to the operating gear corresponding to the second gear threshold. The second gear threshold is the switching threshold for the operating gear below the fan speed of the current operating gear. In other words, the current ratio information needs to be less than the second gear threshold and the difference below the second gear threshold is greater than or equal to the set value before switching to the operating gear corresponding to the second gear threshold. In this way, frequent switching of operating gears in the case of smoke fluctuations can be effectively avoided.

[0136] For example, in some cases, a hysteresis area needs to be set. For example, the range of the proportion of the oil smoke area to the area of ​​the image is 10-40%, which corresponds to the range hood 1 gear, the range of the proportion is 40-70% corresponding to the range hood 2 gear, and the range of the proportion is 70-100% corresponding to the range hood 3 gear. As the smoke becomes larger, that is, the proportion of the oil smoke area to the area of ​​the image increases, so that when the proportion reaches the preset range, the gear is normally upgraded to the corresponding gear. When the smoke becomes smaller, the gear is lowered only when the difference between the current proportion of the oil smoke area to the area of ​​the image and the lower limit of the proportion range corresponding to the range hood gear is greater than 10%. For example, the gear is lowered from 3 gear to 2 gear only when the current proportion of the oil smoke area to the area of ​​the image is less than 60%. In this way, repeated fluctuations in the range hood gear can be avoided.

[0137] In this way, frequent fluctuations and unnecessary switching of the range hood gear can be avoided, the stable operation of the equipment can be ensured, and the impact on users can be reduced.

[0138] The embodiment of the present disclosure is described in detail below with reference to an application example.

[0139] Home range hoods are increasingly equipped with a variety of functions, but their core function remains smoke extraction. In practice, users often forget or fail to activate their range hoods promptly. Current intelligent detection systems typically utilize infrared light, which is limited in detection angle and typically requires the sensor to be placed at the rear (near the range hood's exhaust duct). Regardless of the aforementioned situation, this can lead to untimely smoke extraction, causing smoke to escape.

[0140] Based on this, this application example provides a range hood smoke detection method. This method uses an image-based intelligent smoke detection solution to measure smoke, addressing the latency issue with existing infrared smoke detection. This application example incorporates a camera (the aforementioned visual sensor) on the range hood (i.e., the aforementioned range hood) facing the stove. This camera captures the smoke generated by the stove. Because the color characteristics of the smoke are more distinct than the background, it can capture the characteristics of cooking smoke in real time.

[0141] The RGB color space (the first color space mentioned above) represents color using a linear combination of three color components. Any color is related to these three components, and these components are highly correlated. Therefore, continuously changing colors is not intuitive; adjusting the color of an image requires modifying these three components. Images captured in natural environments are easily affected by natural lighting, occlusion, and shadows, making them sensitive to brightness. However, the three components of the RGB color space are closely related to brightness. Any change in brightness will cause all three components to change accordingly, resulting in a lack of a more intuitive representation. The HSV color space (the second color space mentioned above) is closer to human perception of color than RGB. It intuitively represents the hue, vividness, and brightness of a color, facilitating color comparison. The HSV representation of color images consists of three components: Hue (hue, representing color information, i.e., its position on the spectrum); Saturation (saturation, color purity, indicating the color's proximity to the spectrum; higher saturation indicates a darker color); and Value (value, which determines the brightness of a color within the color space; higher value indicates a brighter color).

[0142] This application example provides a workflow diagram for oil smoke detection of a range hood, as shown in FIG2-2 , including steps 2-201 to 2-211.

[0143] In step 2-201: Initialization.

[0144] In step 2-202: determine whether the range hood is started. If the range hood is started, execute step 2-203; if the range hood is not started, execute step 2-201.

[0145] In step 2-203: collect pictures.

[0146] The image data generated by the visual detection sensor is obtained, as shown in Figure 2-3. Figure 2-3 is a schematic diagram of the oil smoke feature image collected by the camera (i.e., the aforementioned image data).

[0147] In step 2-204: read the image.

[0148] Based on the image, the feature image is read to obtain RGB color data (ie, the aforementioned first color space).

[0149] In step 2-205: the image RGB data is converted to HSV color.

[0150] In some embodiments, color space conversion is performed based on the first color space data of the image data to generate the second color space data of the image data, that is, the RGB data is converted into the HSV color space to generate the aforementioned second color space data.

[0151] In step 2-206: the data is marked according to a preset threshold.

[0152] In some embodiments, based on the second color pixel feature data of the second color space data, the mean and variance values ​​of the second color pixel feature data are determined; based on the mean and variance values, the upper limit and lower limit values ​​of the smoke color interval numerical value are generated; based on the upper limit and lower limit values, the smoke color interval numerical value is determined.

[0153] For example, the mean M and variance S of the H, S, and V components of the feature image are obtained. That is, the upper threshold TH1 for the smoke image is M + S, and the lower threshold TH2 is MS. These upper and lower thresholds are used to separate smoke and background in real-world situations. These upper and lower thresholds are determined through actual testing and debugging and are fixed in the software.

[0154] In step 2-207: data marking is completed, and mask data (ie, the aforementioned third color data) is formed.

[0155] In some embodiments, based on the second color space data and the smoke color interval value, third color space data is determined, and the third color space data is the second color space data within the smoke color interval value;

[0156] After the range hood is turned on, the camera begins operating, capturing images as shown in Figures 2-4 and 2-5. Figure 2-4 shows a cooking appliance without producing any smoke, while Figure 2-5 shows a cooking appliance producing any smoke. The MCU reads the image and converts the data into the HSV color space. It then marks valid data based on the preset upper and lower thresholds TH1 / TH2 (the valid data here refers to the data in the second color space described above), and forms mask data (data between the upper and lower thresholds is set to 255, and data below the lower threshold or above the upper threshold is set to 0).

[0157] In step 2-208: the mask data and the RGB data are bitwise operated.

[0158] In some embodiments, a bit operation is performed on the third color space data after the numerical normalization process and the first color space data to generate smoke data representing the size of the oil smoke area.

[0159] For example, performing a bitwise operation on the mask data and the original RGB data of the image yields an extraction result. However, the extraction result still contains a lot of noise. See Figures 2-6 and 2-7. Figure 2-6 shows the smoke before the bitwise operation, and Figure 2-7 shows the smoke after the bitwise operation.

[0160] In step 2-209: median filtering.

[0161] Apply median filtering to the extracted image. A relatively ideal result is obtained, as shown in Figures 2-8 and 2-9. Figure 2-8 shows the smoke image before median filtering, and Figure 2-9 shows the smoke image after median filtering.

[0162] In step 2-210: calculate the percentage of white dots in the image.

[0163] The size of the oil smoke can be represented by the proportion of the white dot in the image to the entire image. Set the oil smoke size corresponding to different gears of the range hood, and set the hysteresis area. For example, the proportion of the oil smoke area to the area of ​​the image ranges from 10 to 40%, which corresponds to the range hood 1 gear, the proportion range is 40 to 70%, which corresponds to the range hood 2 gear, and the proportion range is 70 to 100%, which corresponds to the range hood 3 gear. As the smoke becomes larger, that is, the proportion of the oil smoke area to the area of ​​the image increases, when the proportion reaches the preset range, it will normally shift to the corresponding gear. When the smoke becomes smaller, the gear will be lowered only when the difference between the current proportion of the oil smoke area to the area of ​​the image and the lower limit of the proportion range corresponding to the range hood gear is greater than 10%. For example, it will be lowered from gear 3 to gear 2 only when the current proportion of the oil smoke area to the area of ​​the image is less than 60%. In this way, repeated fluctuations in the range hood gear can be avoided.

[0164] In step 2-211: adjust the range hood position according to the size of the smoke.

[0165] This application example uses imaging to detect smoke as soon as it occurs, resolving the current issue of delayed infrared smoke detection. This prevents smoke from escaping and improves the user experience.

[0166] The disclosed embodiments provide a control method for a range hood. The range hood includes a visual detection sensor located above a cooktop. A visual detection sensor, as used herein, is a device for collecting and processing visual information and is a key component of a machine vision system. The visual sensor includes a camera or a video camera. The visual sensor is located above the cooktop and primarily monitors smoke and other gases generated during cooking.

[0167] In some embodiments, acquiring image data generated by the visual detection sensor includes acquiring current frame image data generated by the visual detection sensor. Determining smoke data based on the image data includes processing the current frame image data to obtain smoke data, wherein the smoke data includes at least two of first smoke data, second smoke data, and third smoke data; wherein the first smoke data is used to represent smoke concentration data based on color extraction, the second smoke data is used to represent smoke concentration data based on static filtering extraction, and the third smoke data is used to represent smoke concentration data based on dynamic filtering extraction. Controlling the range hood based on the smoke data includes generating smoke detection data based on the smoke data, wherein the smoke detection data represents the current size of the smoke area.

[0168] An embodiment of the present disclosure provides a control method for a range hood, referring to FIG. 1-1 , including steps 1 - 110 to 1 - 130 .

[0169] In step 1-110, current frame image data generated by the visual detection sensor is obtained.

[0170] The current frame image data refers to the image information obtained by the visual detection sensor at the current moment. The frame image data includes: pixel data, image size and timestamp.

[0171] In step 1-120, the current frame image data is processed to obtain smoke data, which includes at least two of the first smoke data, the second smoke data, and the third smoke data; wherein the first smoke data is used to represent the smoke concentration data based on color extraction, the second smoke data is used to represent the smoke concentration data based on static filtering extraction, and the third smoke data is used to represent the smoke concentration data based on dynamic filtering extraction.

[0172] In an embodiment of the present disclosure, processing the current frame image data can generate at least two of the following: first, second, and third smoke data, thereby obtaining smoke data. The at least two of the first, second, and third smoke data can be generated based on smoke characteristics. For example, the first smoke data is used to represent smoke concentration data extracted based on color, the second smoke data is used to represent smoke concentration data extracted based on static filtering, and the third smoke data is used to represent smoke concentration data extracted based on dynamic filtering.

[0173] In step 1-130, smoke detection data is generated based on the smoke data, where the smoke detection data represents the size of the current smoke area.

[0174] At least two of the first smoke data, the second smoke data, and the third smoke data may be selected to generate smoke detection data, where the smoke detection data represents the size of the current smoke area.

[0175] For example, the first smoke data and the second smoke data, or the first smoke data and the third smoke data, or the second smoke data and the third smoke data.

[0176] For example, at least two pieces of smoke data may be subjected to a logical operation (such as "AND", "OR", "NOT", etc.) to generate smoke detection data. This combines multiple pieces of smoke data to generate smoke detection data, thereby improving the accuracy and stability of smoke detection.

[0177] In this way, the first smoke data of the embodiment of the present disclosure can better highlight the color and brightness characteristics of the smoke and reduce the interference of other color information. The second smoke data based on static filtering extraction can better eliminate the influence of static objects, thereby avoiding environmental interference; the third smoke data based on dynamic filtering extraction can realize motion analysis of the smoke and reduce the interference of environmental changes; and based on at least two of the first smoke data, the second smoke data and the third smoke data, smoke detection data is generated. The smoke detection data represents the current size of the smoke area, improves the detection accuracy of oil smoke, and avoids environmental interference.

[0178] In some embodiments, processing the current frame image data to obtain smoke data includes at least two of the following steps:

[0179] Generate first smoke data based on the current frame image data and set threshold data, where the set threshold data is used to characterize the distribution of smoke color intervals;

[0180] generating second smoke data based on the current frame image data and the background image data, wherein the background image data represents a background image in a smoke-free state;

[0181] The third smoke data is generated based on the current frame image data and a set number of historical frame image data before the current frame image data.

[0182] First smoke data is generated based on the current frame image data and set threshold data, and the set threshold data is used to characterize the color interval distribution of the smoke.

[0183] The threshold data is set to characterize the distribution of smoke color intervals. Based on the set threshold data, it can be determined within which color range smoke exists, thereby determining the image data corresponding to the set threshold data in the current frame to generate the first smoke data. The set threshold can be generated based on smoke image data at different levels of density. For example, assuming that the color space type corresponding to the set threshold is HSV, smoke image data at different levels of density can be collected and converted to the HSV color space. The mean μ and standard deviation σ of the sample smoke data are calculated, and the HSV data value of the smoke pixel color is calculated according to the probability density statistical formula. For example, the set threshold data can be expressed as [μ-3*σ,μ+3*σ] to characterize the smoke color distribution.

[0184] Second smoke data is generated based on the current frame image data and the background image data, where the background image data represents a background image in a smoke-free state.

[0185] The background image data represents the background image in a smoke-free state. To prevent the background image data from affecting the smoke data, the background image data is used as a reference for the smoke-free state to separate the possible smoke area from the current frame image data, thereby generating the second smoke data. This can reduce the impact of background noise and improve the accuracy of smoke detection.

[0186] The third smoke data is generated based on the current frame image data and a set number of historical frame image data before the current frame image data.

[0187] The accuracy of smoke detection can be enhanced based on temporal continuity. Smoke data corresponding to images between consecutive frames can be determined based on a set number of historical image frames preceding the current frame and the current frame, generating third smoke data. This better captures the dynamic characteristics of smoke. For example, assuming the set number is five, the third smoke data is generated based on the current frame and the five frames preceding it.

[0188] In this way, the first smoke data of the embodiment of the present disclosure can better highlight the color and brightness characteristics of the smoke and reduce interference from other color information. The second smoke data is generated based on the current frame image data and background image data, which can eliminate the influence of fixed backgrounds and static objects, thereby avoiding environmental interference. The third smoke data combines the current frame and a set number of historical frame data to analyze the movement and diffusion of the smoke and reduce interference from environmental changes. Based on at least two of the first smoke data, the second smoke data, and the third smoke data, smoke detection data is generated. The smoke detection data represents the current size of the smoke area, improves the accuracy of oil smoke detection, and avoids environmental interference.

[0189] In some embodiments, the range hood further includes a fan, and the method further includes:

[0190] Based on the smoke detection data and the set smoke threshold, the fan operation is controlled.

[0191] The user sets one or more smoke thresholds based on actual operating requirements and environmental conditions. For example, smoke detection data is compared with the set smoke thresholds, and a decision to control fan operation is made based on the comparison result between the smoke detection data and the thresholds.

[0192] If the judgment result shows that the smoke exceeds the set threshold, the fan is controlled to operate. For example, the speed, direction, working mode and other parameters of the fan can be adjusted according to the severity and location information of the smoke to achieve the best smoke exhaust effect.

[0193] In some embodiments, generating smoke detection data based on the smoke data includes:

[0194] Perform binarization processing on each type of smoke data to generate binary data corresponding to each type of data;

[0195] Generate smoke detection data based on the binary data.

[0196] It should be noted that binarization can enhance the contrast in an image, making the difference between the smoke area and the background more distinct, which is beneficial for smoke detection and segmentation. Binarization converts an image into a black and white (or binary) image. In this process, the grayscale value or color information of each pixel is simplified to two possible values: usually black (usually represented by 0 or a low level) and white (usually represented by 255 or a high level).

[0197] For example, in a grayscale image, each pixel has a grayscale value between 0 and 255, where 0 represents black, 255 represents white, and values ​​in between represent varying shades of gray. During binarization, a threshold T is selected, and the grayscale value of each pixel is compared with this threshold: if the grayscale value of the pixel is less than or equal to the threshold T, the pixel is assigned a value of 0 (or a set black value). If the grayscale value of the pixel is greater than the threshold T, the pixel is assigned a value of 255 (or a set white value).

[0198] By binarizing the three different types of smoke data, we can extract the smoke features in each data. These features may vary in different data, but all contain smoke information.

[0199] In addition, generating smoke detection data based on at least two binary data sets can achieve data fusion and complementary effects. Different smoke detection methods may perform better in certain situations. By fusing the results of multiple methods, the overall detection performance and robustness can be improved.

[0200] In some embodiments, generating smoke detection data based on the smoke data includes:

[0201] determining whether the third smoke data is valid smoke data, and if it is determined that the third smoke data is valid smoke data, generating smoke detection data based on at least two of the first smoke data, the second smoke data, and the third smoke data;

[0202] If it is determined that the third smoke data is not valid smoke data, smoke detection data is generated based on the first smoke data and the second smoke data.

[0203] It can be understood that the third smoke data is generated by combining multi-frame image information, that is, based on the current frame and the current frame image data and a set number of historical frame image data before the current frame image data. In the process of combining with multi-frame images, additional noise, errors or interference may be introduced.

[0204] By determining whether the third smoke data is valid smoke data, adverse effects can be filtered out, ensuring the final third smoke detection. Determine whether the third smoke data is valid smoke data.

[0205] If the third smoke data is determined to be valid, smoke detection data is generated based on at least two of the first, second, and third smoke data. If the third smoke data is determined not to be valid, the third smoke data is deemed invalid, and the fourth smoke data may be omitted from subsequent smoke fusion with the first and / or second smoke data to generate smoke detection data, thereby avoiding wasting computing resources and improving smoke detection accuracy.

[0206] In some embodiments, the method further comprises:

[0207] generating a standard deviation of a contour center of gravity of the third smoke data based on the third smoke data and an edge detection algorithm, wherein the standard deviation is used to characterize a contour dispersion of the third smoke data;

[0208] If the standard deviation is greater than or equal to the profile dispersion threshold, the third smoke data is determined to be valid smoke data.

[0209] The edge detection algorithm can effectively extract the boundaries of smoke areas in smoke data and generate edge data. After obtaining the edge data, the contour extraction algorithm can be used to obtain the contour of the third smoke data. A contour is a closed path along the edge of an object, reflecting the shape characteristics of the object. For each contour, its center of gravity (or center of mass) can be calculated. The center of gravity is the average of the coordinates of all pixel points in the contour, usually expressed as (x, y) coordinates. The standard deviation of the coordinates of the center of gravity of all contours in the x and y directions is calculated. The standard deviation is a statistic that measures the degree of dispersion of the data distribution. A larger value indicates a more dispersed distribution of data points, while a smaller value indicates a more concentrated distribution of data points.

[0210] Compare the calculated standard deviation to the preset contour dispersion threshold. If the standard deviation is greater than or equal to the contour dispersion threshold, it indicates that the contour distribution of the third smoke data is relatively discrete and may represent true smoke characteristics. Conversely, if the standard deviation is less than the contour dispersion threshold, it may indicate that the contour is too concentrated or affected by noise, and is not sufficient to serve as valid smoke data.

[0211] In this way, the validity of the third smoke data is judged based on its contour dispersion. Data with larger contour dispersion is more likely to represent the actual smoke situation, while data with smaller dispersion may need to be analyzed or excluded, which helps improve the accuracy of smoke detection.

[0212] In some embodiments, generating first smoke data based on current frame image data and set threshold data includes:

[0213] Performing color space conversion based on the first color space data of the current frame image data to generate second color space data of the current frame image data;

[0214] First smoke data is generated based on the second color space data and the set threshold data.

[0215] Perform a color space conversion on the first color space data of the current frame image data. Common color spaces include RGB, HSV, YUV, Lab, etc. The purpose of color space conversion is to convert image data from one color representation to another to facilitate subsequent smoke detection processing.

[0216] For example, assuming that the first color space data is RGB color space, it is converted into HSV color space data, i.e., second color space data. Based on the converted second color space data (such as HSV color space data) and the set threshold data, the first smoke data is generated. The set threshold data is usually used to characterize the distribution of smoke in a specific color range. HSV color space data includes: hue (Hue), saturation (Saturation) and brightness (Value) range in HSV color space,

[0217] In this way, by generating the first smoke data based on data in different color spaces and a set threshold, the smoke area in the image can be more accurately identified and extracted. The choice of color space conversion and threshold application depends on the specific smoke characteristics and application scenario.

[0218] In some embodiments, generating first smoke data based on the second color space data and the set threshold data includes:

[0219] generating initial first smoke data based on the second color space data and the set threshold data;

[0220] Morphological processing is performed on the initial first smoke data to generate first smoke data.

[0221] Due to the fluidity of smoke, morphological processing can also be performed on the initial first smoke data.

[0222] Morphological processing is a shape- and structure-based image analysis method that includes operations such as dilation, erosion, opening, closing, and edge detection. Morphological processing can be used to remove noise, smooth edges, fill holes, and separate connected objects.

[0223] Exemplarily, for the initial first smoke data, one or more of the following morphological operations can be performed: that is, morphological dilation and erosion operations are performed on the initial first smoke data (the number of dilation operations needs to be greater than the number of erosion operations) to obtain the final first smoke data, thereby improving the integrity and accuracy of the first smoke data.

[0224] In some embodiments, generating second smoke data based on the current frame image data and the background image data includes:

[0225] Based on the current frame image data and the background image data, an image difference algorithm is performed to generate pixel data of the difference area;

[0226] Second smoke data is generated based on the pixel data of the difference area.

[0227] Image difference is used to compare the pixel value changes of the current frame image data and the background image data, thereby generating pixel data of the difference area. Based on the pixel data of the difference area, second smoke data can be generated to indicate the existence and changes of smoke.

[0228] In some embodiments, generating the second smoke data based on the pixel data of the difference area includes:

[0229] Extracting pixel data of the difference area that meets the set threshold based on the pixel data of the difference area and the set threshold;

[0230] Based on the background image data and the grayscale threshold, extracting the background image data that meets the grayscale threshold;

[0231] The second smoke data is generated based on the pixel data of the difference area meeting the set threshold and the background image data meeting the grayscale threshold.

[0232] In order to extract significant difference pixel data in the difference area, filter out minor changes and noise, based on the pixel data of the difference area and a set threshold, extract the pixel data of the difference area that meets the set threshold. The set threshold here can be 10.

[0233] Considering that smoke colors are close to a specific gray or white, the background image can be processed. This involves converting it to grayscale and setting a grayscale threshold. Extracting background image data that meets this threshold helps highlight the smoke area. Thresholding the background can reduce its impact on smoke detection. A grayscale threshold of 180 is recommended.

[0234] The pixel data of the difference area that meets the set threshold is subjected to logical operations (such as AND, OR, NOT, etc.) with the background image data that meets the grayscale threshold to highlight the smoke area and suppress the influence of background and noise. The resulting second smoke data can more clearly show the location, shape, and intensity of the smoke.

[0235] For example, considering that the smoke color is close to a specific gray or white, it is necessary to enhance the difference values ​​of the white background area and the color area close to the smoke color. That is, extract the pixel area with a grayscale value greater than 180 in the background frame, and extract the pixel area with a difference value greater than 10 in the difference image. Extracting and filtering the pixel areas in the two images to extract the common parts of the pixel areas that meet the same conditions can improve the accuracy of the second smoke data.

[0236] In some embodiments, generating third smoke data based on the current frame image data and a set number of historical frame image data before the current frame image data includes:

[0237] For a set number of continuous historical frame image data, performing a differential operation based on the current frame image data and each historical frame image data to generate pixel data of multiple difference areas;

[0238] Based on the pixel data of the plurality of difference areas, third smoke data is generated.

[0239] By performing image difference algorithm processing on the current frame image data and the background image data, background interference can be accurately identified and filtered out, and second smoke data can be generated based on the pixel data of the difference area to achieve accurate detection and positioning of the smoke.

[0240] In some embodiments, generating third smoke data based on pixel data of the plurality of difference regions includes:

[0241] Performing superposition operation based on pixel data of multiple difference areas;

[0242] Determine the moving target and the contour information of the moving target based on the pixel data of the multiple difference areas after the superposition operation;

[0243] Based on the contour information and the contour information threshold, third smoke data is generated.

[0244] Based on the pixel data of multiple difference areas, an overlay operation is performed to obtain the pixel data of multiple difference areas after the overlay operation, thereby forming a more comprehensive representation of the running target. During the overlay process, a binarization process can be performed simultaneously. For example, the current frame is differentiated from the previous five frames, and the data grayscale binarization is performed simultaneously. Finally, the pixel data of the five frames after differentiation and data grayscale binarization are superimposed to obtain the final differential binarization data.

[0245] Based on the pixel data from multiple difference regions after superposition, the system can determine the specific position and shape of the moving target and then extract its contour information. Contour information includes the target's boundaries and shape characteristics. In actual working conditions, since smoke particles are relatively small, and frying spoons and pots are relatively large, they will interfere with subsequent smoke detection results.

[0246] The third smoke data is generated based on the extracted contour information and a preset contour information threshold. If the contour information is greater than or equal to the preset contour information threshold, it indicates that the contour information of the moving target is large and needs to be removed. After removal, the third smoke data is generated again. The contour information threshold is the contour area threshold, which is set to 200.

[0247] If the contour information is less than the preset contour information threshold, it indicates that the contour of the moving target is small and will not affect the smoke detection results. The third smoke data can be directly generated. In this way, the interference of the moving target is eliminated, ensuring that the generated third smoke data is more accurate.

[0248] The embodiment of the present disclosure is described in detail below with reference to an application example.

[0249] Home range hoods are increasingly equipped with a variety of functions, but their core function remains smoke extraction. In practice, users often forget or fail to activate their range hoods in time. Current intelligent detection methods typically utilize infrared light, which is limited by its detection angle and typically requires the sensor to be placed at the rear end (near the range hood's exhaust duct). Regardless of the aforementioned situation, this can lead to untimely smoke extraction, causing smoke to escape.

[0250] When detecting cooking smoke through images, changes in the external environment, such as the user's hands or the movement of a cooking spoon, can affect the detection and thus the accuracy of smoke detection. Therefore, how to eliminate environmental interference during the detection process is a key issue that needs to be addressed in the embodiments of the present disclosure.

[0251] This application example provides a smoke detection solution for range hoods 12-1100, primarily using a combination of three different solutions. First, a wide-angle camera (the aforementioned visual detection sensor) is installed on the range hood to capture the cooking conditions on both burners of the stove. Refer to Figures 1-2, which show the current raw image captured by the camera (the aforementioned current frame of image data).

[0252] Below, three solutions are described.

[0253] Solution 1: Color Extraction. Referring to FIG. 1-3 , FIG. 1-3 is a schematic diagram of the workflow of Solution 1. Solution 1 includes steps 1-301 to 1-302.

[0254] In step 1-301: pre-processing.

[0255] In step 1-302: collect samples.

[0256] According to the difference between the color of the smoke and the color of the environment, smoke image data with different density can be collected.

[0257] In step 1-303: calculate the mean μ and variance σ of the sample.

[0258] Generally speaking, smoke image data at different levels of density is in the RGB color space. First, convert it to the HSV color space, calculate the mean μ and standard deviation σ of the sample smoke data, and then calculate the HSV data value of the smoke pixel color based on the probability density statistical formula.

[0259] Step 1-304: Capture a picture of the actual scene (ie, the aforementioned current frame image data).

[0260] Get the current frame image data generated by the visual detection sensor, that is, get the original image shown in Figure 1-2 above.

[0261] In step 1-305: extract smoke pixels according to the probability distribution 3σ criterion.

[0262] [μ-3*σ, μ+3*σ] is used as the aforementioned threshold value. In some embodiments, the RGB pixel space of the captured image is converted to the HSV space, and a color space conversion is performed based on the first color space data of the current frame image data to generate the second color space data of the current frame image data; the initial first smoke data is generated based on the second color space data and the threshold value data.

[0263] Exemplarily, smoke data in the actual test image is extracted based on [μ-3*σ, μ+3*σ] to generate initial first smoke data, and the extracted smoke data (ie, the aforementioned initial first smoke data) is converted into grayscale binarization and saved.

[0264] In step 1-306: Morphological dilation erosion.

[0265] In some embodiments, before grayscale binarization is performed, morphological processing may be performed on the initial first smoke data to generate first smoke image data.

[0266] For example, due to the fluidity of smoke, morphological dilation and erosion operations are performed on the extracted pixel image (the number of dilation operations needs to be greater than the number of erosion operations), ultimately obtaining first smoke data, which is then binarized to obtain image data frame 1 (i.e., the aforementioned first binarized data). Referring to Figures 1-4 , Figures 1-4 are schematic diagrams of a smoke image after color extraction.

[0267] In step 1-307: output color extraction data frame 1.

[0268] In this way, the color extraction scheme can better highlight the color and brightness characteristics of the smoke and reduce the interference of other color information.

[0269] Solution 2: Static Filtering. This solution primarily filters background noise. When the stove is turned on, the range hood camera begins recording the video stream until the stove is turned off. See Figure 1-5 for a workflow diagram of Solution 2, which includes steps 1-501 through 1-507.

[0270] In step 1-501: pre-processing.

[0271] In step 1-502: cache the first frame of picture data as a background frame (ie, the aforementioned background image data).

[0272] In some embodiments, the first frame data is cached as a background frame to facilitate background filtering after comparison with subsequent frames.

[0273] In step 1-503: the current frame is differentiated from the background frame.

[0274] In some embodiments, subsequent data frames are differenced from the first frame.

[0275] In step 1-504: the differential data frame is converted into a grayscale image.

[0276] In step 1-505: a specific area of ​​the differential data frame (corresponding to the white background or the area with a color close to the smoke color) is enhanced.

[0277] Considering the proximity of smoke to a specific gray or white color, we need to enhance the difference values ​​of white background areas and those with colors close to the smoke color. Specifically, we extract pixel regions with grayscale values ​​greater than 180 in the background frame, and those with difference values ​​greater than 10 in the difference image. We then extract and filter the pixel regions in both images to identify the common regions that meet the same criteria. After normalizing the data values ​​of these common regions, we overlay them with the original image to obtain the enhanced, effective difference data. Similarly, we can also extract and enhance background regions with color differences close to the smoke color.

[0278] In step 1-506: the differential data frame is grayscale binarized.

[0279] The added differential data frame (i.e., the aforementioned second smoke data) is also grayscale binarized. The final image data frame 2 (i.e., the aforementioned second binarized data) is obtained through static filtering. Referring to FIG. 1-6 , FIG. 1-6 is a schematic diagram of the image of the static differential frame filtered data frame 2.

[0280] In step 1-507: output the background filtered data frame 2 (ie, the aforementioned second binarized data).

[0281] In this way, static filtering can eliminate the influence of fixed background and static objects, thereby avoiding environmental interference.

[0282] Option 3: Dynamic Filtration. This method primarily considers the mobility of smoke and the impact of factors such as flipping a pan or frying a spoon during filtration testing. Figure 1-7 illustrates the workflow for Option 3. Option 3 includes steps 1-701 through 1-708.

[0283] In step 1-701: pre-processing.

[0284] In step 1-702: cache the first 5 frames of image data.

[0285] Cache the set number of historical frame image data before the current frame image data, the set number is 5.

[0286] In step 1-703: the current frame is differentiated from the previous five frames.

[0287] In some embodiments, based on a set number of consecutive historical frame image data, a differential operation is performed between the current frame image data and each historical frame image data to generate pixel data of a plurality of difference areas.

[0288] For example, in solution three, the dynamic filtering uses a rolling difference method to perform a difference operation to generate pixel data of multiple difference areas.

[0289] In step 1-704: the differential data is converted into a grayscale image and binarized.

[0290] The current frame can be differentiated from the previous five frames, and the data grayscale binarization can be performed synchronously. Finally, the pixel data of the five frames after differentiation and data grayscale binarization can be superimposed to obtain the final differential binarized data.

[0291] In step 1-705: 5 frames of differential data are superimposed and an OR operation is performed.

[0292] In some embodiments, a superposition operation is performed based on pixel data of multiple difference regions.

[0293] In step 1-706: calculate the contour and remove the area with a larger contour area.

[0294] In actual working conditions, since smoke particles are relatively small, while frying spoons and pots are relatively large, it is necessary to detect contours in the superimposed differential data and eliminate pixels in areas with relatively large contour areas (the contour area threshold needs to be set according to the test data, and the threshold used in the test is 200).

[0295] In step 1-707: calculate the contour dispersion to confirm whether it is valid smoke data.

[0296] In some embodiments, before obtaining the final valid data frame three (i.e., the aforementioned third binarized data), it is necessary to calculate the contour dispersion to confirm whether it is valid smoke data. For example, based on the third smoke data and an edge detection algorithm, the standard deviation of the contour centroid of the third smoke data is generated. The standard deviation is used to represent the contour dispersion of the third smoke data. If the standard deviation is greater than or equal to the contour dispersion threshold, the third smoke data is determined to be valid smoke data.

[0297] In step 1-708: output the dynamic filtering data frame three (ie the aforementioned second binarization process).

[0298] If it is determined to be valid smoke data, after binarization processing, dynamic filtering data frame three is output (ie the second binarization processing mentioned above). Referring to Figure 1-8, Figure 1-8 is a schematic diagram of a smoke image of dynamic difference frame data frame three.

[0299] In this way, dynamic filtering can be used to analyze the movement and diffusion of smoke and reduce the interference of environmental changes.

[0300] Based on at least two of the above-mentioned solutions 1, 2, and 3, a result data frame (i.e., the aforementioned smoke detection data) is generated. Referring to FIG. 1-9 , FIG. 1-9 is a schematic diagram of a process for generating a result data frame. The process for generating a result data frame includes steps 1-901 to 1-906.

[0301] In step 1-901: pre-processing.

[0302] In step 1-902: dynamic difference frame discreteness determination.

[0303] In order to ensure that the dynamic difference frame output by scheme three is valid smoke, it is necessary to judge the discreteness of the dynamic difference frame.

[0304] In step 1-903: determine whether it is valid smoke, if it is valid smoke, execute step 1-904, if it is not valid smoke, execute step 1-906.

[0305] The discreteness threshold and the discreteness of the dynamic difference frame can be compared to determine whether the smoke is valid. If the smoke is determined to be invalid based on the discreteness, no processing is performed. Specifically, a result data frame is generated based on the image frames 1 and 2 output from the above-described schemes 1 and 2, the result is output, and steps 1-906 are executed.

[0306] If it is determined to be valid smoke, smoke detection data is generated based on at least two of the above-mentioned solutions 1, 2, and 3, and step 1-904 is executed.

[0307] In step 1-904: perform data processing [data frame one and data frame two or data frame three].

[0308] Because data frames 1, 2, and 3 are all binary data, logical operations can be performed on each data frame to generate a result data frame (i.e., the aforementioned smoke detection data). The result data frame here is also binary data. The logical operation formula is as follows:

[0309] Result data frame = [data frame one and data frame two or data frame three].

[0310] Refer to Figure 1-10, which is a schematic image diagram of the result data frame.

[0311] In step 1-905: calculate the ratio of white pixels to the entire image.

[0312] For example, smoke detection data indicates the current size of the smoke area. Based on this smoke detection data and a set smoke threshold, the fan is controlled. Specifically, in the result data frame, the proportion of white pixels in the entire image is calculated to represent the smoke size. The smoke size is then compared to the set smoke threshold to generate a comparison result. This comparison result is used to determine whether to control the fan.

[0313] In step 1-906: Output the results. In this application example, the combination of the above solutions eliminates the impact of environmental interference on smoke detection. For example, when detecting cooking smoke through images, cooking utensils such as the user's hands or the movement of a frying pan can affect detection. This improves smoke detection efficiency.

[0314] In a second aspect, to implement the method of the present embodiment, the present embodiment further provides a third control device for the range hood 12-1100. The third control device comprises a third acquisition module, a third processing module, and a third generation module. The third acquisition module is configured to acquire image data generated by the visual detection sensor. The third processing module is configured to determine smoke data based on the image data. The third generation module is configured to control the range hood based on the smoke data.

[0315] In some embodiments, to implement the method of the embodiments of the present disclosure, the embodiments of the present disclosure further provide a second control device for the range hood. As shown in Figure 2-10, the second control device 2-1000 of the range hood includes: a second acquisition module 2-1010, a second determination module 2-1020, and a second control module 2-1030. The second acquisition module 2-1010 is used to acquire image data generated by the visual detection sensor; the second determination module 2-1020 is used to determine smoke data corresponding to the image data based on the second color space data; based on the smoke data and the gear thresholds of at least two operating gears, the target operating gear of the fan is determined; and the second control module 2-1030 is used to control the fan to operate at the target operating gear.

[0316] In some embodiments, the second control device 2-1000 of the range hood 12-1100 also includes a second conversion module 2-1040, which is used to perform color space conversion based on the first color space data of the image data to generate second color space data of the image data; the second determination module 2-1020 is also used to determine the smoke data corresponding to the image data based on the second color space data.

[0317] In some embodiments, the second acquisition module 2-1010 is also used to obtain the first color space type corresponding to the first color space data; the second determination module 2-1020 is used to determine the second color space type to be converted based on the first color space type; the second conversion module 2-1040 is used to perform color space conversion on the first color space data based on the conversion rules corresponding to the second color space type to generate second color space data.

[0318] In some embodiments, the second acquisition module 2-1010 is also used to obtain first color pixel feature data corresponding to the first color space data; the second determination module 2-1020 is also used to determine the second color pixel feature data of the second color space data based on the first color pixel feature data and the conversion rule; the second control device 2-1000 of the range hood 12-1100 also includes a second generation module 2-1050, which is used to generate second color space data based on the second color pixel feature data.

[0319] In some embodiments, the second determination module 2-1020 is further used to determine third color space data based on the second color space data and the smoke color interval, where the third color space data is the second color space data within the smoke color interval; the second generation module 2-1050 is further used to generate smoke data based on the third color space data.

[0320] In some embodiments, the second determination module 2-1020 is also used to determine the mean and variance value of the second color pixel feature data based on the second color space data; the second generation module 2-1050 is also used to generate the upper limit value and lower limit value of the smoke color range based on the mean and variance value; the second determination module 2-1020 is also used to determine the smoke color range based on the upper limit value and the lower limit value.

[0321] In some embodiments, the second control device 2-1000 of the range hood 12-1100 also includes a second processing module 2-1060 for performing numerical normalization processing on the third color space data; the second generation module 2-1050 is also used to perform bit operations on the third color space data and the first color space data after numerical normalization processing to generate smoke data representing the size of the oil smoke area.

[0322] In some embodiments, the second determination module 2-1020 is also used to determine the proportion information of the current oil smoke area to the area of ​​the image based on the size information of the image corresponding to the smoke data and the image data; and determine the target operating gear of the fan based on the proportion information and the gear threshold.

[0323] In some embodiments, the second control device 2-1000 of the range hood 12-1100 also includes a second switching module 2-1070, which is used to switch the fan operation to the operating gear corresponding to the first gear threshold if it is determined that the current proportional information is greater than or equal to the first gear threshold, and the first gear threshold is the switching threshold of the operating gear above the fan speed of the current operating gear; if it is determined that the current proportional information is less than the second gear threshold, then after determining that the difference between the second gear threshold and the current proportional information is greater than or equal to the set value, switch the fan operation to the operating gear corresponding to the second gear threshold, and the second gear threshold is the switching threshold of the operating gear below the fan speed of the current operating gear.

[0324] In some embodiments, the second acquisition module 2-1010, the second determination module 2-1020, the second control module 2-1030, the second conversion module 2-1040, the second generation module 2-1050, the second processing module 2-1060, and the second switching module 2-1070 may be implemented by the processor 12-1101 in the second control device 2-1000 of the range hood 12-1100. The processor 12-1101 needs to execute a computer program in the memory 12-1102 to implement its functions.

[0325] It should be noted that the second control device 2-1000 of the range hood 12-1100 provided in the above embodiment is merely an example of the division of the aforementioned program modules when controlling the range hood 12-1100. The aforementioned processing can be distributed among different program modules as needed, i.e., the internal structure of the device can be divided into different program modules to perform all or part of the aforementioned processing. The second control device 2-1000 of the range hood 12-1100 provided in the above embodiment and the control method embodiment of the range hood 12-1100 are based on the same concept. The implementation process is detailed in the method embodiment and will not be further described here.

[0326] In some embodiments, to implement the method of the embodiments of the present disclosure, the embodiments of the present disclosure further provide a first control device 1-1000 for a range hood 12-1100. As shown in FIG1-11 , the first control device 1-1000 of the range hood 12-1100 includes: a first acquisition module 1-1010, a first processing module 1-1020, and a first generation module 1-1030. The first acquisition module 1-1010 is configured to acquire current frame image data generated by a visual detection sensor; the first processing module 1-1020 is configured to process the current frame image data to obtain smoke data, which includes at least two of first smoke data, second smoke data, and third smoke data; wherein the first smoke data is configured to represent smoke concentration data based on color extraction, the second smoke data is configured to represent smoke concentration data based on static filtering extraction, and the third smoke data is configured to represent smoke concentration data based on dynamic filtering extraction; and the first generation module 1-1030 is configured to generate smoke detection data based on the smoke data, wherein the smoke detection data represents the current size of the smoke area.

[0327] In some embodiments, the first processing module 1-1020 is also used to generate the first smoke data based on the current frame image data and set threshold data, and the set threshold data is used to characterize the smoke color interval distribution; generate the second smoke data based on the current frame image data and background image data, and the background image data characterizes the background image in a smoke-free state; and generate the third smoke data based on the current frame image data and a set number of historical frame image data before the current frame image data.

[0328] In some embodiments, the first control device 1-1000 of the range hood 12-1100 includes a first control device 1-1000 for controlling the operation of the fan based on smoke detection data and a set smoke threshold.

[0329] In some embodiments, the first generating module 1-1030 is further configured to perform binarization processing on each type of the smoke data to generate binarized data corresponding to each type of data; and generate the smoke detection data based on the binarized data.

[0330] In some embodiments, the first control device 1-1000 of the range hood 12-1100 also includes a first determination module 1-1050, which is used to determine whether the third smoke data is valid smoke data. If the third smoke data is determined to be valid smoke data, smoke detection data is generated based on at least two of the first smoke data, the second smoke data and the third smoke data; if the third smoke data is determined not to be valid smoke data, smoke detection data is generated based on the first smoke data and the second smoke data.

[0331] In some embodiments, the first determination module 1-1050 is further used to generate a standard deviation of the contour center of gravity of the third smoke data based on the third smoke data and the edge detection algorithm, and the standard deviation is used to characterize the contour discreteness of the third smoke data; if the standard deviation is greater than or equal to the contour discreteness threshold, the third smoke data is determined to be valid smoke data.

[0332] In some embodiments, the first control device 1-1000 of the range hood 12-1100 also includes a first conversion module 1-1060, which is used to perform color space conversion based on the first color space data of the current frame image data to generate second color space data of the current frame image data; the first generation module 1-1030 is also used to generate first smoke data based on the second color space data and set threshold data.

[0333] In some embodiments, the first generating module 1-1030 is further configured to generate initial first smoke data based on the second color space data and set threshold data; and perform morphological processing on the initial first smoke data to generate first smoke data.

[0334] In some embodiments, the first generation module 1-1030 is further used to perform an image difference algorithm based on the current frame image data and the background image data to generate pixel data of the difference area; and generate second smoke data based on the pixel data of the difference area.

[0335] In some embodiments, the first control device 1-1000 of the range hood 12-1100 also includes a first extraction module 1-1070, which is used to extract pixel data of the difference area that meets the set threshold based on the pixel data of the difference area and the set threshold; and extract background image data that meets the grayscale threshold based on the background image data and the grayscale threshold; the first generation module 1-1030 is also used to generate second smoke data based on the pixel data of the difference area that meets the set threshold and the background image data that meets the grayscale threshold.

[0336] In some embodiments, the first generation module 1-1030 is also used to perform a differential operation on the set number of continuous historical frame image data based on the current frame image data and each of the historical frame image data to generate pixel data of multiple difference areas; and generate third smoke data based on the pixel data of the multiple difference areas.

[0337] In some embodiments, the first determination module 1-1050 is further used to perform an overlay operation based on pixel data of multiple difference areas; determine the moving target and the contour information of the moving target based on the pixel data of multiple difference areas after the overlay operation; the first generation module 1-1030 is further used to generate third smoke data based on the contour information and the contour information threshold.

[0338] In some embodiments, the first acquisition module 1-1010, the first processing module 1-1020, the first generation module 1-1030, the first control module 1-1040, the first determination module 1-1050, the first conversion module 1-1060, and the first extraction module 1-1070 can be implemented by the processor 12-1101 in the first control device 1-1000 of the range hood 12-1100. The processor 12-1101 needs to run the computer program in the memory 12-1102 to implement its functions. It should be noted that the first control device 1-1000 of the range hood 12-1100 provided in the above embodiment only uses the division of the above-mentioned program modules as an example to illustrate the control of the range hood 12-1100. In some embodiments, the above-mentioned processing can be assigned to different program modules as needed, that is, the internal structure of the device can be divided into different program modules to complete all or part of the processing described above. The first control device 1-1000 of the range hood 12-1100 provided in the above embodiment and the control method embodiment of the range hood 12-1100 belong to the same concept. The implementation process is detailed in the method embodiment and will not be repeated here.

[0339] Thirdly, based on the hardware implementation of the program modules described above, and in order to implement the methods of the present embodiments, the present embodiments further provide a range hood 12-1100. Figures 1-12 illustrate only exemplary structures of the range hood 12-1100, not all of them. Part or all of the structures shown in Figures 1-12 may be implemented as needed.

[0340] As shown in FIG1-12, the range hood 12-1100 provided in an embodiment of the present disclosure includes: at least one processor 12-1101, a memory 12-1102, a user interface 12-1103, and at least one network interface 12-1104. The various components in the range hood 12-1100 are coupled together via a bus system 12-1105. It will be understood that the bus system 12-1105 is used to enable connection and communication between these components. In addition to including a data bus, the bus system 12-1105 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clarity, in FIG1-12, the various buses are labeled as the bus system 12-1105.

[0341] The user interface 12-1103 may include a display, keyboard, mouse, trackball, click wheel, keys, buttons, touch pad or touch screen, etc.

[0342] The memory 12-1102 in the disclosed embodiment is used to store various types of data to support the operation of the range hood 12-1100. Examples of such data include any computer program used to operate on the range hood 12-1100.

[0343] The control method of the range hood 12-1100 disclosed in the embodiment of the present disclosure can be applied to the processor 12-1101, or implemented by the processor 12-1101. The processor 12-1101 may be an integrated circuit chip with signal processing capabilities. During the implementation process, the various steps of the control method of the range hood 12-1100 can be completed by the hardware integrated logic circuit in the processor 12-1101 or by instructions in the form of software. The above-mentioned processor 12-1101 can be a general-purpose processor, a digital signal processor (DSP, Digital Signal Processor), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 12-1101 can implement or execute the various methods, steps and logic block diagrams disclosed in the embodiment of the present disclosure. The general-purpose processor can be a microprocessor or any conventional processor, etc. The steps of the method disclosed in combination with the embodiment of the present disclosure can be directly embodied as being executed by a hardware decoding processor, or being executed by a combination of hardware and software modules in the decoding processor. The software module may be located in a storage medium, which is located in the memory 12-1102. The processor 12-1101 reads the information in the memory 12-1102 and combines its hardware to complete the steps of the control method of the range hood 12-1100 provided in the embodiment of the present disclosure.

[0344] In an exemplary embodiment, the range hood 12-1100 can be implemented by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontrollers (MCUs), microprocessors, or other electronic components to perform the aforementioned method.

[0345] It is understood that the memory 12-1102 can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a flash memory, a magnetic surface memory, an optical disk, or a compact disc read-only memory (CD-ROM); the magnetic surface memory can be a magnetic disk or a magnetic tape. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), synchronous static random access memory (SSRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct rambus random access memory (DRRAM).The memory 12 - 1102 described in the embodiments of the present disclosure is intended to include, but is not limited to, these and any other suitable types of memory.

[0346] Fourthly, in exemplary embodiments, the present disclosure further provides a storage medium, namely, a computer storage medium, specifically a computer-readable storage medium, such as a memory 12-1102 storing a computer program. The computer program can be executed by the processor 12-1101 of the range hood to perform the steps of the method of the present disclosure. The computer-readable storage medium can be a memory such as a ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface mount storage, optical disk, or CD-ROM.

[0347] It should be noted that: "first", "second", etc. are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0348] In addition, the technical solutions described in the embodiments of the present disclosure can be arbitrarily combined without conflict.

[0349] The above are only specific embodiments of the present disclosure, but the scope of protection of the present disclosure is not limited thereto. Any changes or substitutions that can be easily conceived by any person skilled in the art within the technical scope disclosed in this disclosure should be included in the scope of protection of the present disclosure. Therefore, the scope of protection of the present disclosure should be based on the scope of protection of the claims.

Claims

1. A control method for an oil fume extractor, the oil fume extractor comprising: A vision detection sensor, the vision detection sensor is arranged above the cooking range, and the method includes: Obtaining image data generated by the vision detection sensor; Determining smoke data based on the image data; Controlling the range hood based on the smoke data.

2. The method according to claim 1, wherein The range hood further includes a blower; The determining the smoke data based on the image data includes: determining the smoke data corresponding to the image data based on the color space data of the image data; The controlling the range hood based on the smoke data includes: Determining a target operating gear of the blower based on the smoke data and gear thresholds of at least two operating gears; and Controlling the blower to operate at the target operating gear.

3. The method according to claim 2, wherein The determining the smoke data corresponding to the image data based on the color space data of the image data includes: Performing color space conversion based on the first color space data of the image data to generate second color space data of the image data; Determining the smoke data corresponding to the image data based on the second color space data.

4. The method according to claim 3, wherein, The performing color space conversion based on the first color space data of the image data to generate second color space data of the image data includes: Obtaining a first color space type corresponding to the first color space data; Determining a second color space type to be converted based on the first color space type; Performing color space conversion on the first color space data based on a conversion rule corresponding to the second color space type to generate second color space data.

5. The method according to claim 4, wherein The performing color space conversion on the first color space data based on a conversion rule corresponding to the second color space type to generate second color space data includes: Obtaining first color pixel feature data corresponding to the first color space data; Determining second color pixel feature data of the second color space data based on the first color pixel feature data and the conversion rule; Generating the second color space data based on the second color pixel feature data.

6. The method according to any one of claims 3 to 5, wherein The determining the smoke data corresponding to the image data based on the second color space data includes: Determining third color space data based on the second color space data and smoke color interval values, where the third color space data is the second color space data within the smoke color interval values; Generating the smoke data based on the third color space data.

7. The method according to any one of claims 3 to 6 further includes: Determining an average value and a variance value of the second color pixel feature data based on the second color pixel feature data of the second color space data; Generating an upper limit value and a lower limit value of the smoke color interval values based on the average value and the variance value; Determining the smoke color interval values based on the upper limit value and the lower limit value.

8. The method according to claim 6, wherein The generating the smoke data based on the third color space data includes: Performing numerical normalization processing on the third color space data; Perform a bit operation on the third color space data after normalizing the numerical value and the first color space data to generate the smoke data representing the size of the oil fume area.

9. The method according to any one of claims 2 to 8, wherein Determining the target operating gear of the fan based on the smoke data and the gear thresholds of at least two operating gears includes: Determine the proportional information of the current oil fume area occupying the area of the image based on the smoke data and the size information of the image corresponding to the image data; Determine the target operating gear of the fan based on the proportional information and the gear threshold.

10. The method according to claim 9 further includes: If it is determined that the current proportional information is greater than or equal to the first gear threshold, switch the fan to operate at the operating gear corresponding to the first gear threshold, where the first gear threshold is the switching threshold of the operating gear above the fan speed of the current operating gear; If it is determined that the current proportional information is less than the second gear threshold, after determining that the difference between the second gear threshold and the current proportional information is greater than or equal to the set value, switch the fan to operate at the operating gear corresponding to the second gear threshold, where the second gear threshold is the switching threshold of the operating gear below the fan speed of the current operating gear.

11. The method according to claim 1, wherein Obtaining the image data generated by the visual detection sensor includes: obtaining the current frame image data generated by the visual detection sensor; Determining the smoke data based on the image data includes: processing the current frame image data to obtain smoke data, where the smoke data includes at least two of the first smoke data, the second smoke data, and the third smoke data; where the first smoke data is used to represent the smoke concentration data extracted based on color, the second smoke data is used to represent the smoke concentration data extracted based on static filtration, and the third smoke data is used to represent the smoke concentration data extracted based on dynamic filtration; and Controlling the range hood based on the smoke data includes: Generate smoke detection data based on the smoke data, where the smoke detection data represents the size of the current smoke area.

12. The method according to claim 11, wherein, Processing the current frame image data to obtain smoke data includes at least two of the following steps: Generate the first smoke data based on the current frame image data and the set threshold data, where the set threshold data is used to represent the smoke color interval distribution; Generate the second smoke data based on the current frame image data and the background image data, where the background image data represents the background image in a smokeless state; Generate the third smoke data based on the current frame image data and a set number of historical frame image data before the current frame image data.

13. The method according to claim 11 or 12, wherein The range hood further includes a fan, and the method further includes: Control the operation of the fan based on the smoke detection data and the set smoke threshold.

14. The method according to any one of claims 11 to 13, wherein Generating the smoke detection data based on the smoke data includes: Perform binary processing on each type of data in the smoke data to generate binary data corresponding to each type of data; Generate the smoke detection data based on the binary data.

15. The method according to any one of claims 11 to 14, wherein, Generating smoke detection data based on the smoke data includes: Determining whether the third smoke data is valid smoke data. If it is determined that the third smoke data is valid smoke data, generating smoke detection data based on at least two of the first smoke data, the second smoke data, and the third smoke data; If it is determined that the third smoke data is not valid smoke data, generating smoke detection data based on the first smoke data and the second smoke data.

16. The method according to any one of claims 11 to 15 further includes: Generating a standard deviation of the contour centroid of the third smoke data based on the third smoke data and an edge detection algorithm, where the standard deviation is used to characterize the contour dispersion of the third smoke data; If the standard deviation is greater than or equal to a contour dispersion threshold, determining that the third smoke data is valid smoke data.

17. The method according to any one of claims 11 to 16, wherein, Generating the first smoke data based on the current frame image data and set threshold data includes: Performing color space conversion based on the first color space data of the current frame image data to generate second color space data of the current frame image data; Generating the first smoke data based on the second color space data and the set threshold data.

18. The method according to claim 17, wherein, Generating the first smoke data based on the second color space data and the set threshold data includes: Generating initial first smoke data based on the second color space data and the set threshold data; Performing morphological processing on the initial first smoke data to generate the first smoke data.

19. The method according to any one of claims 11 to 17, wherein, Generating the second smoke data based on the current frame image data and the background image data includes: Performing an image difference algorithm based on the current frame image data and the background image data to generate pixel data of a difference region; Generating the second smoke data based on the pixel data of the difference region.

20. The method according to claim 12 or 19, wherein, Generating the second smoke data based on the pixel data of the difference region includes: Extracting pixel data of the difference region that meets the set threshold based on the pixel data of the difference region and the set threshold; Extracting background image data that meets the gray level threshold based on the background image data and the gray level threshold; Generating the second smoke data based on the pixel data of the difference region that meets the set threshold and the background image data that meets the gray level threshold.

21. The method according to any one of claims 12 to 20, wherein Generating the third smoke data based on the current frame image data and a set number of historical frame image data before the current frame image data includes: For the set number of consecutive historical frame image data, performing a difference operation based on the current frame image data and each of the historical frame image data to generate pixel data of multiple difference regions; Generating the third smoke data based on the pixel data of the multiple difference regions.

22. The method according to any one of claims 12 to 21, wherein, Generating the third smoke data based on the pixel data of the multiple difference regions includes: Performing a superposition operation based on the pixel data of the multiple difference regions; Determining a moving target and contour information of the moving target based on the pixel data of the multiple difference regions after the superposition operation; Generating the third smoke data based on the contour information and a contour information threshold.

23. The third control device of an oil fume extractor, wherein, The range hood includes: a visual detection sensor, the visual detection sensor is disposed above the cooking range, and the third control device includes: a third acquisition module, which acquires the image data generated by the visual detection sensor; a third processing module, which is configured to determine smoke data based on the image data; a third control module, which is configured to control the range hood based on the smoke data.

24. An oil fume extractor, wherein, The range hood includes: a visual detection sensor and a blower, the visual detection sensor is disposed above the cooking range, and the range hood further includes: a processor and a memory for storing a computer program that can run on the processor, wherein, the processor, when running the computer program, executes the steps of the method according to any one of claims 1 to 22.

25. A computer storage medium, on which a computer program is stored, wherein, When the computer program is executed by the processor, the steps of the method according to any one of claims 14 to 22 are implemented.

Images