Human posture recognition method and recognition device for intelligent toilet

Abstract

The application provides a human posture recognition method and recognition device for a smart toilet, temperature array data of a user in a smart toilet area is collected through an infrared array sensor, and pre-processing is performed. A temperature data distribution map is formed; a human contour, a thermal centroid, a thermal area and a thermal distribution shape feature point are extracted from the pre-processed temperature data distribution map; whether a user exists in the smart toilet area is judged according to a heat source detection result in a field of view of the infrared array sensor, and the user is recognized as a standing posture or a sitting posture according to the feature point, and corresponding posture output information is output; precise input is provided for personalized intelligent control of automatic cover opening, ring opening, cover and ring not being closed by a person, automatic cover and ring being closed after a person leaves, flushing and disinfection and the like of the smart toilet, precise and non-contact recognition of a posture of the user in the smart toilet area is realized, and the problems of false triggering, high energy consumption and poor user experience existing in the prior art smart toilet are effectively solved.

Description

Technical Field

[0001] This invention relates to the fields of smart home and sensor technology, and in particular to a human posture recognition method and recognition device for a smart toilet. Background Technology

[0002] With the increasing popularity of smart homes, smart toilets, as an important component of modern bathroom products, are receiving growing attention for their level of intelligence. However, existing smart toilets still have many shortcomings in terms of user experience and energy efficiency. Traditional smart toilets often fail to accurately recognize the user's intentions, unable to distinguish between men urinating standing up and men sitting down, thus failing to provide personalized automatic control. Furthermore, the automatic lid opening / closing function of some existing smart toilets still requires manual operation or relies on simple proximity sensors, which is prone to false triggering. For example, the toilet lid may frequently open when pets or passersby are present, causing unnecessary energy consumption and hygiene hazards. These problems not only affect the user experience but also increase energy consumption.

[0003] Therefore, developing a human posture recognition method and device for smart toilets that can accurately and non-contactly recognize the user's posture in the smart toilet area, solves the problems of accidental closing of existing smart toilets when the user is present, failure to clean and disinfect in a timely manner after the user leaves, and the resulting hygiene hazards, high energy consumption and poor user experience, is of great significance for improving the user experience of smart toilets and achieving energy conservation and environmental protection. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a human posture recognition method and recognition device for smart toilets, which can solve the problems of existing smart toilets that may close accidentally when the user is present, fail to clean and disinfect in time after the user leaves, and the resulting hygiene hazards, high energy consumption and poor user experience.

[0005] To address the aforementioned problems, this invention provides a human posture recognition method for smart toilets, comprising: collecting temperature array data of a user in the smart toilet area using an infrared array sensor to form a temperature data distribution map; performing Gaussian filtering on the temperature data distribution map to remove environmental noise, and performing histogram equalization and normalization processing to form a preprocessed temperature data distribution map; extracting feature points from the preprocessed temperature data distribution map, the feature points including: human contour, thermal centroid, thermal area, and thermal distribution shape; determining whether a user exists in the smart toilet area based on the heat source detection results within the field of view of the infrared array sensor, identifying the user as standing or sitting based on the feature points, and outputting corresponding posture output information to control the smart toilet to perform corresponding operations.

[0006] In some embodiments, the step of identifying whether a user is standing or sitting based on the feature points further includes: when the thermal profile is a vertical human shape, the longitudinal coordinate of the thermal centroid is higher than the longitudinal midpoint of the thermal profile, and the width of the lateral temperature distribution remains relatively stable, it is determined to be a standing posture; when the thermal profile is a specific folded shape, a high-temperature hot spot appears in the seat area, and the longitudinal coordinate of the thermal centroid is lower than the longitudinal midpoint of the thermal profile, it is determined to be a sitting posture.

[0007] In some embodiments, in the step of determining whether a user exists in the smart toilet area based on the heat source detection result within the field of view of the infrared array sensor, when the heat source within the field of view of the infrared array sensor disappears, it is determined that the user has left. The posture output information includes user departure information. The smart toilet control system performs a delayed closing of the toilet lid and seat ring based on the user departure information, and performs flushing and disinfection operations.

[0008] In some embodiments, the posture output information is used to control the opening or closing of the toilet lid, the opening or closing of the seat, the flushing intensity of the flushing pump, the opening or closing of the disinfection module, and the seat temperature of the seat heater of the smart toilet.

[0009] To address the aforementioned technical problems, this application also provides a human posture recognition device for a smart toilet, comprising: an infrared array sensor for real-time acquisition of temperature array data of a user in the smart toilet area and forming a temperature data distribution map; a processing module connected to the infrared array sensor for receiving and processing the temperature data distribution map and outputting the user's posture output information, the processing module including a preprocessing unit, a feature extraction unit, and a posture classification unit, the preprocessing unit performing Gaussian filtering on the temperature data distribution map to remove environmental noise, and performing histogram equalization and normalization processing to form a preprocessed temperature data distribution map, the feature extraction unit extracting feature points from the preprocessed temperature data distribution map, the feature points including: human body contour, thermal centroid, thermal area, and thermal distribution shape, the posture classification unit determining whether a user exists in the smart toilet area based on the heat source detection results within the field of view of the infrared array sensor and identifying the user as standing or sitting based on the feature points, and outputting corresponding posture output information; and an output module connected to the processing module and the smart toilet control system for sending the posture output information to the control system for the control system to perform corresponding control.

[0010] In some embodiments, the infrared array sensor is an 8×8, 16×16, or 32×32 pixel infrared thermopile sensor.

[0011] In some embodiments, the infrared array sensor is installed behind or to the side of the toilet seat of the smart toilet, and the installation position ensures that the field of view of the infrared array sensor covers the area in front of the toilet.

[0012] In some embodiments, when the posture classification unit identifies a user as standing or sitting based on the feature points, it specifically includes: when the thermal profile is a vertical human shape, the longitudinal coordinate of the thermal centroid is higher than the longitudinal midpoint of the thermal profile, and the width of the lateral temperature distribution remains relatively stable, it is determined to be a standing posture; when the thermal profile has a specific folded shape, high-temperature hot spots appear in the seat area, and the longitudinal coordinate of the thermal centroid is lower than the longitudinal midpoint of the thermal profile, it is determined to be a sitting posture.

[0013] In some embodiments, the posture output information is used to control the opening or closing of the toilet lid, the opening or closing of the toilet seat, the flushing intensity of the flushing pump, and the seat temperature of the seat heater of the smart toilet.

[0014] In some embodiments, when the heat source disappears within the field of view of the infrared array sensor, the attitude output information output by the attitude classification unit includes user departure information. The output module sends the user departure information to the smart toilet control system, which controls the toilet lid and seat to close with a delay and performs flushing and disinfection operations.

[0015] The above technical solution collects temperature array data of the user in the smart toilet area using an infrared array sensor, and preprocesses it to form a preprocessed temperature data distribution map. It then extracts human contour, thermal centroid, thermal area, and thermal distribution shape feature points from the preprocessed temperature data distribution map. Based on the heat source detection results within the field of view of the infrared array sensor, it determines whether a user is present in the smart toilet area and identifies the user's posture (standing or sitting) based on the feature points, outputting corresponding posture information. This provides precise input for personalized intelligent control of the smart toilet, such as automatic lid opening, lid opening, automatic lid and lid closing when the user is present, automatic lid and lid closing after the user leaves, flushing, and disinfection. It achieves accurate, non-contact recognition of the user's posture in the smart toilet area, effectively solving the problems of false triggering, high energy consumption, and poor user experience existing in smart toilets.

[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Techniques, methods, and apparatus known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and apparatus should be considered part of the specification. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a human posture recognition method for a smart toilet provided in an embodiment of the present invention; Figure 2 This is a control decision logic diagram for human posture recognition in a smart toilet provided by an embodiment of the present invention; Figure 3A This is a schematic diagram of the heat source outline of a human standing posture provided in an embodiment of the present invention; Figure 3B This is a schematic diagram of the heat source outline of a human sitting posture provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of a human posture recognition device for a smart toilet provided in an embodiment of the present invention. Detailed Implementation

[0019] The technical solutions in the embodiments of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Figure 1 This is a flowchart of a human posture recognition method for a smart toilet according to an embodiment of the present invention. Figure 1 As shown, the human posture recognition method for a smart toilet includes: Step S1, collecting temperature array data of the user in the smart toilet area using an infrared array sensor to form a temperature data distribution map; Step S2, performing Gaussian filtering on the temperature data distribution map to remove environmental noise, and performing histogram equalization and normalization processing to form a preprocessed temperature data distribution map; Step S3, extracting feature points from the preprocessed temperature data distribution map, the feature points including: human body contour, thermal centroid, thermal area, and thermal distribution shape; Step S4, determining whether a user exists in the smart toilet area based on the heat source detection results within the field of view of the infrared array sensor, identifying the user as standing or sitting based on the feature points, and outputting corresponding posture output information to control the smart toilet to perform corresponding operations.

[0021] Regarding step S1, temperature array data of the user in the smart toilet area is collected by an infrared array sensor to form a temperature data distribution map.

[0022] The infrared array sensor is an 8×8, 16×16, or 32×32 pixel infrared thermopile sensor, which synchronously acquires temperature field images of the target area through a multi-pixel array (such as 8×8, 16×16, or 32×32). The temperature measurement range acquired in this step is from 10℃ to 50℃, and the sampling frequency is greater than 1Hz.

[0023] In this embodiment, the infrared array sensor is a VMT88 infrared thermopile sensor, which has an 8×8 pixel array and can collect temperature data from 64 pixels at a sampling frequency of 5Hz to form an 8×8 thermal map. This sensor has a wide 60° field of view, power consumption of less than 20mW, a response time of less than 200ms, and an operating temperature range of -20℃ to 85℃, making it suitable for non-contact human body thermal signal acquisition in a smart toilet environment.

[0024] The infrared array sensor is installed behind or to the side of the toilet seat of the smart toilet, and the installation position ensures that the field of view of the infrared array sensor covers the area in front of the toilet.

[0025] In this embodiment, the infrared array sensor is installed behind the toilet seat of the smart toilet, which can cover the user's body and leg area, and realize the detection of imperceptible approach movements.

[0026] The temperature array data is a two-dimensional temperature value matrix arranged according to row and column rules, and each matrix element corresponds to the average temperature value of a sub-region within the sensor's field of view.

[0027] Regarding step S2, the temperature data distribution map is subjected to Gaussian filtering to remove environmental noise, and histogram equalization and normalization are performed to form a preprocessed temperature data distribution map.

[0028] This step is used to preprocess the temperature data distribution map to provide high-quality input data for subsequent heat source detection and attitude recognition.

[0029] In this embodiment, the temperature data distribution map is an 8×8 temperature matrix. A 3×3 two-dimensional Gaussian kernel with a specific standard deviation is used to convolve the temperature data distribution map. Centered on each target pixel, the temperature values ​​within its 3×3 neighborhood are weighted and averaged according to a Gaussian weight distribution. The weights decrease normally as the pixel's distance from the center increases. This operation can suppress high-frequency random noise from the sensor, eliminate isolated pixel temperature jumps, smooth temperature gradients, and avoid scattered noise in subsequent threshold segmentation, while preserving the overall spatial distribution of the temperature field and avoiding excessive blurring of thermal contour edges. This also improves the stability of thermally connected component extraction. After filtering, the temperature matrix still maintains an 8×8 array dimension, with only numerical smoothing correction performed.

[0030] Histogram equalization is performed to enhance image contrast. In this embodiment, each temperature value in the 8×8 temperature matrix is ​​treated as a gray level. First, the distribution histogram of all temperature values ​​in the current frame is calculated, and the cumulative distribution function corresponding to each temperature level is calculated. Then, the original temperature distribution is stretched to a uniform dynamic range through a mapping function. Since the temperature difference between the human body and the background environment in daily scenes is usually only 3℃~5℃, the dynamic range of the original temperature matrix is ​​narrow, and the distinction between the human body thermal outline and the background boundary is low. After histogram equalization, the distribution range of temperature values ​​is linearly stretched, the difference between the high-temperature human body area and the low-temperature background area is amplified, and the edge features of the thermal outline are significantly enhanced. This can effectively reduce the interference of the overall fluctuation of the ambient temperature on target recognition and improve the accuracy of subsequent heat source localization and outline morphology analysis.

[0031] Linear normalization is performed on the equalized temperature matrix, and all temperature values ​​are uniformly mapped to the [0,1] interval, which can eliminate the influence of baseline drift under different ambient temperatures and standardize the data input range.

[0032] Regarding step S3, feature points are extracted from the preprocessed temperature data distribution map. The feature points include: human body outline, thermal centroid, thermal area, and thermal distribution shape.

[0033] The analysis scope is defined by human contour features. When extracting the human contour, a binary segmentation method combining background subtraction and a fixed temperature rise threshold is used: the temperature value of each pixel in the current frame is calculated by subtracting from the pre-stored ambient background temperature benchmark. Pixels with a temperature difference higher than the threshold are marked as foreground pixels of the human heat source, and those lower than the threshold are marked as background pixels. For example, in this embodiment, the binarization result is processed by an eight-neighbor connected component analysis algorithm to extract the highest-temperature connected component as the candidate human body region, excluding scattered small-area high-temperature noise points around it. Furthermore, a boundary tracking algorithm is used to traverse the outermost pixels of this connected component and connect them sequentially in spatial order to form a closed human thermal contour. The contour coordinates correspond one-to-one with the row and column coordinates of the temperature array. This contour can intuitively reflect the outer boundary shape of the human heat source and is a basic feature for distinguishing vertical human figures, folded human figures, and other postures.

[0034] The thermal centroid feature reflects the center of gravity of temperature distribution. When extracting the thermal centroid, a temperature-weighted centroid algorithm is used to calculate the coordinates of the thermal centroid of the human body's heat source, which closely matches the actual location of the core heating area of ​​the human body. For example, with the upper left corner of an 8×8 temperature array as the origin, the column direction as the x-axis and the row direction as the y-axis, the vertical coordinate Cy of the thermal centroid can directly reflect the height of the heat source: when standing, the human torso is vertically extended, the core heating area is higher, and the vertical position of the thermal centroid is higher; when sitting, the human torso is folded, the heating area is lower, and the vertical position of the thermal centroid is lower, which is the core judgment parameter for posture classification.

[0035] Thermal area features refer to the total number of effective foreground pixels contained in the high-temperature connected domain of the human body. The pixel-level thermal area can be obtained by directly counting the number of pixels whose temperature difference exceeds the segmentation threshold. Combining the installation height of the infrared array sensor and the field of view of a single pixel, the pixel area can be further converted into the actual physical area. Thermal area features can be used to eliminate interference from non-human heat sources. For example, if the thermal area of ​​a small, high-temperature object is less than 3 pixels, it can be directly identified as an invalid heat source. It can also help distinguish between human posture and distance; for example, the projected area of ​​a person in a seated position is larger, and the thermal area value is higher than that of a standing person.

[0036] The shape of the heat distribution describes the overall spatial geometry of a high-temperature heat source within a temperature array. In this embodiment, the shape of the heat distribution is quantified based on the geometric parameters of the human body's thermal profile and connected regions, extracting two indicators: aspect ratio and local hotspot distribution. First, the minimum bounding rectangle of the human body's connected regions is calculated to obtain the maximum horizontal width (pixel span in the column direction) and the maximum vertical height (pixel span in the row direction). The ratio of these two values ​​yields the aspect ratio of the heat distribution. Second, the average temperature of each horizontal column is calculated to determine the degree of temperature distribution fluctuation and identify any significantly higher local hotspots than the surrounding areas. In a standing posture, the heat distribution is vertically elongated with a small aspect ratio, and the horizontal temperature distribution width is uniform and stable, with no obvious local hotspots. In a sitting posture, the human torso and legs are folded, resulting in a wide and short heat distribution with a large aspect ratio. Furthermore, the overlapping area of ​​the torso and legs is prone to forming local hotspots. This feature is a key morphological basis for distinguishing between standing and sitting postures.

[0037] Regarding step S4, based on the heat source detection results within the field of view of the infrared array sensor, it is determined whether there is a user in the smart toilet area, and the user is identified as standing or sitting based on the feature points. Corresponding posture output information is output to control the smart toilet to perform corresponding operations.

[0038] Figure 2 This is a control decision logic diagram for human posture recognition in a smart toilet, provided by an embodiment of the present invention. Figure 2 As shown, this step further includes: determining whether a user exists based on the heat source detection results within the field of view of the infrared array sensor. If a heat source exists within the field of view of the infrared array sensor, the user is determined to exist; if the heat source disappears within the field of view of the infrared array sensor, the user is determined to have left. When a user exists, the user's posture is identified as standing or sitting based on the feature points obtained in step S3. Accordingly, the posture output information includes whether the user's posture is standing or sitting. When a user leaves, the posture output information includes user departure information.

[0039] Figure 3A This is a schematic diagram of the heat source contour of a human standing posture according to an embodiment of the present invention. (Reference) Figure 3Aα is the field of view of the infrared array sensor, which is 60° in this embodiment. The heat source profile when the user is standing is shown in the dotted circle. Specifically, when the user is standing, the standing thermal signal characteristics include: the thermal profile is a vertical human shape, the longitudinal coordinate of the thermal centroid is higher than the longitudinal midpoint of the thermal profile, and the lateral temperature distribution width remains relatively stable. Here, "vertical human shape" means that the projection of the human body in the sensor's field of view when standing is a long, narrow shape; "the longitudinal coordinate of the thermal centroid is higher than the longitudinal midpoint of the thermal profile" means that when standing, the core heat-generating areas (torso, head) of the human body are concentrated in the upper half; "the lateral temperature distribution width remains relatively stable" means that when standing, the torso is uniformly thick from top to bottom, and the lateral width of the corresponding high-temperature areas does not change abruptly, remaining generally stable. These standing thermal signal characteristics can be used to determine if the user is standing.

[0040] Figure 3B This is a schematic diagram of the heat source contour of a human sitting posture according to an embodiment of the present invention. (Reference) Figure 3B The heat source contour when the user is standing is shown as a dotted circle. Specifically, when the user is sitting, the sitting posture thermal signal characteristics include: the thermal contour exhibiting a specific folded shape, high-temperature hotspots appearing in the seat area, and the thermal centroid's longitudinal coordinate being lower than the longitudinal midpoint of the thermal contour. Specifically, the specific folded shape of the thermal contour refers to the torso and thighs being folded when the user sits down, resulting in a short, wide, and low-profile overall contour. High-temperature hotspots appearing in the seat area refer to concentrated high-temperature pixels within the seat area with temperatures significantly higher than the surrounding background. The thermal centroid's longitudinal coordinate being lower than the longitudinal midpoint of the thermal contour means that when sitting, the torso folds downwards, the overall heating area sinks, and the thermal centroid falls below the longitudinal midpoint. These sitting posture thermal signal characteristics can be used to determine if the user is standing.

[0041] The posture output information can be a simple status signal (such as "standing" or "sitting") or more detailed data for use by the smart toilet control system.

[0042] The posture output information is used to control the opening or closing of the toilet lid, the opening or closing of the seat, the flushing intensity of the flushing pump, the opening or closing of the disinfection module, and the seat temperature of the seat heater.

[0043] In this embodiment, opening the toilet seat means flipping it up to expose the seat ring and the inner cavity of the toilet; closing the toilet seat means closing it down to completely cover the seat ring; opening the seat ring means flipping it up and away from the toilet seat to expose the inner wall of the toilet, which is commonly used for men to urinate standing up; closing the seat ring means falling down and lying flat on the toilet seat, in a ready state for the user to sit down.

[0044] The intelligent toilet control system controls relevant components to perform delayed closing of the toilet lid and seat based on the user's departure information, and to perform flushing and disinfection operations.

[0045] Continue to refer to Figure 2 When the posture output information includes standing posture information, the intelligent toilet control system controls relevant components to open the toilet seat and the seat according to the standing posture information; when the posture output information includes sitting posture, the intelligent toilet control system controls relevant components to open the toilet seat and close the seat according to the sitting posture information; when the posture output information includes user departure information, the intelligent toilet control system controls relevant components to perform delayed closing of the toilet seat and the seat according to the user departure information, and performs flushing and disinfection operations.

[0046] The above technical solution collects temperature array data of the user in the smart toilet area using an infrared array sensor, and preprocesses it to form a preprocessed temperature data distribution map. It then extracts human contour, thermal centroid, thermal area, and thermal distribution shape feature points from the preprocessed temperature data distribution map. Based on the heat source detection results within the field of view of the infrared array sensor, it determines whether a user is present in the smart toilet area and identifies the user's posture (standing or sitting) based on the feature points, outputting corresponding posture information. This provides precise input for personalized intelligent control of the smart toilet, such as automatic lid opening, lid opening, automatic lid and lid closing when the user is present, automatic lid and lid closing after the user leaves, flushing, and disinfection. It achieves accurate, non-contact recognition of the user's posture in the smart toilet area, effectively solving the problems of false triggering, high energy consumption, and poor user experience existing in smart toilets.

[0047] Based on the same inventive concept, this application also provides a human posture recognition method and recognition device for a smart toilet.

[0048] Figure 4 This is a schematic diagram of a human posture recognition device for a smart toilet according to an embodiment of the present invention. Figure 4 As shown, the human posture recognition device for smart toilets includes: an infrared array sensor 11, a processing module 12, and an output module 13.

[0049] The infrared array sensor 11 is used to collect temperature array data of the user in the smart toilet area in real time and form a temperature data distribution map. The infrared array sensor is an 8×8, 16×16, or 32×32 pixel infrared thermopile sensor. In this embodiment, the infrared array sensor is a VMT88 infrared thermopile sensor, which has an 8×8 pixel array, a wide-angle 60° field of view, power consumption less than 20mW, a response time less than 200ms, and an operating temperature range of -20℃ to 85℃.

[0050] The infrared array sensor 11 is installed behind or to the side of the toilet seat of the smart toilet, and the installation position ensures that the field of view of the infrared array sensor 11 covers the area in front of the toilet. In this embodiment, the temperature data distribution map is an 8×8 temperature matrix.

[0051] The processing module 12 is connected to the infrared array sensor 11 and is used to receive and process the temperature data distribution map and output the user's posture output information. The processing module 12 includes a preprocessing unit 121, a feature extraction unit 122, and a posture classification unit 123.

[0052] The preprocessing unit 121 performs Gaussian filtering on the temperature data distribution map to remove environmental noise, and performs histogram equalization and normalization to form a preprocessed temperature data distribution map. Gaussian filtering can effectively suppress high-frequency random noise from the sensor, eliminate temperature jumps in isolated pixels, smooth temperature gradients, avoid scattered noise in subsequent threshold segmentation, and improve the stability of thermally connected component extraction. After filtering, the temperature matrix still maintains an 8×8 array dimension, and only numerical smoothing correction is performed. Histogram equalization can amplify the difference between the high-temperature human body area and the low-temperature background area, enhance the edge features of the thermal contour, reduce the interference of overall environmental temperature fluctuations on target recognition, and improve the accuracy of subsequent heat source localization and contour morphology analysis. Linear normalization mapping uniformly maps all temperature values ​​to the [0,1] interval, which can eliminate the influence of baseline drift under different environmental temperatures.

[0053] The feature extraction unit 122 is used to extract feature points from the preprocessed temperature data distribution map. The feature points include: human body contour, thermal centroid, thermal area, and thermal distribution shape. The human body contour feature is used to define the analysis range and reflect the outer boundary shape of the human body heat source. It is a basic feature for distinguishing between vertical human figures, folded human figures, and other postures. The thermal centroid feature reflects the center of gravity of the temperature distribution. In a standing posture, the human torso is vertically extended, the core heating area is higher, and the thermal centroid is positioned higher vertically. In a sitting posture, the human torso is folded, the heating area is shifted downward, and the thermal centroid is positioned lower vertically. It is a core parameter for posture classification. The thermal area feature refers to the total number of effective foreground pixels contained in the high-temperature connected domain of the human body. It can be used to exclude interference from non-human body heat sources and to help distinguish between human posture and distance. The thermal distribution shape feature describes the overall spatial geometry of the high-temperature heat source in the temperature array and is a key feature for distinguishing between standing and sitting postures.

[0054] The posture classification unit 123 is used to determine whether a user exists within the smart toilet area based on the heat source detection results within the field of view of the infrared array sensor 11, and to identify whether the user is standing or sitting based on the feature points, outputting corresponding posture output information. When a heat source is present within the field of view of the infrared array sensor 11, it is determined that a user exists; when the heat source disappears from the field of view of the infrared array sensor 11, it is determined that the user has left. When a user exists, the user's posture is identified as standing or sitting based on the feature points. Specifically, when the thermal profile is a vertical human shape, the longitudinal coordinate of the thermal centroid is higher than the longitudinal midpoint of the thermal profile, and the lateral temperature distribution width remains relatively stable, it is determined to be a standing posture; when the thermal profile has a specific folded shape, a high-temperature hot spot appears in the seat area, and the longitudinal coordinate of the thermal centroid is lower than the longitudinal midpoint of the thermal profile, it is determined to be a sitting posture.

[0055] Among these, a vertical human-shaped thermal profile refers to a long, narrow shape projected onto the sensor's field of view when the human body is standing; a thermal center of mass with its vertical coordinate higher than the midpoint of the thermal profile indicates that the core heat-generating area of ​​the human body is concentrated in the upper half when standing; and a relatively stable lateral temperature distribution width means that the human torso is uniformly thick from top to bottom when standing, and the lateral width of the corresponding high-temperature area does not change abruptly, remaining generally stable. A thermal profile with a specific folded shape means that when the human body is seated, the torso and thighs are folded, resulting in a short, wide, and low-profile shape with a short vertical span and a wide horizontal span; high-temperature hotspots appearing in the seat area indicate concentrated high-temperature pixels with temperatures significantly higher than the surrounding background within the seat area; and a thermal center of mass with its vertical coordinate lower than the midpoint of the thermal profile indicates that when sitting, the torso folds downwards, the overall heat-generating area sinks, and the thermal center of mass falls below the midpoint of the vertical coordinate.

[0056] When the user's posture is standing or sitting, the posture output information includes whether the user's posture is standing or sitting. When the user leaves, the posture output information includes user departure information.

[0057] The output module 13 is connected to the processing module 12 and the smart toilet control system 101, and is used to send the posture output information to the smart toilet control system so that the smart toilet control system 101 can perform corresponding control.

[0058] Continue to refer to Figure 4The intelligent toilet control system 101 is connected to the lid motor 102, seat motor 103, flush pump 104, disinfection module 105, and seat heater 106. The intelligent toilet control system 101 controls these components based on the posture output information. For example, it controls the lid motor 102 to open or close the toilet lid, controls the seat motor 103 to open or close the seat, controls the flush pump 104 to adjust the flushing intensity, controls the disinfection module 105 to perform disinfection, and controls the seat heater 106 to adjust the seat temperature.

[0059] The above technical solution uses an infrared array sensor to collect temperature array data of the user in the smart toilet area in real time to form a temperature data distribution map. The processing module receives and processes the temperature data distribution map and outputs the user's posture output information. The output module sends the posture output information to the control system so that the control system can perform corresponding control. This achieves accurate and non-contact recognition of the user's posture in the smart toilet area, effectively solving the problems of false triggering, high energy consumption and poor user experience of existing smart toilets.

[0060] It should be noted that references to "an embodiment," "an embodiment," "an exemplary embodiment," "some embodiments," etc., in the specification indicate that the described embodiments may include specific features, structures, or characteristics, but each embodiment may not necessarily include that specific feature, structure, or characteristic. Furthermore, such phrases do not necessarily refer to the same embodiment. In addition, when a specific feature, structure, or characteristic is described in connection with an embodiment, whether explicitly described or not, implementing such a feature, structure, or characteristic in conjunction with other embodiments is within the knowledge of those skilled in the art.

[0061] It should be noted that the terms "comprising" and "having," and their variations, used in this invention document are intended to cover non-exclusive inclusion. The terms "first," "second," etc., are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence, unless explicitly indicated by the context. It should be understood that such data used interchangeably where appropriate. Furthermore, embodiments and features within embodiments of this invention can be combined with each other unless otherwise specified. In addition, descriptions of well-known components and technologies have been omitted in the above description to avoid unnecessarily obscuring the concepts of this invention. In the various embodiments described above, each embodiment focuses on its differences from other embodiments; similar or identical parts between embodiments can be referred to interchangeably.

[0062] The above description is only a preferred embodiment of the present invention. It should be noted that those skilled in the art can make several improvements and modifications without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for human posture recognition in a smart toilet, characterized in that, include: Temperature data distribution map is generated by collecting temperature array data of the user in the smart toilet area through infrared array sensors. The temperature data distribution map is subjected to Gaussian filtering to remove environmental noise, and histogram equalization and normalization are performed to form a preprocessed temperature data distribution map. Feature points are extracted from the preprocessed temperature data distribution map. The feature points include: human body outline, thermal centroid, thermal area, and thermal distribution shape. Based on the heat source detection results within the field of view of the infrared array sensor, it is determined whether there is a user in the smart toilet area, and the user is identified as standing or sitting based on the feature points, and the corresponding posture output information is output to control the smart toilet to perform the corresponding operation.

2. The method according to claim 1, characterized in that, The step of identifying whether a user is standing or sitting based on the feature points further includes: When the thermal profile is a vertical human figure, the longitudinal coordinate of the thermal centroid is higher than the longitudinal midpoint of the thermal profile, and the width of the lateral temperature distribution remains relatively stable, it is judged to be a standing posture. When the thermal profile has a specific folded shape, a high-temperature hot spot appears in the seat area, and the longitudinal coordinate of the thermal centroid is lower than the longitudinal midpoint of the thermal profile, it is judged to be a sitting posture.

3. The method according to claim 1, characterized in that, In the step of determining whether a user exists in the smart toilet area based on the heat source detection results within the field of view of the infrared array sensor, when the heat source within the field of view of the infrared array sensor disappears, it is determined that the user has left. The posture output information includes user departure information. Based on the user departure information, the smart toilet control system performs a delayed closing of the toilet lid and seat, and performs flushing and disinfection operations.

4. The method according to claim 1, characterized in that, The posture output information is used to control the opening or closing of the toilet lid, the opening or closing of the seat, the flushing intensity of the flushing pump, the opening or closing of the disinfection module, and the seat temperature of the seat heater.

5. A human posture recognition device for a smart toilet, characterized in that, include: Infrared array sensors are used to collect temperature array data of the user in the smart toilet area in real time and generate a temperature data distribution map. The processing module, connected to the infrared array sensor, is used to receive and process the temperature data distribution map and output the user's posture output information. The processing module includes a preprocessing unit, a feature extraction unit, and a posture classification unit. The preprocessing unit is used to perform Gaussian filtering on the temperature data distribution map to remove environmental noise, and to perform histogram equalization and normalization to form a preprocessed temperature data distribution map. The feature extraction unit is used to extract feature points from the preprocessed temperature data distribution map. These feature points include: human body contour, thermal centroid, thermal area, and thermal distribution shape. The posture classification unit is used to determine whether there is a user in the smart toilet area based on the heat source detection results in the field of view of the infrared array sensor, and to identify whether the user is standing or sitting based on the feature points, and output the corresponding posture output information. An output module, connected to the processing module and the smart toilet control system, is used to send the posture output information to the control system so that the control system can perform corresponding control.

6. The apparatus according to claim 5, characterized in that, The infrared array sensor is an 8×8, 16×16, or 32×32 pixel infrared thermopile sensor.

7. The apparatus according to claim 5, characterized in that, The infrared array sensor is installed behind or to the side of the toilet seat of the smart toilet, and the installation position ensures that the field of view of the infrared array sensor covers the area in front of the toilet.

8. The apparatus according to claim 5, characterized in that, When the posture classification unit identifies whether a user is standing or sitting based on the feature points, it specifically includes: When the thermal profile is a vertical human figure, the longitudinal coordinate of the thermal centroid is higher than the longitudinal midpoint of the thermal profile, and the width of the lateral temperature distribution remains relatively stable, it is judged to be a standing posture. When the thermal profile has a specific folded shape, a high-temperature hot spot appears in the seat area, and the longitudinal coordinate of the thermal centroid is lower than the longitudinal midpoint of the thermal profile, it is judged to be a sitting posture.

9. The apparatus according to claim 5, characterized in that, The posture output information is used to control the opening or closing of the toilet lid, the opening or closing of the toilet seat, the flushing intensity of the flushing pump, and the seat temperature of the seat heater.

10. The apparatus according to claim 5, characterized in that, When the heat source disappears within the field of view of the infrared array sensor, the attitude output information output by the attitude classification unit includes user departure information. The output module sends the user departure information to the smart toilet control system, which controls the toilet lid and seat to close with a delay and performs flushing and disinfection operations.

Images