Water intake detection method and device and intelligent equipment

Abstract

The invention discloses a water intake detection method and device and intelligent equipment. The method comprises the following steps: acquiring a behavior state diagram when a target user has a water drinking behavior; and according to a user portrait model to which the target user belongs and the behavior state diagram, determining the single water intake corresponding to the current water drinking behavior of the target user. According to the method and the device provided by the embodiment of the invention, the user portrait model related to the water intake is constructed based on the user portrait, the user portrait model is used for predicting the water intake of the current user, and whether the current water intake of the user meets the preset requirement or not can be judged based on the prediction result, so that the water intake of the user is automatically identified, and the user experience is improved. The defects of low efficiency and low intelligence caused by manual input of the user are avoided, and convenience is provided for the life of the user.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent electronic devices, and in particular to a method and device for detecting water intake, and an intelligent device. Background Art

[0002] Daily drinking water has become a habit. With the increasing emphasis on health, it is particularly important for people to identify the amount of water they drink, especially those who have a certain requirement for drinking water due to illness. Smart terminal developers have also begun to develop applications for water intake detection.

[0003] The water intake detection application in the related art generally requires the user to customize the daily water intake in the application, and then input the daily water intake into the application according to the user's own water intake situation. The application then determines whether the user's daily water intake meets the requirements based on the customized daily water intake and the water intake input by the user. Therefore, the water intake detection in this method mainly relies on the user to manually input information, and the device cannot automatically obtain the user's water intake. In addition, the information input method is difficult to collect due to input errors, user forgetfulness, etc., which leads to the accuracy of water intake detection failing to meet the requirements.

[0004] Therefore, the accuracy of water intake detection needs to be further improved. Summary of the invention

[0005] The purpose of the present invention is to provide a method, device and intelligent equipment for detecting water intake, so as to overcome the defect that water intake cannot be automatically detected.

[0006] The technical solution adopted by the present invention to solve the technical problem is as follows:

[0007] In a first aspect, this embodiment discloses a method for detecting drinking water amount, which includes:

[0008] Obtain the behavior state diagram of the target user when drinking water;

[0009] The single drinking amount corresponding to the current drinking behavior of the target user is determined according to the user portrait model to which the target user belongs and the behavior state diagram.

[0010] Optionally, after the step of determining the single drinking amount corresponding to the current drinking behavior of the target user according to the user portrait model to which the target user belongs and the behavior state diagram, the step further includes:

[0011] The total amount of water drunk by the target user within a preset time period is compared with the preset amount of water drunk to obtain a water drinking comparison result, and the water drinking comparison result is stored or output; wherein the total amount of water drunk is the sum of multiple single amounts of water drunk corresponding to the drinking behavior within the preset time period.

[0012] Optionally, after the step of comparing the total amount of water consumed by the target user within the preset time period with the preset amount of water consumed to obtain a water consumption comparison result, the step further includes:

[0013] Determine whether the current scene is in a do not disturb scene and / or determine whether the current time is in a do not disturb time period;

[0014] If the current scene is in the do not disturb scene and / or the current time is in the do not disturb time period, it is prohibited to output the drinking water reminder information corresponding to the drinking water amount comparison result.

[0015] Optionally, before the step of determining the single drinking amount corresponding to the current drinking behavior of the target user according to the user portrait model to which the target user belongs and the behavior state diagram, the step further includes:

[0016] Get the target user's setting information;

[0017] The user portrait model to which the target user belongs is determined according to the setting information of the target user.

[0018] Optionally, before the step of obtaining the behavior state diagram of the target user when drinking water, the step further includes:

[0019] Calling the camera to identify the user's posture behavior and determine whether the user's posture behavior is a drinking action;

[0020] If the user's posture is a drinking action, the step of obtaining a behavior state diagram of the target user when the drinking action occurs is entered.

[0021] Optionally, the step of calling a camera to identify the user's gesture behavior and determining whether the user's gesture behavior is a drinking action includes:

[0022] Call the camera to identify the user's posture behavior and obtain the target user's hand image information;

[0023] When it is determined that a water cup exists in the hand image information, determining a distance between the water cup and the camera according to the hand image information;

[0024] If the distance between the water cup and the camera gets closer and closer, it is determined that the target user's gesture behavior is a drinking action.

[0025] Optionally, the step of determining the single drinking amount corresponding to the current drinking behavior of the target user according to the user portrait model to which the target user belongs and the behavior state diagram includes:

[0026] Obtain the drinking water information set corresponding to the user portrait model to which the target user belongs;

[0027] Extracting drinking water information corresponding to the current drinking water behavior from the behavior state diagram;

[0028] The current single drinking amount of the target user is determined based on the drinking information contained in the drinking information set and the drinking information corresponding to the current drinking behavior of the target user.

[0029] Optionally, the step of determining the current single drinking amount of the target user according to the drinking information contained in the drinking information set and the drinking information corresponding to the current drinking behavior of the target user includes:

[0030] The drinking information corresponding to the current drinking behavior of the target user is compared with each group of drinking information, the drinking information group with the highest similarity is screened out, and the drinking amount contained in the matched drinking information group is used as the single drinking amount of the current target user.

[0031] In a second aspect, this embodiment further discloses a water intake detection device, which includes:

[0032] An information collection module is used to obtain a behavior state diagram of the target user when he / she drinks water;

[0033] The water drinking amount detection module is used to determine the single water drinking amount corresponding to the current drinking behavior of the target user according to the user portrait model to which the target user belongs and the behavior state diagram.

[0034] In the third aspect, this embodiment also provides a smart device, which is used to receive a behavior state diagram transmitted by a smart wearable device, and determine the single water drinking amount corresponding to the current drinking behavior of the target user based on the user portrait model to which the target user belongs and the behavior state diagram; the behavior state diagram is an image of the target user's drinking behavior captured by the smart wearable device using a camera when the target user drinks water.

[0035] Beneficial effects:

[0036] The present embodiment discloses a method, device and intelligent device for detecting water intake, which obtains a behavior state diagram of a target user when drinking water; and determines the single water intake corresponding to the target user's current drinking behavior according to the user portrait model to which the target user belongs and the behavior state diagram. The method provided in the present embodiment constructs a user portrait model related to the user's water intake based on the user portrait, and uses the user portrait model to predict the current user's water intake. Based on the prediction result, it can be determined whether the user's current water intake meets the predetermined requirements, so as to realize automatic identification of the user's water intake, avoid the defects of low efficiency and low intelligence caused by the need for manual input by the user, and provide convenience for the user's life. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a flowchart of the steps of the method for detecting water intake based on user portrait in an embodiment of the present invention;

[0038] Figure 2 A flowchart of a specific application example of the method described in the embodiment of the present invention;

[0039] Figure 3 It is a schematic diagram of the principle of a specific application example method of the method described in the embodiment of the present invention;

[0040] Figure 4 Principle block diagram of the water consumption detection device in the embodiment of the invention;

[0041] Figure 5 4 is a principle block diagram of an intelligent device in an embodiment of the present invention. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0043] The Dietary Guidelines for Chinese Residents states that adults should drink 1500ml to 2000ml of water every day, including both water drunk directly and water contained in food. However, some users forget to replenish water or do not like to drink water, resulting in a daily water intake far lower than the above figure. In order to keep these users’ water intake at the predetermined requirements, it is necessary to design smart devices with water intake recording and water intake reminder functions for such users to help them achieve their goals and help them develop a daily water drinking habit.

[0044] In the related art, users can record their daily water intake goals on an application, and the application will compare the daily water intake recorded by the user with the daily water intake goal to determine whether the daily water intake goal has been achieved. At the same time, the application will also record the user's daily water intake data to allow the user to view the remaining progress towards the goal in real time; the user can also receive a message notification of the unfinished amount of water intake, or that today's water intake has been reached within a customized reminder time range, thereby allowing the user to develop a habit of replenishing water every day. However, the daily water intake recorded in the application in the prior art is manually input by the user, and the application cannot automatically determine the user's water intake based on the user's behavior. The user may not be able to accurately identify the amount of water consumed or the user may record an incorrect amount of water consumed, resulting in inaccurate daily water intake reminder information, which cannot provide the user with a good reminder effect.

[0045] In order to solve the problem in the related art that the accuracy of water drinking reminder is low and the water drinking information needs to be manually input, this embodiment provides a water drinking detection method based on user portrait, which judges the human body behavior state, identifies whether the user makes a drinking action, and then identifies the user type based on the constructed user portrait, and identifies the accurate water drinking amount based on the user portrait model corresponding to the user type, so that the application can automatically obtain the total amount of water drunk by the user on that day, and then provide the user with a water drinking reminder, and supervise the user to develop a good drinking habit.

[0046] The following is a detailed description of a water intake detection method, device and intelligent device provided in this embodiment in conjunction with the accompanying drawings.

[0047] This embodiment discloses a method for detecting drinking water amount. Figure 1 As shown, including:

[0048] Step S1: Obtain a behavior state diagram of a target user when he or she drinks water.

[0049] The method disclosed in this embodiment is applied on a smart device, which may be a smart bracelet, a smart watch, a smart phone, a smart headset or a smart glasses, etc. In this step, the camera of the smart device is used to capture the user's action behavior, and the captured action is judged to identify whether the action is a drinking action. If the action is a drinking action, a behavior status diagram of the current target user is captured.

[0050] In order to accurately identify whether the action initiated by the user is a drinking action, this step includes: calling the camera to implement recognition of the user's posture behavior to determine whether the user's posture behavior is a drinking action; if the user's posture is a drinking action, then entering the step of obtaining at least one behavior state diagram in the process of the current drinking action.

[0051] Specifically, the smart device uses a camera to monitor the target user in real time, obtains the target user's image information, and simultaneously extracts the target user's body morphology features in the image information, and compares the body morphology features with the body morphology features of drinking water to determine whether they match. If they match, it is determined to be a drinking behavior issued by the user. Furthermore, in order to improve the efficiency of recognition, multiple drinking body morphology features can be pre-stored in the system, so that only the extracted body morphology features of the target user and the stored drinking body morphology features have a similarity that exceeds a preset threshold, and then it can be determined as a drinking behavior.

[0052] Furthermore, the behavior state diagram corresponding to the user's drinking behavior can be obtained through the camera of the smart wearable device worn on the user, or the behavior state diagram of the target user can be obtained by using cameras set at different positions to improve the accuracy of recognition.

[0053] Specifically, when determining whether the user's gesture behavior is a drinking action, this step also includes:

[0054] Call the camera to identify the user's posture behavior and obtain the target user's hand image information;

[0055] When it is determined that a water cup exists in the hand image information, determining a distance between the water cup and the camera according to the hand image information;

[0056] If the distance between the water cup and the camera gets closer and closer, it is determined that the target user's gesture behavior is a drinking action.

[0057] In this step, a series of images between the target user and the water cup are captured, and it is determined based on the captured images whether the distance between the user and the water cup is getting closer. If so, it means that the user is close to the water cup and may drink water.

[0058] In one embodiment, if the smart wearable device is an AR glasses, the image captured by the camera in the AR glasses will only include the user's hand image information, which includes the user's hand image and the water cup image. It can be understood that during the user's drinking process, the user's hand holding the water cup and the water cup will get closer and closer to the user's face, that is, the distance between the water cup and the camera will get closer and closer. Therefore, under the same shooting parameters, during the user's drinking process, the user's hand image information and / or the water cup in the hand image information in the captured image will become larger and larger, so that the captured hand image information can be used to determine whether the distance between the water cup and the camera is getting closer and closer. If the distance between the water cup and the camera is getting closer and closer, it is determined that the target user's posture behavior is a drinking action.

[0059] In one embodiment, the series of images captured are images in which the target user and the water cup are in the same image. When the target user and the water cup are in the same 2D image, it is determined whether the distance between the target user and the water cup is getting smaller in the images captured in consecutive time periods. This can be used to determine whether the distance between the target user and the water cup is getting closer.

[0060] If you want to accurately calculate the distance between the target user and the water cup, you can also calculate the distance between the target user and the water cup in this series of continuous images, and determine whether the distance between the target user and the water cup is getting closer based on the calculated distance value. Specifically, the algorithm for calculating the distance between the target user and the water cup in a single image is: use the length mark of the "internal standard object" pre-set in the image to calculate the distance between the target user and the water cup. The "internal standard object" is an object with a known length when the camera captures the image. The distance between the target user and the water cup is calculated based on the length of the "internal standard object" in the captured image and the length between the target user and the water cup in the image.

[0061] If the target user and the water cup are in different images in a series of captured images, the area where the target user is located can be calibrated first, and then it can be determined whether the water cup enters the calibrated area where the target user is located. If the water cup enters the calibrated area, an image of the target user and the water cup in the same picture is captured, and then the above method of determining whether the distance between the target user and the water cup is getting closer is used to determine the distance change state between the target user and the water cup.

[0062] If the target user uses the camera of the smart wearable device to capture the image information of the user's hand, it can be determined whether the target user is drinking water based on whether the distance between the water cup and the camera is getting closer.

[0063] In another embodiment, when determining whether the user's gesture behavior is a drinking action, this step further includes: obtaining hand motion information of the target user;

[0064] The hand motion trajectory of the target user is inferred based on the sensor data corresponding to the hand motion information; if the hand motion trajectory matches the drinking motion trajectory, it is determined that the target user's gesture behavior is a drinking motion.

[0065] In addition, it is also possible to determine whether drinking water occurs by obtaining hand motion data. In one embodiment, the user may wear a finger clip on his hand, in which a sensor is provided, and the sensor can collect finger motion data. The finger clip is connected to a smart wearable device (such as smart glasses) in communication, and the finger motion data collected in the sensor is transmitted to the smart wearable device. Then, the smart wearable device can determine whether the hand has a drinking water action according to the finger motion data. Specifically, the hand motion trajectory of the target user can be analyzed through the motion data, and the hand motion trajectory of the target user obtained by the analysis is compared with the hand motion trajectory of the preset drinking water action to determine whether they match. If the hand motion trajectory of the target user matches the hand motion trajectory of the preset drinking water action, it is determined that the target user has a drinking water action. If the hand motion trajectory of the target user does not match the hand motion trajectory of the preset drinking water action, it is determined that the target user has not had a drinking water action. It is understandable that after the smart wearable device establishes a communication connection with the finger clip, the finger clip should be worn on the hand of the user holding the water cup. When the user picks up the water cup to drink water, the distance between the finger clip and the smart wearable device will become closer and closer. The hand movement trajectory of the target user can also be obtained by the change in the distance between the finger clip and the smart wearable device.

[0066] Step S2: determining the single drinking amount corresponding to the current drinking behavior of the target user according to the user portrait model to which the target user belongs and the behavior state diagram.

[0067] The amount of water and drinking habits of users of different ages and genders vary greatly. Therefore, in order to accurately identify the amount of water drunk by the target user, before this step, the following steps are also included:

[0068] Build different types of user portrait models.

[0069] Specifically, the steps for setting different types of user portrait models include:

[0070] Collect drinking behavior status information of a large number of different types of users; wherein different types of users include: children, elderly, adult males and adult females; the drinking behavior status information includes: drinking time, drinking duration, shape and capacity of water cup, drinking speed and single drinking amount;

[0071] Different types of user portrait models are constructed according to the drinking behavior status information of different types of users.

[0072] In this step, a large amount of drinking behavior status information of the above-mentioned different types of users is first collected, and based on age, child users, elderly users, adult male users and adult female users are classified respectively to obtain drinking information data corresponding to various types of users.

[0073] The key information extracted from the drinking water information data corresponding to different types of users is used to extract common keywords reflecting the commonality of each type of user, that is, to obtain the labels corresponding to each type. Common keywords include: the amount of water drunk by this type of user in a certain time period, the maximum amount of water drunk by this type of user in a certain time period of a day, the most drinking times of this type of user in a certain time period, the most cup shapes used by this type of user, and the most water capacity of the cup used by this type of user. Take the common keywords corresponding to children's users as an example: the maximum amount of water drunk is at 11:00-12:00 in the morning and 3:00-4:00 in the afternoon, the most number of straw cups used, and the maximum capacity of the cup used is 500ml. Elderly users may drink the most water in the morning and evening, use the most tea cups, and the maximum capacity of the cup is 350ml.

[0074] Based on the extracted labels corresponding to each user type, a user portrait model corresponding to each user type is constructed.

[0075] After constructing different types of user portrait models, this step includes:

[0076] Get the target user's setting information;

[0077] The user portrait model to which the target user belongs is determined according to the setting information of the target user.

[0078] Since the target user will set up the device and enter user information when using the smart device, in this step, the user can be matched with a corresponding type of user portrait model based on the user's age information and gender information obtained from the target user's setting information.

[0079] On the other hand, if the setting information of the target user cannot be obtained or the setting information of the target user does not contain the above-mentioned user profile information, the user's age and gender can also be judged by obtaining the user's photo or full-body image information, and then the user portrait model corresponding to the target user can be determined.

[0080] Furthermore, when the user portrait model corresponding to the current user is determined, the determined user portrait model and the behavioral state diagram obtained by shooting are used to predict the single water drinking amount corresponding to the current drinking behavior of the target user.

[0081] The step of determining the single drinking amount corresponding to the current drinking behavior of the target user according to the user portrait model to which the target user belongs and the behavior state diagram comprises:

[0082] Obtain a drinking water information set corresponding to the user portrait model to which the target user belongs; the drinking water information set includes multiple groups of drinking water information with different information, each group of drinking water information includes: water cup capacity, drinking time, drinking duration, and drinking water volume corresponding to the water cup capacity, drinking time, and drinking duration;

[0083] Drinking water information corresponding to the current drinking water behavior is extracted from the behavior state diagram; the drinking water information includes: drinking water time, water cup capacity and drinking water duration.

[0084] The current drinking amount of the target user is determined based on the drinking information contained in the drinking information set and the drinking information corresponding to the current drinking behavior of the target user.

[0085] Specifically, different types of user portrait types correspond to a drinking information set, which contains the drinking information corresponding to this type of user. The drinking information is constructed based on different common keywords, and different drinking times, different drinking durations, and different water cup capacities all correspond to different single drinking amounts. Therefore, in this step, the drinking information corresponding to the target user is obtained from the behavior state diagram, and the drinking information corresponding to the user is matched with each group of drinking information contained in the drinking information set in the user portrait model corresponding to the user to obtain the drinking information group closest to the target user, and the single drinking amount contained in the drinking information group is used as the drinking amount of the target user this time.

[0086] In order to more accurately predict the drinking amount of the target user, the drinking information set includes multiple groups of drinking information with different information, each group of drinking information includes: water cup capacity, drinking time, drinking duration and single drinking amount;

[0087] The step of determining the current drinking amount of the target user according to the drinking information contained in the drinking information set and the drinking information corresponding to the current drinking behavior of the target user comprises:

[0088] The drinking information corresponding to the current drinking behavior of the target user is compared with each group of drinking information, the drinking information group with the highest similarity is screened out, and the drinking amount contained in the screened drinking information group is used as the single drinking amount of the current target user.

[0089] Specifically, based on multiple behavior state diagrams, the target user's drinking information is extracted, and the target user's drinking time, drinking time and water cup capacity are analyzed to obtain.

[0090] In detail, in this embodiment, the drinking information corresponding to the current user is matched with each group of drinking information contained in the drinking information set to obtain the drinking information group that is closest to the current target user: the drinking information contained in each group of drinking information is sequentially compared with the drinking information contained in the drinking information corresponding to the current target user, and the drinking information group that is most similar to the current drinking behavior of the target user is screened out. Since each drinking information group and the drinking information corresponding to the current drinking of the target user extracted also contain: drinking time, drinking duration and water cup capacity information, the drinking time, drinking duration and water cup capacity are sequentially matched step by step to obtain the most similar drinking information group, and the drinking amount contained in the drinking information group is used as the single drinking amount of the current drinking of the target user.

[0091] First, the drinking time contained in the drinking information group is compared with the drinking time in the corresponding drinking information of the target user, and the drinking information group with the same or similar drinking time is screened out. Then, from the drinking information group with the same or similar drinking time, the drinking information group with the same or similar drinking time and the drinking time contained in the corresponding drinking information of the target user is screened out. After the drinking information group with the same or similar drinking time and drinking time is screened out, the water cup capacity is screened out from the drinking information group with the same or similar drinking time and drinking time, and finally the drinking information group with the same or similar drinking time, drinking time and water cup capacity is obtained. It can be imagined that the higher the similarity of the drinking time, drinking time and water cup capacity in the screened drinking information group, the closer the corresponding drinking water amount in the drinking information group is to the value of the single drinking water amount of the target user.

[0092] When screening, if the drinking time, drinking duration, or cup capacity contained in the drinking information group is different from one or more of the drinking time, drinking duration, or cup capacity contained in the drinking information corresponding to the target user, the matching value can be calculated based on the degree of proximity of the information in the drinking information corresponding to the target user, that is, the closer the value, the higher the matching value, and vice versa. The degree of proximity can be reflected by the difference between the data, and the larger the difference, the lower the degree of proximity.

[0093] For example, if the target user is a child, then the drinking water information set of the user portrait model type is obtained as a child, and a large number of drinking water information groups contained in the drinking water information set are obtained. If the target user drinks at 3 pm this time, the extracted drinking time is 5 seconds, and the water cup capacity is 500 ml, then the drinking water information group containing the drinking time at 3 pm is filtered out in sequence, and the drinking water information group containing the drinking time at 3 pm is filtered out, and the drinking water information group containing the drinking time at 3 pm is filtered out, and the drinking water information group containing the drinking time at 3 pm and the drinking time is 5 seconds is filtered out. The drinking water information group with a capacity of 500 ml is finally filtered out: the drinking water information group that satisfies: the drinking time is 3 pm, the drinking time is 5 seconds, and the water cup capacity is 500 ml. Since the drinking water information group also contains information on the amount of water drunk, the drinking water information contained in the drinking water information group is used as the single drinking amount corresponding to the target user this time. It is understandable that the target user's water cup capacity can be pre-set by the user in the smart wearable device.

[0094] In this step, matching is performed based on the corresponding big data information in the user portrait model, and the drinking water information group with the highest similarity is matched. The single drinking water amount in the most matching drinking water information group is used as the drinking water amount of the current target user, thereby realizing the prediction of the target user's current drinking water amount.

[0095] After the step of determining the single drinking amount corresponding to the current drinking behavior of the target user according to the user portrait model to which the target user belongs and the behavior state diagram, the method of this embodiment further includes:

[0096] The total amount of water drunk by the target user within a preset time period is compared with the preset amount of water drunk to obtain a water drinking comparison result, and the water drinking comparison result is stored or output; wherein the total amount of water drunk is the sum of multiple single amounts of water drunk corresponding to the drinking behavior within the preset time period.

[0097] Since the target user's single drinking amount can be predicted in this embodiment, the single drinking amount within a certain time period can be superimposed to obtain the total drinking amount of multiple single drinking amounts within a certain time period, and then the total drinking amount within the time period can be compared with the preset drinking amount to obtain the drinking amount comparison result. For example, the preset time period can be one day. After obtaining the sum of multiple single drinking amounts within one day, the total drinking amount can be compared with the preset drinking amount for one day to determine whether the drinking amount on that day meets the requirement. If the drinking amount on that day meets the requirement of the preset drinking amount, the target user is reminded that the drinking amount has reached the standard, otherwise the user is reminded that the drinking amount is still insufficient, so as to facilitate the user to replenish water in time.

[0098] In order to prevent the reminder information issued by the application of the smart device from interfering with the user's life, in one embodiment, after comparing the total amount of water consumed by the target user within the preset time period with the preset amount of water consumed, and obtaining the water consumption comparison result, the method further includes:

[0099] Determine whether the current scene is in a do not disturb scene and / or determine whether the current time is in a do not disturb time;

[0100] If the current scene is in the do not disturb scene or the current time is in the do not disturb time, it is prohibited to output the drinking water reminder information corresponding to the drinking water amount comparison result.

[0101] In one embodiment, a smart scene detection function can also be enabled for the camera of the smart device. The smart scene detection function is used to determine whether the scene where the current target user is located is in a do not disturb scene. If the current scene is in a do not disturb scene, the output of drinking water reminder information is prohibited.

[0102] Specifically, the camera can be used to determine whether the current target user is in a non-disturbance scene such as a meeting, class, sleep or rest. If the target user is in the above non-disturbance scene, the drinking water reminder information is prohibited. When the camera monitors that the target user is not in the above non-disturbance scene, the drinking water reminder information is output.

[0103] It is conceivable that the drinking water reminder information can also change the output mode of the voice prompt information to a text reminder to avoid disturbing the target user in the do not disturb scenario.

[0104] In addition, users can also achieve message do not disturb by setting the do not disturb time period. If the current time is within the do not disturb time period, the control will prohibit the output of drinking water reminder information. For example: set the period from 10 pm to 6 am as the message do not disturb time period. When the time is within the above do not disturb time period, the drinking water reminder information will not be output, or the drinking water reminder information will be changed to text message output.

[0105] The method provided in this embodiment can be implemented by a single smart device, or can be implemented by two smart devices based on information interaction.

[0106] When a single smart device is used to perform the above method, the target user's behavior can be monitored using the camera set on the smart device itself. When the user drinks water, a behavior state diagram is captured, and the behavior state diagram is processed and analyzed by a processor to obtain the single drinking amount corresponding to the target user's drinking behavior. In order to facilitate real-time monitoring of the target user's behavior, preferably, the smart device can be a device such as smart glasses or a mobile phone that the user carries with him / her, is equipped with a camera, and the camera facilitates monitoring of the user's behavior when in use.

[0107] Take smart glasses as an example: when a user uses smart glasses, if the drinking water detection function is turned on, the smart glasses will monitor the user's behavior in real time to determine whether drinking water occurs. If drinking water occurs, the user's current drinking behavior state diagram is photographed, and the behavior state diagram is combined with the user's corresponding user portrait model to predict the single drinking water volume of this drinking behavior. The sum of multiple single drinking water volumes predicted within a day is counted, and the user's daily drinking water volume is judged based on the sum of the drinking water volume, and a drinking water reminder notification is issued based on the judgment result.

[0108] When two smart devices are used to implement the steps in the above method in an information exchange manner, one smart device is used to monitor user behavior and take a behavior status graph, and the other smart device focuses on analyzing the behavior status graph and outputs drinking water reminder information based on the analysis results. Therefore, in this usage scenario, based on the fact that the user carries the smart wearable device with him when in use, the smart wearable device can be used to monitor user behavior and take a behavior status graph. Preferably, the smart wearable device can be smart glasses or smart headphones, etc., and the other smart device can preferably be a mobile phone. When the mobile phone receives the behavior status graph taken by the smart wearable device, the processor can be used to process the behavior status graph, and the obtained information can be displayed using the application, and the drinking water reminder information has been issued based on the analysis results.

[0109] The method provided by the present invention is further described below in conjunction with specific application examples of this embodiment.

[0110] Taking the smart device used in the method of this embodiment as smart glasses and the mobile terminal on the application side as a mobile phone as an example, Figure 2 and Figure 3 As shown, the embodiments of the present invention are further described in more detail.

[0111] Step H1, first, the user sets the daily water intake target value in the water intake detection application of the mobile phone, so that the intelligence has a judgment standard for the daily water intake. It can be imagined that the water intake target value can also be achieved through the default setting, that is, when the user is not sure about the amount of water he needs to drink, the application can also set a more suitable water intake target for the user by default based on the basic information entered by the user, such as age, height, occupational identity (athlete, teacher or office white-collar worker, outdoor worker, etc.), to provide convenience for the user.

[0112] Users can also set the drinking water reminder time, whether to record the amount of water consumed, and whether to turn on the smart scene detection function on the application side, so that the application can remind users to drink water based on the user's customized settings.

[0113] Step H2, secondly, the smart glasses monitor the user's behavior in real time, capture the user's current actions through the camera, and determine whether the user has taken a drinking action. If a drinking action occurs, a behavior status diagram of the current user is taken to accurately identify the capacity and shape of the water cup, so as to match the closest drinking amount according to the user portrait model.

[0114] In specific implementation, whether the water content in the water cup changes can also be combined with the hand movement to determine whether drinking water is present. For example, the camera is used to determine whether the user has picked up the water cup to drink water. When the user picks up the water cup, a capacity comparison of the water cup is required. When the user drinks water, the camera needs to determine the difference between the water content in the cup and the top of the cup body, and estimate the water content before drinking water. When the user has finished drinking water, the water content after drinking water is determined to be different from the water content before drinking water to calculate the remaining water content. If the user's hand movement is consistent with the drinking action, and the amount of water in the user's water cup has changed, it can be determined that the current user has initiated a drinking action.

[0115] Step H3, again, the smart glasses transmit the captured user behavior status graph to the mobile phone, and the application of the mobile phone receives the behavior status graph and matches the accurate amount of water intake based on the behavior status graph and the user portrait model corresponding to the user.

[0116] H4. Finally, the mobile phone application calculates whether the current user's water intake meets the standard based on the predicted water intake, and feeds back the calculation result to the user.

[0117] After the current user's water intake is predicted, the application installed in the mobile phone for detecting the user's water intake can calculate the remaining water intake to be consumed on the day based on the set preset water intake and the total water intake predicted for the day, determine whether the water intake on the day reaches the preset water intake based on the remaining water intake, and feed back the calculation result to the user.

[0118] The water drinking method disclosed in this embodiment predicts the water drinking amount of the user's current drinking behavior based on the big data obtained from the user portrait, so as to overcome the defect that the water drinking amount needs to be manually detected, thereby improving the intelligence of the water drinking amount detection.

[0119] In addition, this embodiment also discloses a water intake detection device, such as Figure 4 As shown, including:

[0120] The information collection module 100 is used to obtain the behavior state diagram of the target user when he / she drinks water; its function is as described in step S1.

[0121] The water intake determination module 200 is used to determine the water intake corresponding to the current drinking behavior of the target user according to the user portrait model to which the target user belongs and the behavior state diagram. Its function is as described in step S2.

[0122] In one embodiment, the water intake detection device further includes:

[0123] The comparison module is used to compare the total amount of water drunk by the target user within a preset time period with the preset amount of water drunk, obtain a comparison result of the amount of water drunk, and store or output the comparison result of the amount of water drunk; wherein the total amount of water drunk is the sum of multiple single amounts of water drunk corresponding to the drinking behavior within the preset time period.

[0124] In one embodiment, the water intake detection device further includes:

[0125] A first judgment module is used to judge whether the current scene is in a do not disturb scene and / or whether the current time is in a do not disturb time period;

[0126] The information processing module is used to prohibit the output of the drinking water reminder information corresponding to the drinking water amount comparison result if the current scene is in the do not disturb scene and / or the current time is in the do not disturb time period.

[0127] In one embodiment, the water intake detection device further includes:

[0128] The acquisition module is used to obtain the setting information of the target user;

[0129] The user portrait model to which the target user belongs is determined according to the setting information of the target user.

[0130] In one embodiment, the water intake detection device further includes:

[0131] The second judgment module is used to call the camera to identify the user's posture behavior and determine whether the user's posture behavior is a drinking action;

[0132] The processing module is used to enter the step of obtaining a behavior state diagram of the target user when the target user has a drinking behavior if the user's posture is a drinking action.

[0133] In one embodiment, the second judgment module is also used to call the camera to identify the user's posture behavior and obtain the target user's hand image information; when it is determined that there is a water cup in the hand image information, the distance between the water cup and the camera is determined according to the hand image information; if the distance between the water cup and the camera is getting closer and closer, it is determined that the target user's posture behavior is a drinking action.

[0134] In one embodiment, the water drinking amount determination module 200 is also used to obtain a drinking information set corresponding to the user portrait model to which the target user belongs; extract drinking information corresponding to the current drinking behavior from the behavior state diagram; and determine the current single drinking amount of the target user based on the drinking information contained in the drinking information set and the drinking information corresponding to the current drinking behavior of the target user.

[0135] In one embodiment, the water drinking amount determination module 200 is also used to compare the drinking information corresponding to the current drinking behavior of the target user with each group of drinking information, filter out the drinking information group with the highest similarity, and use the drinking amount contained in the filtered drinking information group as the single drinking amount of the current target user.

[0136] This embodiment also provides a smart device, such as Figure 5 As shown, the smart device 520 is used to receive the behavior status diagram transmitted by the smart wearable device 510, and determine the amount of water consumed corresponding to the current drinking behavior of the target user based on the user portrait model to which the target user belongs and the behavior status diagram; the behavior status diagram is an image of the target user's drinking behavior captured by the smart wearable device 510 using a camera when the target user drinks water.

[0137] Specifically, the smart wearable device 510 is smart glasses, smart headphones or a smart bracelet worn on the target user. The smart wearable device is provided with a camera to monitor the behavior of the target user. When the smart wearable device detects that the target user is drinking water, a behavior status diagram of the target user when drinking water is taken.

[0138] The smart device 520 can be a mobile terminal, such as a mobile phone or a tablet, or a computer, etc. A communication connection is established between the smart wearable device 510 and the smart device 520, and the communication connection can be any one of wifi, Bluetooth or 5G communication. When the smart wearable device 510 captures the behavior state diagram of the target user when drinking water, the behavior state diagram is sent to the smart device 520. The smart device 520 can combine the obtained behavior state diagram with the user portrait model of the target user stored inside it to predict the single water drinking amount corresponding to the drinking behavior of the target user.

[0139] Specifically, the smart device 520 first receives the behavior state diagram sent by the smart wearable device 510, and then uses the user portrait model and behavior state diagram corresponding to the target user determined by the processor to analyze and obtain the water intake corresponding to the current drinking behavior of the target user.

[0140] Before the smart device 520 receives the behavior state diagram, the smart wearable device 510 is also used to judge whether the user has issued a drinking behavior, and collect the behavior state diagram of the user's drinking behavior. In specific applications, the smart wearable device 510 can be a smart bracelet, smart glasses or smart headphones. When the user puts on the smart wearable device, the camera on the smart wearable device is used to obtain the current motion behavior information of the target user, and monitor in real time whether the user has a drinking behavior. For example: when the user wears smart glasses, the smart glasses can obtain the distance between the glasses worn by the user and the water cup through the camera. When the distance between the water cup and the glasses is monitored to be closer and closer, and the distance between the glasses and the water cup is less than the preset distance, it is determined that the user has issued a drinking behavior. Or it is also possible to analyze the motion trajectory corresponding to the hand movement based on the acquired hand motion data, and judge whether the target user is drinking water based on the hand motion trajectory. If drinking water, it triggers the shooting of the behavior state diagram, and detects the user's drinking water based on the behavior state and user portrait model.

[0141] Based on the behavior state diagram, the user can obtain the user's current drinking information, including: the capacity and shape of the water cup used by the target user, the time corresponding to the current drinking, the drinking duration corresponding to the current drinking, etc. The above drinking information is matched with the information contained in each drinking information in the corresponding user portrait model to obtain the set of drinking information with the highest matching value, and the drinking amount contained in the set of drinking information with the highest matching value is used as the drinking amount of the user this time.

[0142] Furthermore, if the target user does not drink water within a preset time, the application may also output a reminder message, which may include the remaining amount of water, the drinking time, and the like.

[0143] In order to prevent the reminder information issued by the application of the smart device from interfering with the user's life, in one embodiment, after comparing the total amount of water consumed by the target user within the preset time period with the preset amount of water consumed, and obtaining the water consumption comparison result, the method further includes:

[0144] Determine whether the current scene is in a do not disturb scene and / or determine whether the current time is in a do not disturb time;

[0145] If the current scene is in the do not disturb scene or the current time is in the do not disturb time, it is prohibited to output the drinking water reminder information corresponding to the drinking water amount comparison result.

[0146] In one embodiment, the smart scene detection function can also be enabled for the camera of the smart wearable device. The smart scene detection function is used to determine whether the scene where the current target user is located is in a do not disturb scene. If the current scene is in a do not disturb scene, the output of drinking water reminder information is prohibited.

[0147] Specifically, the camera can be used to determine whether the current target user is in a non-disturbance scene such as a meeting, class, sleep or rest. If the target user is in the above non-disturbance scene, the drinking water reminder information is prohibited. When the camera monitors that the target user is not in the above non-disturbance scene, the drinking water reminder information is output.

[0148] It is conceivable that the drinking water reminder information can also change the output mode of the voice prompt information to a text reminder to avoid disturbing the target user in the do not disturb scenario.

[0149] In addition, users can also achieve message do not disturb by setting the do not disturb time period. If the current time is within the do not disturb time period, the control will prohibit the output of drinking water reminder information. For example: set the period from 10 pm to 6 am as the message do not disturb time period. When the time is within the above do not disturb time period, the drinking water reminder information will not be output, or the drinking water reminder information will be changed to text message output.

[0150] The water intake identification method, device and intelligent device provided by the present invention predict the user's water intake based on the user portrait model, thereby overcoming the defect that the user needs to open the APP to manually record after drinking water every day. The accuracy of water intake identification is improved through automatic identification and recording by the device. In addition, the water intake identification detection on the device side detects the non-perceptual behavior, thereby better providing detection and calculation of changes in the user's daily water intake, so as to prompt and assist the user to develop a good drinking habit. The method of this embodiment is easy to use, records the user's needs, and the wearable device can be used alone, thereby improving the accuracy of the user's water intake detection.

[0151] Those skilled in the art will readily appreciate other embodiments of the invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art that are not disclosed in this disclosure. The description and examples are to be considered exemplary only, and the true scope and spirit of the present invention are indicated by the following claims.

[0152] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for detecting water intake, characterized in that: include: Obtain the behavior state diagram of the target user when drinking water; The single drinking amount corresponding to the current drinking behavior of the target user is determined according to the user portrait model to which the target user belongs and the behavior state diagram.

2. The method for detecting drinking water amount according to claim 1, characterized in that: After the step of determining the single drinking amount corresponding to the current drinking behavior of the target user according to the user portrait model to which the target user belongs and the behavior state diagram, the method further includes: The total amount of water drunk by the target user within a preset time period is compared with the preset amount of water drunk to obtain a water drinking comparison result, and the water drinking comparison result is stored or output; wherein the total amount of water drunk is the sum of multiple single amounts of water drunk corresponding to the drinking behavior within the preset time period.

3. The method for detecting drinking water amount according to claim 2, characterized in that: After the step of comparing the total amount of water consumed by the target user within the preset time period with the preset amount of water consumed to obtain the water consumption comparison result, the method further includes: Determine whether the current scene is in a do not disturb scene and / or determine whether the current time is in a do not disturb time period; If the current scene is in the do not disturb scene and / or the current time is in the do not disturb time period, it is prohibited to output the drinking water reminder information corresponding to the drinking water amount comparison result.

4. The method for detecting drinking water amount according to claim 1, characterized in that: Before the step of determining the single drinking amount corresponding to the current drinking behavior of the target user according to the user portrait model to which the target user belongs and the behavior state diagram, the method further includes: Get the target user's setting information; The user portrait model to which the target user belongs is determined according to the setting information of the target user.

5. The method for detecting drinking water amount according to claim 1, characterized in that: Before the step of obtaining the behavior state diagram of the target user when drinking water, the method further includes: Calling the camera to identify the user's posture behavior and determine whether the user's posture behavior is a drinking action; If the user's posture is a drinking action, the step of obtaining a behavior state diagram of the target user when the drinking action occurs is entered.

6. The method for detecting drinking water amount according to claim 5, characterized in that: The step of calling the camera to identify the user's gesture behavior and determining whether the user's gesture behavior is a drinking action includes: Call the camera to identify the user's posture behavior and obtain the target user's hand image information; When it is determined that a water cup exists in the hand image information, determining a distance between the water cup and the camera according to the hand image information; If the distance between the water cup and the camera gets closer and closer, it is determined that the target user's gesture behavior is a drinking action.

7. The method for detecting drinking water amount according to claim 1, characterized in that: The step of determining the single drinking amount corresponding to the current drinking behavior of the target user according to the user portrait model to which the target user belongs and the behavior state diagram comprises: Obtain the drinking water information set corresponding to the user portrait model to which the target user belongs; Extracting drinking water information corresponding to the current drinking water behavior from the behavior state diagram; The current single drinking amount of the target user is determined based on the drinking information contained in the drinking information set and the drinking information corresponding to the current drinking behavior of the target user.

8. The method for detecting drinking water amount according to claim 7, characterized in that: The step of determining the current single drinking amount of the target user according to the drinking information contained in the drinking information set and the drinking information corresponding to the current drinking behavior of the target user comprises: The drinking information corresponding to the current drinking behavior of the target user is compared with each group of drinking information, the drinking information group with the highest similarity is screened out, and the drinking amount contained in the screened drinking information group is used as the current single drinking amount of the target user.

9. A water intake detection device, characterized in that: include: An information collection module is used to obtain a behavior state diagram of the target user when he / she drinks water; The water drinking amount detection module is used to determine the single water drinking amount corresponding to the current drinking behavior of the target user according to the user portrait model to which the target user belongs and the behavior state diagram.

10. A smart device, characterized in that: The smart wearable device is used to receive the behavior state graph transmitted by the smart wearable device, and determine the current drinking state of the target user according to the user portrait model to which the target user belongs and the behavior state graph. The water behavior corresponds to the single water intake; The behavior state diagram is when the target user drinks water. The smart wearable device uses a camera to collect images of the target user's drinking behavior.

Images