Intelligent hand washing and disinfecting system and method

Abstract

The invention provides an intelligent hand washing and disinfection system and method, and relates to the technical field of hand disinfection, the system comprises a data acquisition module, a cleaning sensing module, a gesture recognition module, a timing module, a control module and an alarm module. Intelligent monitoring and management of the hand washing and disinfecting process of medical staff are achieved, the sanitation and safety level of a ward can be effectively improved, and the risk of cross infection is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of hand disinfection, and in particular to an intelligent hand washing disinfection system and method. Background Art

[0002] Hand hygiene, encompassing key aspects of professional life for medical staff, including handwashing, sanitary hand disinfection, and surgical hand disinfection, is an essential line of defense for ensuring medical and nursing safety. Hospital-acquired infections, a key issue impacting medical quality and nursing safety, remain a top priority for hospital management. Among numerous prevention and control measures, hand hygiene management is the most effective and direct means of improving hospital infection prevention and control.

[0003] Traditional hand washing and disinfection methods mostly rely on manually pressing the disinfectant bottle. However, the manual pressing method is not only cumbersome to operate, but the hand may come into contact with the bottle during the pressing process, posing a risk of secondary contamination.

[0004] The Chinese invention patent with application publication number CN107158421A overcomes the problems of cumbersome operation and secondary contamination of manually pressing the disinfectant bottle to a certain extent, and adopts a fixed automatic sensing disinfector to disinfect the hands. However, the fixed sensing disinfection device lacks effective monitoring of the hand cleaning process, and cannot accurately judge whether the hand washing posture is standard or whether the hand washing time meets the standard. It can easily lead to incomplete hand washing and disinfection of medical staff, making it difficult to achieve the expected cleaning effect and unable to provide reliable infection prevention and control protection. Summary of the Invention

[0005] In order to improve the infection prevention and control level of medical staff, this application provides an intelligent hand washing and disinfection system and method.

[0006] In the first aspect, the present application provides an intelligent hand washing and disinfection system, which adopts the following technical solutions: An intelligent hand washing and disinfection system, comprising: Data collection module, used to collect user gestures and real-time hand washing posture data; Cleaning sensing module, including cleaning unit and identification unit, The cleaning unit includes a bottle body for containing disinfectant, a pump head threadedly connected to the bottle body, and a drive mechanism connected to the pump head; The recognition unit is used to recognize the user's gesture and send the recognition result to the control module; A gesture recognition module, which is in communication with the control module and is used to recognize real-time hand washing gesture data; A timing module is connected to the control module and is used to count the hand washing time; The control module is used to control the opening and closing of the drive mechanism according to the recognition results; it is used to analyze the real-time hand washing posture data and hand washing duration, and issue an alarm instruction when the analysis results do not meet expectations; The alarm module is used to send out an alarm signal according to the alarm instruction.

[0007] This application includes data acquisition, cleaning sensing (including cleaning unit and recognition unit), gesture recognition, timing, control and alarm modules. The data acquisition module collects user gestures and hand washing posture data to provide a basis for subsequent analysis; the cleaning unit in the cleaning sensing module realizes automatic liquid discharge, and the recognition unit cooperates with the acquisition module to transmit gesture information; the gesture recognition module accurately analyzes the hand washing posture; the timing module records the hand washing time; the control module serves as the core to coordinate the work of each module, control the liquid discharge and analyze the data to issue an alarm; the alarm module converts the alarm instruction into an alarm signal that the user can perceive. The modules cooperate with each other to realize hand washing assistance and standard reminders. When the user does not meet the requirements of a certain link, the control module will issue an alarm signal to remind the user to wash hands and disinfect in a standardized manner, thereby improving the infection prevention and control level of medical staff.

[0008] Optionally, the data acquisition module further includes: A TOF sensor is communicatively connected to the control module, and the TOF sensor is used to detect the contour data of the user and send the contour data to the control module.

[0009] Optionally, the system further comprises: The temperature sensor is used to measure the water flow temperature and send the water flow temperature to the control module; Turbidity sensor, used for measuring water turbidity and sending the water turbidity to the control module; The heater is used to heat the water flow.

[0010] In a second aspect, the present application provides an intelligent hand washing and disinfection method, which adopts the following technical solutions: An intelligent hand washing and disinfection method, comprising: The data acquisition module collects user gestures, the recognition unit recognizes the user gestures, obtains the recognition results, and sends the recognition results to the control module. The control module determines whether the recognition results meet expectations. If so, the control module controls the driving mechanism to open, and the driving mechanism drives the pump head to output the disinfectant. After the preset volume of disinfectant is output, the control module controls the driving mechanism to close; If not, no action will be taken; The data acquisition module collects real-time hand washing posture data, and the timing module starts timing when the data acquisition module starts collecting real-time hand washing posture data to obtain the hand washing time; the hand washing time is transmitted to the control module in real time, and the control module determines whether the hand washing time is greater than the preset time threshold. If so, no action will be taken; If not, the control module controls the alarm module to send an alarm signal.

[0011] This application uses a data acquisition module to collect user gestures, and the recognition unit performs recognition and transmits the results to the control module. If it meets expectations, the control module can accurately control the drive mechanism to open, and then drive the pump head to output a preset volume of disinfectant, and then automatically close. This application uses an automated control method to reduce the risk of cross-infection caused by manually pressing the pump head, and improve the hygiene level. At the same time, the output of the preset volume also ensures the relative stability of the amount of disinfectant used each time.

[0012] While the data acquisition module collects real-time hand-washing posture data, the timing module starts timing, and the control module determines whether the hand-washing duration is greater than a preset time threshold. If the threshold is not reached, the control alarm module sends an alarm signal to remind the user to extend the hand-washing time, which helps to standardize the user's hand-washing behavior and ensure that the hand-washing effect meets hygiene standards. The real-time collection of hand-washing posture data and real-time monitoring of hand-washing duration enable this application to keep abreast of the user's hand-washing status and provide immediate feedback, enhancing the user's attention to and self-monitoring of the hand-washing process.

[0013] Optionally, the method further includes: Obtain historical posture data covering the entire hand disinfection process, add labels to the historical posture data, and establish a correspondence between the labels and the hand washing time threshold; build a CNN model, use the labeled historical posture data to train the CNN model to obtain a trained CNN model; input the real-time hand washing posture data into the trained CNN model, and output the classification result; based on the classification result, retrieve the hand washing time threshold, and determine whether the preset time threshold is greater than the hand washing time threshold. If so, no action will be taken; If not, the hand washing time threshold is updated to the preset time threshold.

[0014] This application obtains historical posture data covering the entire process of hand disinfection and adds labels, and then establishes a corresponding relationship between the labels and the hand washing time threshold, so that corresponding hand washing time standards can be formulated for different hand washing posture patterns. Everyone's hand washing posture habits are different. This personalized management method can more accurately measure the hand washing time requirements under different postures, so that the hand washing effect meets the hygiene standards. After retrieving the hand washing time threshold based on the classification results, if the preset time threshold is not greater than the actual required hand washing time threshold, this application will automatically update the hand washing time threshold, so that the hand washing time threshold can be dynamically adjusted according to the actual situation, reducing the problem of insufficient or excessive hand washing time caused by fixed thresholds, and improving the rationality and effectiveness of time management.

[0015] A CNN model was constructed and trained using labeled historical handwashing posture data, enabling it to learn the characteristics of different handwashing postures. When real-time handwashing posture data was fed into the trained CNN model, it quickly and accurately output classification results. The CNN model's powerful image and data processing capabilities improve the accuracy and efficiency of handwashing posture classification, providing a reliable basis for subsequent time determination.

[0016] The CNN model's classification results are used to determine and update the handwashing time threshold, enabling intelligent decision-making. Compared to traditional manual judgment or simple fixed rules, this application's model-based analysis method is more scientific and accurate, and can better adapt to various complex handwashing scenarios.

[0017] Optionally, the method further includes: Obtain the previous n frames of historical handwashing posture data of the real-time handwashing posture data, input the previous n frames of historical handwashing posture data into the trained CNN model in sequence, and output the category of each frame of historical handwashing posture data in sequence, which is recorded as the historical label; Based on the historical tags, historical hand washing posture data identical to the classification result is obtained, recorded as first data, and the first data is sorted in the order of collection time to obtain a first data sequence; historical posture data identical to the classification result is obtained, recorded as second data, and the second data is sorted in the order of collection time to obtain a second data sequence; Calculate the similarity between the first data sequence and the second data sequence, and determine whether the similarity is greater than a preset similarity threshold. If so, no action will be taken; If not, an alarm signal is output.

[0018] Optionally, the method further includes: Based on the timestamp of the second data, determine whether the second data belongs to the same hand washing process, If so, no action will be taken; If not, the second data belonging to the same hand washing process are arranged in the order of collection time to obtain the second data sequence of each hand washing process.

[0019] Optionally, the method further includes: When the water temperature is lower than the preset minimum water temperature threshold, the heater is controlled to heat the water; when the water temperature is higher than the preset maximum water temperature threshold, the heater is controlled to be turned off; When the water turbidity is higher than the preset turbidity threshold, an alarm signal is issued.

[0020] Optionally, the method further includes: The control module builds a 3D model based on the contour data collected by the TOF sensor and calculates the required amount of disinfectant based on the 3D model; A mapping relationship between the required amount of disinfectant and the amount of pump head reduction is established, and the control module adjusts the parameters of the driving mechanism according to the mapping relationship.

[0021] The control module uses the contour data collected by the TOF sensor to construct a 3D model, accurately capturing the three-dimensional shape of the user's hand. Because hands vary in size and shape, calculating disinfectant requirements based on this 3D model allows for personalized calculations tailored to each user's specific hand conditions. This ensures that the amount of disinfectant used each time is more in line with actual needs, minimizing the risk of excessive waste or insufficient disinfectant.

[0022] Accurate calculation of disinfectant demand helps ensure that all parts of the hands are covered with an appropriate amount of disinfectant, improving the comprehensiveness and effectiveness of disinfection, thereby better protecting the user's health and safety. Subsequently, a mapping relationship between the disinfectant demand and the pump head downward adjustment amount is constructed, so that the control module can accurately adjust the parameters of the drive mechanism according to the calculated disinfectant demand, thereby controlling the downward adjustment amount of the pump head. This application adopts a precise control method to match the amount of liquid discharged each time with the actual demand, thereby improving the accuracy and stability of liquid discharge control.

[0023] Optionally, after the cleaning unit outputs the disinfectant, ultraviolet light is used to disinfect the pump head.

[0024] Optionally, the method further includes: Build electronic files for each medical staff separately; Through image recognition algorithm, it is determined whether the current handwashing person is a medical staff. If yes, the pass rate of this hand washing process is calculated based on the recognition result and hand washing time, and the pass rate is stored in the electronic file of the current hand washing person; If not, no action will be taken; Regularly check the electronic files of each medical staff. If the pass rate is continuously lower than the preset threshold, send an alarm signal to the medical staff whose pass rate is continuously lower than the preset threshold, and increase training on standard hand washing posture for the medical staff whose pass rate is continuously lower than the preset threshold.

[0025] This application establishes an electronic file for each medical staff member, and then uses an image recognition algorithm to determine whether the current handwashing person is a medical staff member. If so, the pass rate of the current handwashing person's handwashing process is calculated and stored in the electronic file of the medical staff member; otherwise, no action is taken. This application also helps reduce the risk of hospital infection by regularly checking electronic files to promptly identify medical staff whose handwashing pass rate is continuously below a preset threshold. By adopting the above-mentioned regular verification and intervention mechanism, this application encourages medical staff to continuously improve their handwashing behavior, forming a virtuous cycle of continuous improvement and improving the infection control quality of the entire hospital.

[0026] Optionally, the hand washing pass rate is calculated as follows: ; Where A is the qualified rate; n is the total number of labels; Indicates that the recognition result corresponding to the i-th label meets expectations; is the hand washing time of the i-th label; is the preset time threshold of the i-th tag.

[0027] In summary, this application includes at least one of the following beneficial technical effects: 1. This application includes data acquisition, cleaning sensing (including cleaning unit and recognition unit), gesture recognition, timing, control and alarm modules. The data acquisition module collects user gestures and hand washing posture data to provide a basis for subsequent analysis; the cleaning unit in the cleaning sensing module realizes automatic liquid discharge, and the recognition unit cooperates with the acquisition module to transmit gesture information; the gesture recognition module accurately analyzes the hand washing posture; the timing module records the hand washing time; the control module serves as the core to coordinate the work of each module, control the liquid discharge and analyze the data to issue an alarm; the alarm module converts the alarm instruction into an alarm signal that the user can perceive. The modules cooperate with each other to realize hand washing assistance and standard reminders. When the user does not meet the requirements of a certain link, the control module will issue an alarm signal to remind the user to wash hands and disinfect in a standardized manner, thereby improving the infection prevention and control level of medical staff.

[0028] 2. The control module uses the contour data collected by the TOF sensor to construct a 3D model, accurately capturing the three-dimensional shape of the user's hand. Because different users have varying hand sizes and shapes, calculating disinfectant requirements based on this 3D model allows for personalized calculations tailored to each user's specific hand conditions. This ensures that the amount of disinfectant used each time is more in line with actual needs, minimizing the risk of excessive disinfectant waste or insufficient disinfection.

[0029] 3. This application establishes an electronic file for each medical staff member and uses an image recognition algorithm to determine whether the current handwashing person is a medical staff member. If so, the pass rate of the current handwashing person's handwashing process is calculated and stored in the electronic file of the medical staff member; otherwise, no action is taken. This application also helps reduce the risk of hospital infection by regularly checking electronic files to promptly identify medical staff whose handwashing pass rate is continuously below the preset threshold. By adopting the above-mentioned regular verification and intervention mechanism, this application encourages medical staff to continuously improve their handwashing behavior, forming a virtuous cycle of continuous improvement and improving the infection control quality of the entire hospital. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 This is a schematic structural diagram of Example 1 of the present application; Figure 2 This is a flow chart of the method of Example 2 of the present application. DETAILED DESCRIPTION

[0031] The following combination Figure 1 and Figure 2 This application is described in further detail.

[0032] Example 1: This example discloses an intelligent hand washing and disinfection system. Figure 1 The system includes: a data acquisition module, a cleaning sensing module, a gesture recognition module, a timing module, a control module and an alarm module.

[0033] The data acquisition module can collect user gestures and real-time hand washing posture data through a depth camera or a normal camera.

[0034] The data acquisition module also includes a time-of-flight (TOF) sensor in communication with the control module. The TOF sensor is used to detect the user's profile data and transmit it to the control module. The TOF (Time of Flight) sensor calculates the distance between the sensor and the object by emitting a light pulse and measuring the time it takes for the light pulse to be reflected back from the object, thereby detecting the user's profile data.

[0035] The cleaning sensing module includes a cleaning unit and an identification unit. The cleaning sensing module is responsible for spraying disinfectant and recognizing user gestures, so that the disinfection operation can be carried out accurately and timely.

[0036] The cleaning unit comprises: a bottle body, a pump head and a driving mechanism.

[0037] The bottle body is used to contain disinfectant and is a storage container for the disinfectant.

[0038] The pump head is threadedly connected to the bottle body, and through the mechanical structure of the pump head, its own suction tube can draw out the disinfectant in the bottle. The suction tube can be extended into the bottle body and contact the bottom of the bottle.

[0039] The drive mechanism is connected to the pump head. Under the control of the control module, the drive mechanism can drive the pump head to operate, thereby achieving automatic spraying of the disinfectant. For example, the drive mechanism can use a small motor, and the rotation of the motor drives the piston of the pump head to move, thereby squeezing out the disinfectant.

[0040] The recognition unit is used to identify user gestures. When a user places their hand near the handwashing area, the unit detects the hand movement through an infrared sensor and sends the recognition result to the control module. The recognition result is transmitted to the control module in the form of an electrical signal. The control module then decides whether to activate the drive mechanism to spray disinfectant based on the recognition result.

[0041] The gesture recognition module, in communication with the control module, identifies real-time handwashing gesture data. It can recognize various hand gestures used during handwashing and determine whether the user is following the correct handwashing steps. The gesture recognition module then sends the identified real-time handwashing gesture data to the control module, which analyzes and determines the data based on pre-set handwashing gesture standards.

[0042] The gesture recognition module can identify various hand gestures during handwashing, including those based on image recognition or sensor data. For example, cameras installed around the handwashing area capture images of the user's hands, then use image processing algorithms to analyze and identify gestures. Alternatively, multiple sensors can collect hand motion data in different directions, and algorithms can convert this data into gesture information.

[0043] The timing module, in communication with the control module, measures handwashing duration. Correct handwashing duration is crucial for effective handwashing. According to hygiene standards, handwashing requires a specific duration to effectively remove pathogens from the hands. The timing module starts counting when the user begins handwashing, continuously records the time throughout the handwashing process, and transmits this data to the control module in real time. Based on this data, the control module can determine whether the user's handwashing process meets the required timeframe.

[0044] The control module controls the opening and closing of the drive mechanism based on the recognition results sent by the recognition unit. When the recognition unit detects that the user intends to obtain disinfectant, the control module sends a start signal to the drive mechanism, driving the pump head to spray the disinfectant. After spraying a preset volume of disinfectant, the control module sends a stop signal to the drive mechanism to stop spraying the disinfectant.

[0045] The control module analyzes real-time handwashing posture data and duration, comparing it with preset correct handwashing posture standards to determine whether the user's handwashing posture is correct. It also compares the duration of the current handwashing session with the prescribed time to determine whether the required duration is met. If the analysis results indicate that the user's handwashing posture is incorrect or the handwashing duration does not meet the expected time, the control module will issue an alarm command, notifying the alarm module to send an alarm signal, reminding the user to correct their handwashing behavior promptly.

[0046] The alarm module is used to send out an alarm signal according to the alarm instruction.

[0047] A temperature sensor is used to measure the water flow temperature and send the water flow temperature to the control module.

[0048] Turbidity sensor used to measure the turbidity of water flow and send the water flow turbidity to the control module.

[0049] The heater is used to heat the water flow.

[0050] This embodiment realizes intelligent monitoring and management of the hand washing and disinfection process of medical staff through the collaborative work of various modules, which can effectively improve the health and safety level of medical staff and reduce the risk of cross infection.

[0051] Example 2: Reference Figure 2 This embodiment provides an intelligent hand washing and disinfection method, which is applicable to the system described in Example 1 and includes: S1 recognition: First, the data acquisition module collects pictures containing user gestures. Then, the recognition unit extracts key points in the pictures containing user gestures to recognize the user gestures, obtains recognition results, and sends the recognition results to the control module.

[0052] The control module matches the key point with the predefined gesture template to obtain a matching result. Then, the control module determines whether the recognition result meets the expectations based on the matching result. That is, if the key point can be successfully matched with the gesture template, it means that the recognition result meets the expectations. Otherwise, it means that the recognition result does not meet the expectations. If the recognition result is as expected, the control module sends an opening instruction to the driving mechanism to control the driving mechanism to open, and the driving mechanism drives the pump head to output disinfectant. After outputting the preset volume of disinfectant, the control module sends a closing instruction to the driving mechanism to control the driving mechanism to close.

[0053] If the recognition result does not meet expectations, no processing will be performed.

[0054] S2 duration judgment: the data acquisition module collects real-time hand washing posture data, and the timing module starts timing when the data acquisition module starts to collect real-time hand washing posture data to obtain the hand washing duration.

[0055] Historical posture data covering the entire hand disinfection process is obtained and labeled. The entire hand disinfection process includes applying disinfectant, rubbing the hands (including fingers, palms, backs of hands, wrists, etc.), and rinsing. The labels include the steps of applying disinfectant, rubbing the hands (including fingers, palms, backs of hands, wrists, etc.), and rinsing. The historical posture data described in this embodiment refers to historical posture data that complies with disinfection standards.

[0056] Determine the time thresholds required for different stages and postures of hand disinfection based on relevant hygiene standards and specifications. Hygiene standards stipulate that the application time for disinfectant should be no less than 5 seconds, and the time for rubbing hands should be no less than 15 seconds. Establish a correspondence between these labels and handwashing time thresholds.

[0057] Construct a CNN model, which consists of an input layer, a convolutional layer, a pooling layer, a fully connected layer, and an output layer. The input layer receives historical posture data; the convolutional layer extracts local features of the data using convolution kernels; the pooling layer downsamples the output of the convolutional layer to reduce data dimensionality and computational complexity; the fully connected layer integrates and classifies the extracted features; and the output layer outputs the classification result, which is the category of the hand posture in the historical posture data.

[0058] Subsequently, the CNN model is trained using the labeled historical posture data to obtain a trained CNN model. The real-time hand washing posture data is input into the trained CNN model to output a classification result (i.e., one of all labels). Based on the classification result, the hand washing time threshold is retrieved to determine whether the preset time threshold is greater than the hand washing time threshold. If so, no action will be taken; If not, the hand washing time threshold is updated to the preset time threshold.

[0059] The hand washing time is transmitted to the control module in real time, and the control module determines whether the hand washing time is greater than the preset time threshold. If so, no processing is performed and then S3 posture judgment is executed.

[0060] If not, the control module controls the alarm module to send an alarm signal.

[0061] S3 posture judgment, obtain the historical hand washing posture data of the previous n frames of real-time hand washing posture data. The historical hand washing posture data can reflect the changes in the user's posture over a period of time during the hand washing process. This embodiment obtains all historical hand washing posture data of the user before the real-time hand washing posture data. In other embodiments, other amounts of historical hand washing posture data can also be obtained.

[0062] The first n frames of historical handwashing posture data are sequentially fed into the trained CNN model, and the category of each frame of historical handwashing posture data is output in sequence, recorded as a historical label. The historical label can be applying disinfectant, rubbing hands (including fingers, palms, backs of hands, wrists, etc.), rinsing, etc.

[0063] Based on the historical labels, historical hand washing posture data identical to the classification results is obtained and recorded as the first data. The first data is sorted in chronological order of collection time to obtain a first data sequence. Historical posture data identical to the classification results is obtained and recorded as the second data.

[0064] Based on the timestamp of the second data, determine whether the second data belongs to the same hand washing process, If the second data belongs to the same hand washing process, no processing is performed.

[0065] If the second data do not belong to the same hand washing process, the second data belonging to the same hand washing process are arranged in chronological order according to the collection time to obtain a second data sequence for each hand washing process.

[0066] Whether the second data belongs to the same hand washing process can be determined based on the continuity of the timestamps of the second data. When the timestamps are discontinuous and the interval time is greater than a certain time length (for example, 10 minutes), it is considered that they do not belong to the same hand washing process.

[0067] A similarity algorithm is used to calculate the similarity between the first data sequence and each second data sequence, and it is determined whether there is a similarity greater than a preset similarity threshold. If so, no action will be taken.

[0068] If not, an alarm signal is output.

[0069] By adopting the above technical solution, this embodiment can realize the recognition of user gestures, monitoring of hand washing time and standardized inspection of hand washing posture, thereby improving the level of hygiene and safety and reducing the risk of cross infection.

[0070] In other embodiments, the method further comprises: The TOF sensor collects contour data and converts it into point cloud data. The control module then constructs a 3D model based on the point cloud data. If multiple TOF sensors are present, this embodiment also utilizes an ICP algorithm or other method to register the point cloud data corresponding to the multiple TOF sensors. The control module then constructs a 3D model based on the registered point cloud data.

[0071] To simplify calculations, this embodiment also simplifies the constructed 3D model by extracting key hand features, such as finger length and thickness, palm size, etc., and approximating the 3D model to a combination of a series of simple geometric shapes (such as cylinders, spheres, and cuboids).

[0072] Using the surface area formula for geometric shapes, calculate the surface area of ​​each simple geometric shape separately and then add them together to obtain the approximate surface area of ​​the hand. Subsequently, determine the amount of disinfectant required per unit area of ​​the hand based on relevant hygiene standards and experimental data. Finally, multiply the surface area of ​​the hand by the amount of disinfectant required per unit area to obtain the total disinfectant requirement.

[0073] In other embodiments, the amount of disinfectant required per unit area may also be determined based on multiple tests and disinfection effects.

[0074] A mapping relationship between the demand for disinfectant and the downward adjustment of the pump head is constructed, and the control module adjusts the parameters of the driving mechanism according to the mapping relationship. The downward adjustment of the pump head refers to the distance that the piston or diaphragm moves downward each time the pump head works, which determines the volume of disinfectant output each time by affecting the volume change of the pump chamber. In this embodiment, the downward adjustment of the pump head is determined by converting the demand for disinfectant into the volume change of the pump chamber. Subsequently, the parameter value that the driving mechanism should adjust is calculated based on the corresponding relationship between the downward adjustment of the pump head and the parameters of the driving mechanism (which can be obtained through experiments or theoretical calculations). The control module sends the calculated parameter values ​​to the driving mechanism, and the driving mechanism adjusts its own operating state according to these parameter values ​​so that the pump head reaches the corresponding downward adjustment amount, thereby achieving accurate disinfectant output.

[0075] After the cleaning unit outputs the disinfectant, ultraviolet light is used to disinfect the pump head.

[0076] Example 3: This example differs from Example 2 in that the method further comprises: The control module continuously receives temperature signals from the temperature sensor and compares them with a preset minimum or maximum water temperature threshold. When the water temperature falls below the minimum threshold, the heater is activated to heat the water. When the water temperature rises above the maximum threshold, the heater is deactivated.

[0077] In other embodiments, the control module may also use a closed-loop control algorithm (such as a PID control algorithm) to adjust the power of the heater according to the temperature signal fed back by the temperature sensor, so that the water temperature gradually increases and stabilizes within an appropriate range.

[0078] When the water turbidity is higher than the preset turbidity threshold, an alarm signal is issued.

[0079] The above are all preferred embodiments of the present application, and are not intended to limit the scope of protection of the present application. Therefore, any equivalent changes made based on the structure, shape, and principle of the present application should be included in the scope of protection of the present application.

Claims

1. An intelligent hand washing and disinfection system, characterized in that: include: Data collection module, used to collect user gestures and real-time hand washing posture data; Cleaning sensing module, including cleaning unit and identification unit, The cleaning unit includes a bottle body for containing disinfectant, a pump head threadedly connected to the bottle body, and a drive mechanism connected to the pump head; The recognition unit is used to recognize the user's gesture and send the recognition result to the control module; A gesture recognition module, which is in communication with the control module and is used to recognize real-time hand washing gesture data; A timing module is connected to the control module and is used to count the hand washing time; A control module, used to control the opening and closing of the driving mechanism according to the recognition result; Used to analyze real-time hand washing posture data and hand washing duration, and issue an alarm command when the analysis results do not meet expectations; The alarm module is used to send out an alarm signal according to the alarm instruction.

2. The intelligent hand washing and disinfection system according to claim 1, characterized in that: The data acquisition module also includes: A TOF sensor is communicatively connected to the control module, and the TOF sensor is used to detect the contour data of the user and send the contour data to the control module.

3. The intelligent hand washing and disinfection system according to claim 1 or 2, characterized in that: The system further comprises: The temperature sensor is used to measure the water flow temperature and send the water flow temperature to the control module; Turbidity sensor, used for measuring water turbidity and sending the water turbidity to the control module; The heater is used to heat the water flow.

4. An intelligent hand washing and disinfection method, the method being applicable to the system according to any one of claims 1 to 3, characterized in that: include: The data acquisition module collects user gestures, the recognition unit recognizes the user gestures, obtains the recognition results, and sends the recognition results to the control module. The control module determines whether the recognition results meet expectations. If so, the control module controls the driving mechanism to open, and the driving mechanism drives the pump head to output the disinfectant. After the preset volume of disinfectant is output, the control module controls the driving mechanism to close; If not, no action will be taken; The data acquisition module collects real-time hand washing posture data, and the timing module starts timing when the data acquisition module starts collecting real-time hand washing posture data to obtain the hand washing time; the hand washing time is transmitted to the control module in real time, and the control module determines whether the hand washing time is greater than the preset time threshold. If so, no action will be taken; If not, the control module controls the alarm module to send an alarm signal.

5. The intelligent hand washing and disinfection method according to claim 4, characterized in that: The method further comprises: Obtain historical posture data covering the entire hand disinfection process, add labels to the historical posture data, and establish a correspondence between the labels and the hand washing time threshold; build a CNN model, use the labeled historical posture data to train the CNN model to obtain a trained CNN model; input the real-time hand washing posture data into the trained CNN model, and output the classification result; based on the classification result, retrieve the hand washing time threshold, and determine whether the preset time threshold is greater than the hand washing time threshold. If so, no action will be taken; If not, the hand washing time threshold is updated to the preset time threshold.

6. The intelligent hand washing and disinfection method according to claim 5, characterized in that: The method further comprises: Obtain the previous n frames of historical handwashing posture data of the real-time handwashing posture data, input the previous n frames of historical handwashing posture data into the trained CNN model in sequence, and output the category of each frame of historical handwashing posture data in sequence, which is recorded as the historical label; Based on the historical tags, historical hand washing posture data identical to the classification result is obtained, recorded as first data, and the first data is sorted in the order of collection time to obtain a first data sequence; historical posture data identical to the classification result is obtained, recorded as second data, and the second data is sorted in the order of collection time to obtain a second data sequence; Calculate the similarity between the first data sequence and the second data sequence, and determine whether the similarity is greater than a preset similarity threshold. If so, no action will be taken; If not, an alarm signal is output.

7. The intelligent hand washing and disinfection method according to claim 6, characterized in that: The method further comprises: Based on the timestamp of the second data, determine whether the second data belongs to the same hand washing process, If so, no action will be taken; If not, the second data belonging to the same hand washing process are arranged in the order of collection time to obtain the second data sequence of each hand washing process.

8. The intelligent hand washing and disinfection method according to claim 4, characterized in that: The method further comprises: When the water temperature is lower than the preset minimum water temperature threshold, the heater is controlled to heat the water; when the water temperature is higher than the preset maximum water temperature threshold, the heater is controlled to be turned off; When the water turbidity is higher than the preset turbidity threshold, an alarm signal is issued.

9. The intelligent hand washing and disinfection method according to claim 4, characterized in that: The method further comprises: The control module builds a 3D model based on the contour data collected by the TOF sensor and calculates the required amount of disinfectant based on the 3D model; A mapping relationship between the required amount of disinfectant and the amount of pump head reduction is established, and the control module adjusts the parameters of the driving mechanism according to the mapping relationship.

10. The intelligent hand washing and disinfection method according to claim 9, characterized in that: After the cleaning unit outputs the disinfectant, ultraviolet light is used to disinfect the pump head.

Images