An eating behavior detection method, a wearable device, a storage medium and a program product

Abstract

The application discloses a kind of eating behavior detection method, wearable device, storage medium and program product, for reducing the constraint of user, it is not required that user must wear wearable device on the wrist of user's eating habitual hand, even if it is worn on the wrist of user's non-eating habitual hand, the eating behavior of user can be detected, and user experience is improved.For example, the wearable device determines the wearing hand of the user currently wearing the wearable device, and determines the eating habitual hand of the user;if the wearing hand is different from the eating habitual hand, a first detection method is used to detect the eating behavior of the user;if the wearing hand is the same as the eating habitual hand, a second detection method is used to detect the eating behavior of the user;the first detection method is different from the second detection method.

Description

Technical Field

[0001] This application relates to the field of terminal technology, and in particular to a method for detecting eating behavior, a wearable device, a storage medium, and a program product. Background Technology

[0002] Scientific research shows that diet has a significant impact on human health. For example, irregular eating habits can disrupt the body's insulin secretion rhythm, leading to various gastrointestinal diseases. Therefore, recording eating behavior is crucial, helping users understand their eating habits and prevent problems before they arise. Over the years, the industry has made several attempts to record eating behavior, with a more advanced method involving wearable devices that detect and record these behaviors. However, current methods for detecting eating behavior using wearable devices have limitations: they require users to wear the device on their dominant eating wrist. For example, if a user is right-handed, the device must be worn on their right wrist to detect eating behavior; if worn on their left wrist, it cannot detect eating behavior. Clearly, this method is too restrictive for users, offering little freedom and a poor user experience. Summary of the Invention

[0003] This application provides a method, wearable device, storage medium, and program product for detecting eating behavior, which can conveniently detect a user's eating behavior through a wearable device and improve the user experience.

[0004] Firstly, a method for detecting eating behavior is provided, applied to wearable devices, such as watches. The method includes: determining the user's current hand on which the wearable device is worn; determining the user's dominant hand for eating; if the hand on which the device is worn is different from the dominant hand for eating, a first detection method is used to detect the user's eating behavior; if the hand on which the device is worn is the same as the dominant hand for eating, a second detection method is used to detect the user's eating behavior; wherein the first detection method and the second detection method are different.

[0005] Therefore, in this embodiment, the user's eating behavior can be detected regardless of whether the hand on which the wearable device is worn is the same as the user's dominant eating hand. In other words, the user is not required to wear the wearable device on their dominant eating wrist. Even if the wearable device is worn on the user's non-dominant eating wrist, the user's eating behavior can still be detected, reducing constraints on the user, giving the user the right to wear the device freely on either hand, and improving the user experience. Moreover, the first detection method is used when the wearing hand is different from the dominant eating hand, and the second detection method is used when the wearing hand is the same as the dominant eating hand. The first and second detection methods are different to ensure the accuracy of eating behavior detection under different circumstances.

[0006] In one possible design, the first detection method for detecting the user's eating behavior includes: detecting the user's eating behavior based on the user's physiological data and / or current scene data.

[0007] Therefore, in this embodiment, the user's eating behavior can be detected even when the hand being worn is different from the hand used for eating. For example, the user's eating behavior can be detected based on the user's physiological data and / or current scene data, ensuring the accuracy of the eating behavior detection as much as possible.

[0008] In one possible design, detecting a user's eating behavior based on their physiological data includes: repeatedly collecting the user's physiological data, which includes at least one of heart rate, energy, blood oxygen, body temperature, blood pressure, respiration, pulse, and blood glucose; determining changes in the user's physiological data based on the repeatedly collected physiological data; and detecting the user's eating behavior based on the changes in the physiological data.

[0009] It is understandable that users' physiological data will change before and after eating in real-world scenarios. Therefore, in this embodiment, when the hand being worn is different from the hand used for eating, the user's eating behavior can be detected based on changes in their physiological data, thus ensuring the accuracy of the eating behavior detection as much as possible.

[0010] In one possible design, detecting the user's eating behavior based on the changes in physiological data includes: if the changes in physiological data are consistent with preset changes, then it is determined that eating behavior exists, wherein the preset changes are used to describe the changes in the user's physiological data before and after eating.

[0011] It should be noted that in real-world scenarios, a user's physiological data changes before and after eating; these changes constitute the preset changes. Therefore, in this embodiment, the wearable device can determine the changes in the user's physiological data and compare them with the preset changes. If they match, it is determined that eating behavior has occurred; otherwise, it is determined that no eating behavior has occurred. This method ensures the accuracy of eating behavior detection as much as possible.

[0012] In one possible design, the method further includes: determining the start and / or end time of the eating behavior based on the changes in the physiological data. Therefore, in this embodiment, the wearable device can determine the timing of a user's eating behavior, such as the start and / or end time of eating, with high accuracy and a good user experience.

[0013] In one possible design, detecting the user's eating behavior based on current scene data includes: determining that the current scene is a eating scene based on at least one of the current geographical location, surrounding environment data, and user's driving path, and then determining that eating behavior exists.

[0014] Therefore, in this embodiment, in scenarios where the wearing hand is different from the eating hand, the wearable device can detect the user's eating behavior based on the current scenario. For example, if the current scenario is an eating scenario, then eating behavior is determined to exist. In this way, even if the wearable device is worn on the user's non-eating wrist, it can still detect the user's eating behavior, reducing constraints on the user, giving the user the right to wear the device freely on either hand, and improving the user experience.

[0015] In one possible design, before determining that eating behavior exists, the method further includes: determining that the amount of movement of the wearing hand is less than a preset amount of movement.

[0016] It should be noted that in real-world scenarios, the movement of the non-eating dominant hand is relatively small during eating. For example, if a user eats with their right hand, the movement of their left hand is minimal. Therefore, in this embodiment, even when the wearing hand is different from the eating dominant hand, if the wearable device determines that the current scenario is an eating scenario and determines that the movement of the wearing hand is minimal, then it determines that eating behavior exists. Thus, in this way, even if the wearable device is worn on the user's non-eating dominant wrist, it can still detect the user's eating behavior with high accuracy.

[0017] In one possible design, the method further includes: determining the start time and / or end time of eating; the start time of eating is the moment when the user enters the eating scene or the moment when the movement of the wearing hand changes from greater than a preset movement amount to less than a preset movement amount; the end time of eating is the moment when the user leaves the eating scene or the moment when the movement of the wearing hand changes from less than a preset movement amount to reach a preset movement amount. Therefore, in this embodiment of the application, the wearable device can also determine the eating time of the user's eating behavior, such as the start time and / or end time of eating, with high accuracy and a good user experience.

[0018] In one possible design, after detecting a user's eating behavior, the method further includes: outputting a first prompt message, the first prompt message being used to prompt the user to check relevant information about the eating behavior, the relevant information including the start time and / or end time of the eating behavior; if a user confirmation operation is received, then confirming the relevant information about the eating behavior; if a user correction operation is received, then correcting the relevant information about the eating behavior.

[0019] Therefore, in this embodiment, the user can verify the eating behavior detected by the wearable device, such as correcting the start and / or end times of the eating behavior. This improves the accuracy of the recorded eating behavior.

[0020] In one possible design, after detecting the user's eating behavior, the method further includes: detecting the user's blood glucose level; and verifying the eating behavior based on the blood glucose level.

[0021] Therefore, in this embodiment, after the wearable device detects eating behavior, it can detect the user's blood glucose level and use the blood glucose level to verify the eating behavior, thereby determining whether the detection of eating behavior is accurate. This improves the accuracy of eating behavior detection.

[0022] In one possible design, after detecting the user's eating behavior, the method further includes displaying an eating behavior identifier, which indicates the presence of eating behavior. Therefore, in this embodiment, the user can know that their eating behavior has been recorded through the eating behavior identifier displayed on the wearable device, resulting in a better user experience.

[0023] In one possible design, the method further includes displaying the time when the eating behavior occurs. Therefore, in this embodiment, the user can also see their eating time on the wearable device, which helps improve the user experience.

[0024] In one possible design, after detecting the user's eating behavior, the method further includes: displaying an eating behavior identifier at a first location, the eating behavior identifier indicating the presence of eating behavior, and the first location indicating the time of occurrence of the eating behavior. Therefore, in this embodiment, the user can know that their eating behavior has been recorded and can know the time of occurrence of their eating behavior through the eating behavior identifier displayed at the first location on the wearable device, which helps improve the user experience.

[0025] In one possible design, the first position is located within a first interface, which includes multiple time scales indicating different times of day. The first position is close to a first time scale, and the time corresponding to the first time scale is the time when the eating behavior occurred. For example, the wearable device is a watch, which can display an eating behavior indicator on its watch face (i.e., the first interface), with the indicator near the first time scale, representing that the eating behavior occurred at the time indicated by the first time scale. In this way, users can know that their eating behavior has been recorded and when it occurred through the eating behavior indicator on the watch face, thus improving the user experience.

[0026] In one possible design, the eating behavior identifier is displayed in a first display mode, which indicates that the eating behavior corresponding to the identifier has not been confirmed by the user. This method reminds the user to confirm the eating behavior, improving the accuracy of the user's eating behavior recorded by the wearable device.

[0027] In one possible design, the method further includes: receiving an operation on the eating behavior identifier; displaying relevant information about the eating behavior corresponding to the eating behavior identifier, the relevant information including the start time and / or end time of the eating behavior; and after receiving a confirmation operation on the relevant information of the eating behavior, displaying the eating behavior identifier in a second display mode, the second display mode indicating that the eating behavior corresponding to the first identifier has been confirmed by the user, and the second display mode being different from the first display mode. Therefore, in this embodiment, the user can confirm the eating behavior by touching the eating behavior identifier. After the user confirms, the eating behavior identifier is displayed in another display mode, indicating that the user has confirmed it. In this way, the user can know which eating behaviors have been confirmed and which have not been confirmed, resulting in a better user experience.

[0028] In one possible design, the method further includes: displaying motion behavior identifiers and / or sleep behavior identifiers, wherein the motion behavior identifiers indicate the presence of motion behavior, and the sleep time identifiers indicate the presence of sleep behavior. Therefore, in this embodiment, users can know that their sleep behavior is being recorded through the sleep behavior identifiers displayed on the wearable device; and they can know that their motion behavior is being recorded through the motion behavior identifiers displayed on the wearable device, resulting in a better user experience.

[0029] In one possible design, the method further includes: when displaying the exercise behavior identifier, also displaying the time of occurrence of the exercise behavior corresponding to the exercise behavior identifier; when displaying the sleep behavior identifier, also displaying the time of occurrence of the sleep behavior corresponding to the sleep behavior identifier. Therefore, in this embodiment, users can also see their exercise time and / or eating time on the wearable device, which helps to improve the user experience.

[0030] In one possible design, the method further includes: displaying a motion behavior identifier at a second location, the motion behavior identifier indicating the presence of motion behavior, the second location indicating the time of occurrence of the motion behavior; and / or, displaying a sleep behavior identifier at a third location, the sleep behavior identifier indicating the presence of sleep behavior, the third location indicating the time of occurrence of the sleep behavior.

[0031] Therefore, in this embodiment, users can know that their movement behavior has been recorded and when it occurred by the movement behavior indicator displayed at the second location on the wearable device. Furthermore, users can also know that their sleep behavior has been recorded and when it occurred by the sleep behavior indicator displayed at the third location on the wearable device, resulting in a better user experience.

[0032] In one possible design, the second and third positions are located within a first interface, which includes multiple time scales indicating different times of day. The second position is near a second time scale, and the time corresponding to the second time scale is the time when the exercise occurred. The third position is near a third time scale, and the time corresponding to the third time scale is the time when the sleep occurred. For example, the wearable device is a watch. The watch can display exercise and / or sleep indicators on the watch face (i.e., the first interface). The exercise indicator is displayed near the second time scale, indicating that the exercise occurred at the time indicated by the second time scale, and the sleep indicator is displayed near the third time scale, indicating that the sleep occurred at the time indicated by the third time scale. In this way, users can know that their exercise and sleep behaviors are being recorded and the times when they occurred through the exercise and sleep indicators on the watch face, which helps improve the user experience.

[0033] In one possible design, determining the user's dominant eating hand includes: displaying a questionnaire interface showing the dominant eating hand, where the questionnaire is used to input the dominant eating hand; and determining the dominant eating hand based on the user's input. Therefore, in this embodiment, the wearable device can determine the user's dominant eating hand through a questionnaire survey with high accuracy.

[0034] In one possible design, determining the user's dominant eating hand includes: after determining the wearing hand, outputting query information to inquire whether the wearing hand is the dominant eating hand; and determining the dominant eating hand based on user input. Therefore, in this embodiment, after the wearable device determines the wearing hand, it can determine the user's dominant eating hand by inquiring whether the wearing hand is the dominant eating hand, which has high accuracy.

[0035] In one possible design, determining the user's dominant hand for eating includes: after determining the wearing hand, if the current scenario is detected as an eating scenario, determining whether the action of the wearing hand is an eating action; if the action of the wearing hand is an eating action, determining the dominant hand for eating is the wearing hand; if the action of the wearing hand is not an eating action, determining the dominant hand for eating is another hand other than the wearing hand. Therefore, in this embodiment of the application, the wearable device can automatically detect the user's dominant hand for eating without user input, improving convenience and providing a better user experience.

[0036] In one possible design, the method further includes: generating statistical results of the user's eating behavior based on the user's eating behavior within a statistical period, the statistical results including at least one of the user's average number of meals, average eating speed, post-meal exercise, and post-meal sleep within the statistical period; and outputting the statistical results of the eating behavior. Therefore, in this embodiment, the wearable device can also output statistical results of the user's eating behavior, allowing the user to know whether their daily eating is healthy, such as whether they have all three meals, resulting in a better user experience.

[0037] In one possible design, a first detection method is used to detect the user's eating behavior, including: using a first detection model to detect the user's eating behavior, the input parameters of the first detection model including: the user's physiological data and / or current scene data, and the output parameters of the first detection model including a first confidence level, which is used to indicate the probability of eating behavior; a second detection method is used to detect the user's eating behavior, including: using a second detection model to detect the user's eating behavior, the input parameters of the second detection model including: the action data of the wearing hand, and the output parameters of the second detection model including a second confidence level, which is used to indicate the probability of eating behavior.

[0038] Therefore, in this embodiment, the wearable device can detect the user's eating behavior through a detection model. Furthermore, the detection model and corresponding input parameters differ depending on whether the wearing hand is different from the user's dominant eating hand, or whether they are the same. This ensures the accuracy of eating behavior detection under different circumstances.

[0039] In one possible design, the method further includes: if the user's eating behavior is detected by the first detection model, then the eating behavior is verified; if it is determined that the eating behavior does not exist, then the first detection model is retrained; or, if the user's eating behavior is detected by the second detection model, then the eating behavior is verified; if it is determined that the eating behavior does not exist, then the second detection model is retrained.

[0040] Therefore, in this embodiment, the wearable device can detect the user's eating behavior through a detection model. If the eating behavior detected by the detection model is incorrect, the detection model can be retrained to improve accuracy.

[0041] Secondly, a wearable device is also provided for performing the method as described in the first aspect above.

[0042] Thirdly, an electronic device is also provided for performing the method as described in the first aspect above. Exemplarily, the electronic device may be a wearable device.

[0043] Fourthly, a computer-readable storage medium is also provided for storing a computer program that, when run on a computer, causes the computer to perform the method provided in the first aspect above.

[0044] Fifthly, a computer program product is also provided, comprising a computer program that, when run on a computer, causes the computer to perform the method provided in the first aspect above.

[0045] In a sixth aspect, a chip or chip system is also provided for performing the method as described in the first aspect above.

[0046] For the technical effects that can be achieved in the second to sixth aspects mentioned above, please refer to the description of the technical effects that can be achieved by the corresponding design scheme in the first aspect mentioned above. This application will not repeat them here. Attached Figure Description

[0047] Figure 1 A schematic diagram of the display interface of a wearable device provided in an embodiment of this application;

[0048] Figure 2 A schematic flowchart of an eating behavior detection method provided in an embodiment of this application;

[0049] Figure 3 Another schematic diagram of the display interface of a wearable device provided in an embodiment of this application;

[0050] Figure 4 This is a schematic diagram of a detection method for wearing a hand, provided in an embodiment of this application.

[0051] Figure 5 Another schematic diagram of the display interface of a wearable device provided in an embodiment of this application;

[0052] Figure 6 A schematic flowchart illustrating a method for detecting the dominant hand for eating, provided in an embodiment of this application;

[0053] Figure 7 Another schematic diagram of the display interface of a wearable device provided in an embodiment of this application;

[0054] Figures 8A to 8B A schematic diagram illustrating user confirmation of the wearing hand and the dominant hand for eating, provided as an embodiment of this application;

[0055] Figure 9A A schematic diagram of an energy change curve provided in an embodiment of this application;

[0056] Figure 9B A schematic diagram of a heart rate change curve provided in an embodiment of this application;

[0057] Figure 9C A schematic diagram of a body temperature change curve provided in an embodiment of this application;

[0058] Figure 9D This is a schematic diagram of a blood glucose change curve provided in an embodiment of this application;

[0059] Figure 10 A schematic diagram of algorithm model 1 provided in an embodiment of this application;

[0060] Figure 11 This is another flowchart illustrating an embodiment of the eating behavior detection method provided in this application;

[0061] Figure 12 A schematic diagram of algorithm module 2 provided in an embodiment of this application;

[0062] Figure 13 This is another flowchart illustrating an embodiment of the method for detecting eating behavior provided in this application;

[0063] Figure 14 A schematic diagram of algorithm model 3 provided in an embodiment of this application;

[0064] Figures 15A to 15G A schematic diagram illustrating user confirmation of eating behavior provided in an embodiment of this application;

[0065] Figures 16 to 18 A schematic diagram illustrating the display of eating behavior indicators on a wearable device according to an embodiment of this application;

[0066] Figure 19 A schematic diagram illustrating the display of motion behavior indicators on a wearable device according to an embodiment of this application;

[0067] Figure 20 A schematic diagram illustrating the display of sleep behavior indicators by a wearable device according to an embodiment of this application;

[0068] Figure 21A schematic diagram illustrating the statistical results of eating behavior provided in an embodiment of this application;

[0069] Figures 22A to 22C A schematic diagram of a user eating profile provided in an embodiment of this application;

[0070] Figure 23 This is another flowchart illustrating an embodiment of the method for detecting eating behavior provided in this application;

[0071] Figure 24 This is a schematic diagram of the structure of a wearable device provided in an embodiment of this application;

[0072] Figure 25 This is another structural schematic diagram of a wearable device provided in an embodiment of this application;

[0073] Figure 26 This is another structural schematic diagram of a wearable device provided in an embodiment of this application. Detailed Implementation

[0074] The eating behavior detection scheme of this application embodiment will be described in detail below with reference to the accompanying drawings.

[0075] The eating detection solution provided in this application can be applied to wearable devices. Exemplary examples include wrist-worn wearable devices, head-worn wearable devices, and clothing-related wearable devices. For example, wrist-worn wearable devices may include watches, bracelets, gloves, wristbands, rings, and necklaces worn on the hand / wrist / arm. Head-worn wearable devices may include helmets, glasses, headbands, earplugs, headphones, earrings, and necklaces worn on the head. Clothing-related devices may include clothes, pants, shoes, belts, buttons, and ties. It should be noted that the various types of wearable devices listed above are merely examples and do not constitute a limitation of this application. Other wearable devices besides those listed above can also use the technical solutions of this application's embodiments without limitation. For ease of understanding, the following embodiments primarily use a watch as an example of a wearable device.

[0076] As described in the background technology description, currently, detecting a user's eating behavior through wearable devices requires the user to wear the device on the wrist of their dominant eating hand. If the wearable device is worn on the wrist of the user's non-dominant eating hand, the user's eating behavior cannot be detected. This method is too restrictive for the user, which can easily cause user resentment and affect the user experience.

[0077] In view of this, embodiments of this application provide a method for detecting eating behavior. In this method, the wearable device can determine the user's current wearing hand and the user's dominant eating hand. If the wearing hand is different from the dominant eating hand, the user's eating behavior is detected using a first detection method. If the wearing hand is the same as the dominant eating hand, the user's eating behavior is detected using a second detection method. The first detection method and the second detection method are different.

[0078] On the one hand, the eating behavior detection scheme provided in this application does not require the user to wear the wearable device on the wrist of the user's dominant eating hand. Even if the wearable device is worn on the wrist of the user's non-dominant eating hand, it can still detect the user's eating behavior, reducing the constraints on the user, giving the user the right to wear it freely on both hands, and improving the user experience.

[0079] On the other hand, the eating behavior detection scheme provided in this application uses a first detection method when the wearing hand is different from the dominant eating hand, and a second detection method when the wearing hand and the dominant eating hand are the same. The first and second detection methods are different. For example, when the wearing hand is the dominant eating hand, eating behavior is detected by analyzing the movements of the wearing hand (i.e., the second detection method); if the wearing hand is not the dominant eating hand, it is difficult to detect eating behavior through the second detection method, i.e., analyzing the movements of the wearing hand, then eating behavior is detected through the user's physiological data and / or current scene data (i.e., the first detection method). In this way, the accuracy of eating behavior detection can be guaranteed when the wearing hand and the dominant eating hand are different.

[0080] The eating behavior detection method provided in this application embodiment can be integrated into a first application of a wearable device. The first application can be a diet-related application (APP), such as a health monitoring / research application, a blood glucose control / research application, a fitness / body shaping application, or a habit / self-discipline development application, etc. For example, a health monitoring / research application could be the Huawei Innovation Research APP. Similarly, a blood glucose control / research application could be the Huawei Blood Glucose Research APP. Etc. Taking fitness / body shaping apps as an example, such as the Huawei Health app, etc. Taking habit / self-discipline development apps as an example, such as memos, notes, etc. These are apps used to record users' daily behavior. It should be noted that the various types of applications listed above are for illustrative purposes only and do not constitute a limitation of this application. Of course, in addition to the types of applications listed above, the first application can also be other types of applications, which will not be listed one by one. Furthermore, the first application can be a system application or a third-party application, without limitation.

[0081] Taking the eating behavior detection method of this application embodiment integrated into the Huawei Innovation Research APP as an example, please refer to [link to relevant documentation]. Figure 1 In (a), the wearable device displays an application list, which includes icons for each application (e.g., an icon for one application may be contained within a circle), including icon 100 for the Huawei Innovation Research app. Upon receiving an action on icon 100, the wearable device launches the Huawei Innovation Research app, for example, displaying... Figure 1 The interface shown in (b) includes options 101 for eating behavior research, 102 for blood glucose health research, and 103 for heart health research. After receiving an operation on option 101 for eating behavior research, the wearable device displays the following... Figure 1 The interface in (c) can be used to interpret and explain research on eating behavior. For example, Figure 1 The interface in (c) includes: an eating behavior research title 104, a content summary 105, and an "Add to Study" button 106. The content summary 105 explains the eating behavior research, for example, it may include text such as "To ensure your health, this function is used to study your daily eating behavior." Optionally, it may also include other forms of content summary such as images or videos, without limitation. The "Add to Study" button 106 is used to start the eating behavior research. That is, after the wearable device detects the operation on the "Add to Study" button 106, it starts the eating behavior detection method provided in this embodiment, for example, by starting the execution of the following... Figure 2 The process is illustrated. It should be noted that, in this embodiment of the application, the content displayed on any interface of the wearable device can be displayed through other devices such as mobile phones and tablets connected to the wearable device.

[0082] Please see Figure 2 This is a flowchart illustrating an embodiment of an eating behavior detection method provided in this application. This method can be applied to wearable devices, such as... Figure 1 The wearable device shown. (e.g.) Figure 2 As shown, the method may include steps 201 to 205.

[0083] Step 201: Detect the hand on which the user is currently wearing the wearable device.

[0084] Optionally, step 201 may include two implementation methods, referred to as method A and method B. Method A involves the user actively inputting the hand to be worn, while method B involves the device automatically detecting the hand being worn.

[0085] Method A: The user actively inputs the hand they are wearing.

[0086] One possible approach is to use a questionnaire to allow users to actively input their hand type. For example, ... Figure 3In (a), the wearable device displays an interface (for a description of this interface, please refer to the preceding text). Figure 1 (c) (not repeated here), this interface includes a "Join Research" button 106. After receiving an operation on the "Join Research" button 106, the wearable device initiates a questionnaire survey, for example, displaying... Figure 3 The questionnaire interface for the wearing hand (b) displays the title "Please enter the wearing hand" (300), along with options 301 (left hand), 302 (right hand), and 303 (uncertain). The user can select the corresponding option. If the wearable device determines that option 301 (left hand) is selected, then the wearing hand is determined to be the left hand. If the wearable device determines that option 302 (right hand) is selected, then the wearing hand is determined to be the right hand. If the wearable device determines that option 303 (uncertain) is selected, then the wearing hand could be either the right or left hand. In this case, the wearable device can use other methods (e.g., method B) to determine the wearing hand.

[0087] Understandably, in practical applications, users may change the hand on which the device is worn, for example, from the left wrist to the right. To ensure the accuracy of the hand recorded by the wearable device, if the device detects that the device has been removed and then put back on, it will display a prompt message to ask the user whether to change the hand. If the device receives a no-change instruction, it is determined that the hand on which the device is worn has not changed; if it receives a change instruction, it is determined that the hand on which the device is worn has changed to the other hand.

[0088] Method B: The device automatically detects the wearer's hand.

[0089] For example, wearable devices can detect the wearing hand based on at least one of the following: the orientation of the watch face and the face image.

[0090] Take, for example, a wearable device that detects the wearer's hand based on the watch face's orientation. Figure 4 In (a), the dial orientation can be understood as the direction from the center point of the dial to the 24 o'clock mark, indicated by a dashed arrow in the diagram. Wearable devices store an absolute coordinate system, which does not change with the device's posture, position, etc., and is a fixed coordinate system. For example... Figure 4 In (b), when the wearable device is worn on the user's left wrist, the angle between the dial orientation (indicated by the dotted arrow) and the X-axis is obtuse. For example... Figure 4 In (c), when the wearable device is worn on the user's right wrist, the angle between the dial orientation (indicated by the dotted arrow) and the X-axis is acute. Therefore, the wearable device can determine whether the wearer's hand is left or right based on the angle between the dial orientation and the X-axis in the absolute coordinate system. If the angle is obtuse, the wearer's hand is determined to be left; if the angle is acute, the wearer's hand is determined to be right.

[0091] Let's take the example of a wearable device detecting the wearing hand based on a facial image. For instance, a wearable device has a camera. When the device is worn on the user's left wrist, the left hand is typically closer to the left side of the face and farther from the right. Therefore, in the facial image captured by the camera, the left side of the face appears larger than the right. Similarly, when the device is worn on the user's right wrist, the right hand is typically closer to the right side of the face and farther from the left. Thus, in the facial image captured by the camera, the right side of the face appears larger than the left. Therefore, the wearable device can determine whether the wearing hand is the left or right hand based on the facial image captured by the camera. For example, if the left side of the face appears larger than the right, the wearing hand is determined to be the left hand; otherwise, it is determined to be the right hand.

[0092] The above lists two methods for wearable devices to automatically detect the wearer's hand. In addition to these two methods, there are other ways to automatically detect the wearer's hand, which will not be listed here.

[0093] In this embodiment, the wearable device can automatically detect the wearing hand using an algorithm model, which can be called a "wearing hand detection model." For example, the input parameters of the wearing hand detection model may include the watch face orientation and / or a face image, and the output parameter is the wearing hand. The principle of the wearing hand detection model detecting the wearing hand based on the watch face orientation and / or face image has been described in Method B above and will not be repeated here. Optionally, after detecting the wearing hand using the wearing hand detection model, the wearable device can display a prompt message to ask the user to confirm whether the wearing hand detection is accurate. For example, if the wearable device detects that the wearing hand is a left hand using the wearing hand detection model, it will display the prompt message "Your wearing hand is detected as a left hand, correct it?" If the user corrects the wearing hand to a right hand, then the wearing hand is confirmed to be a right hand. If the wearable device determines that the model detection result (i.e., left hand) is inconsistent with the user's correction result (i.e., right hand), it can retrain the wearing hand detection model to improve accuracy. Optionally, the wearing hand detection model can be pre-installed in the wearable device, for example, the wearing device is configured with a wearing hand detection model at the factory. Alternatively, the hand detection model can be pre-installed in the application package of the first application. After the wearable device downloads the application package of the first application, the application package includes the hand detection model.

[0094] Considering that wearable devices require a certain amount of time to automatically detect the wearer's hand, to ensure a good user experience, the device can prompt the user to wait. For example, ... Figure 5 In (a), the wearable device displays an interface (see above for more information about this interface). Figure 1 (c) (not repeated here), this interface includes the "Add Research" button 106. After receiving the operation on the "Add Research" button 106, the wearable device begins to automatically detect the wearing hand. The detection of the wearing hand requires a certain amount of time, and the wearable device can display as follows: Figure 5The interface in (b) displays the message "Detecting, please wait" 500.

[0095] Step 202: Detect the user's dominant hand for eating.

[0096] Optionally, step 202 may include two implementation methods, denoted as method C and method D. Method C involves the user actively inputting their dominant hand for eating, while method D involves the device automatically detecting the dominant hand for eating.

[0097] Method C: The user actively inputs their dominant hand for eating.

[0098] One possible approach is to use a questionnaire to allow users to actively input their dominant hand for eating. For example, ... Figure 3 In section (c), the wearable device displays a questionnaire interface for the dominant eating hand. This interface includes a title 304 stating "Please enter your dominant eating hand," and options 305 (left hand), 306 (right hand), and 307 (not sure). The user can select the corresponding option. If the wearable device determines that option 305 is selected, then the dominant eating hand is determined to be the left hand. If the wearable device determines that option 306 is selected, then the dominant eating hand is determined to be the right hand. If option 307 is selected, then the dominant eating hand could be either the right or left hand. In this case, the wearable device can use other methods (e.g., method D below) to determine the dominant eating hand.

[0099] Method C can be combined with Method A or Method B mentioned above. Taking the combination of Method C and Method A as an example, both the dominant hand for eating and the wearing hand are inputted through a questionnaire. In this case, the processing flow of the wearable device may include: Figure 3 In (a), the wearable device displays an interface including a "Join Research" button 106. After receiving an operation on the "Join Research" button 106, the wearable device displays the following: Figure 3 The questionnaire interface for the wearing hand (b) shows the user's input of the wearing hand information. After receiving the setting operation, the wearable device displays the following: Figure 3 The user inputs their dominant eating hand in the questionnaire interface (c) of the example. Optionally, the setting operation can be various types of operation such as swiping up or left; or, the setting operation can be omitted, for example... Figure 3 In (b) of the above, if the wearable device detects that the user has selected to wear a wrist, it will automatically enter... Figure 3 The interface in (c) is shown. It should be noted that... Figure 3In this example, the questionnaire on the wearing hand is displayed first, followed by the questionnaire on the user's dominant eating hand. Alternatively, the questionnaire on the user's dominant eating hand can be displayed first, followed by the questionnaire on the wearing hand; or the questionnaires on the wearing hand and the user's dominant eating hand can be on the same screen, and so on. Taking a combination of method B and method A as an example, for instance, after the wearable device automatically detects the wearing hand, the questionnaire on the user's dominant eating hand pops up, where the user enters their dominant eating hand. Alternatively, the wearable device can also first display the questionnaire on the user's dominant eating hand, where the user enters their dominant eating hand, and then the wearable device automatically detects the wearing hand.

[0100] Another possible implementation is that, after detecting the wearing hand, the wearable device can ask the user if the wearing hand is their dominant hand for eating. For example, ... Figure 5 In step (c), the wearable device displays the query message "We have detected that you are currently wearing a watch on your left hand. Is your left hand your dominant eating hand?" It can also display "Yes" and "No" buttons. If the wearable device receives an operation for the "Yes" button, it determines that the dominant eating hand is the wearing hand. If the wearable device receives an operation for the "No" button, it determines that the dominant eating hand is another hand besides the wearing hand. This method of querying whether the dominant eating hand is the wearing hand occurs after the wearable device has detected the wearing hand. The process of the wearable device detecting the wearing hand has been described in step 201, including method A (questionnaire survey method) and method B (automatic device detection method). If step 201 adopts method A (i.e., questionnaire survey method), then after determining the wearing hand through the questionnaire survey, query information (e.g., ...) can be output. Figure 5 (c) Inquiry information), to inquire whether the dominant eating hand is the wearing hand. If step 201 adopts method B (i.e., the device automatically detects), then the processing flow of the wearable device may include: Figure 5 In (a), the wearable device displays an interface including an "Add to Research" button 106. Upon receiving an operation on the "Add to Research" button 106, the wearable device automatically detects the wearing hand and displays the following: Figure 5 The interface shown in (b) displays the message "Detecting, please wait" (500). After the wearable device detects the wearing hand, it displays the following... Figure 5 The question in (c) asks whether the eating hand is the wearing hand.

[0101] Method D: The device automatically detects the dominant hand used for eating.

[0102] Please see Figure 6 This is a schematic diagram of a process for a wearable device to detect the dominant hand used for eating, provided in an embodiment of this application. Figure 6 It can be understood as Figure 2The refinement of step 202, therefore, Figure 6 The steps in the text are represented by steps 202-a, 202-b, etc. For example... Figure 6 The process includes steps 202a to 202-e.

[0103] Step 202-a: Determine if the current scenario is a feeding scenario. If it is, proceed to step 202-b; otherwise, repeat step 202-a.

[0104] In this embodiment, the wearable device can determine whether the current scene is a eating scene based on current scene data. Optionally, the current scene data may include at least one of the following: current geographical location, surrounding environment data, and user's driving path.

[0105] Take, for example, a wearable device determining whether a current scene is a eating scene based on its current geographical location. For instance, the wearable device detects its current geographical location using a positioning module, which can optionally be a Global Positioning System (GPS) or a BeiDou Navigation Satellite System (BDS). If the current geographical location is at a eating location, the current scene is determined to be a eating scene; otherwise, it is determined not to be a eating scene. Exemplary eating locations may include restaurants, canteens, convenience stores, fast food restaurants, staff canteens, etc. In some embodiments, if the wearable device determines that its current geographical location is at a eating location, it directly determines that the current scene is a eating scene. In other embodiments, after determining that its current geographical location is at a eating location, the wearable device can also combine other information to determine whether the current scene is a eating scene. This other information may include the current time and / or the duration of stay at the eating location. For example, after determining that its current geographical location is at a eating location, if the wearable device determines that the current time is within a preset eating time range and / or the duration of stay at the eating location exceeds a preset duration, then the current scene is determined to be a eating scene. The preset eating time range can be a preset breakfast time range (e.g., 6:00-9:00), a lunch time range (e.g., 11:00-13:00), a dinner time range (e.g., 17:00-19:00), etc. For example, the preset eating time range can be pre-set in the wearable device; for instance, the wearable device includes the preset eating time range when it leaves the factory.

[0106] Take, for example, a wearable device determining whether a current scene is a eating scene based on surrounding environmental data. For instance, the wearable device collects environmental data from its surroundings using sensors. This environmental data may include image data and / or sound data. Based on this environmental data, the device determines whether the current scene is a eating scene. For example, after collecting image data, if the wearable device determines that the image data includes image features related to eating, such as food, tableware, or a restaurant, then the current scene is determined to be a eating scene. Similarly, after collecting sound data, if the wearable device determines that the sound data includes sound features related to eating, such as chewing sounds, shouts, or announcements, then the current scene is determined to be a eating scene. In some embodiments, if the wearable device determines that the image data includes image features related to eating and / or the sound data includes sound features related to eating, then the current scene is directly determined to be a eating scene. Optionally, the wearable device can also combine other information to determine whether the current scene is a eating scene. This other information may include: the current time, and / or the duration of stay in the eating scene. For example, if the wearable device determines that the current time falls within a preset eating time range, and / or the duration of stay in the eating scenario exceeds a preset duration, then the current scenario is determined to be a eating scenario. Please refer to the preceding description for information on preset eating time ranges; further details will not be repeated here.

[0107] Taking the example of a wearable device determining whether a current scenario is a eating scenario based on a user's driving path, the wearable device records the user's driving path. If this driving path matches a preset driving path, the current scenario is determined to be a eating scenario. The preset driving path can be the user's daily eating path, such as the path from the office to the company's dining area (cafeteria, break room, etc.). In some embodiments, if the wearable device determines that the user's driving path matches a preset driving path, it directly determines the current scenario to be a eating scenario. Optionally, the wearable device can also combine other information to determine whether the current scenario is a eating scenario. This other information may include: the current time, and / or the duration of stay in the eating scenario. For example, if the wearable device determines that the current time is within a preset eating time range, and / or the duration of stay in the eating scenario exceeds a preset duration, it determines the current scenario to be a eating scenario. Regarding the preset eating time range, please refer to the previous description, which will not be repeated here.

[0108] The above three methods can be used individually or in combination. Taking a combination of the three methods as an example, on weekdays, when users are in the office, they typically eat in dining areas such as the cafeteria or break room. In this case, the wearable device can determine whether the current scenario is a eating scenario based on the user's travel path. On holidays, when users eat in restaurants or outdoors (e.g., camping in a park), the device can determine whether the current scenario is a eating scenario based on the current geographical location or surrounding environmental data.

[0109] Step 202-b: Collect motion data.

[0110] For example, the motion data can be data collected by sensors such as accelerometers (ACC) and gyroscopes in wearable devices. Taking accelerometers as an example, accelerometers can collect multi-dimensional data, such as three-axis acceleration data along the X, Y, and Z axes, which can be used for motion analysis of the wearer's hand.

[0111] Step 202-c: Based on the motion data, determine whether the action of wearing the hand is a feeding action. If so, proceed to step 202-d; otherwise, proceed to step 202-e.

[0112] For example, based on motion data, the wearable device determines that the wearer's hand has lifted and lowered itself, and if the number of times this lifting and lowering action occurs within a certain time period is higher than a preset number, then the wearer determines that the hand action is an eating action; otherwise, the wearer determines that the hand action is not an eating action. The certain time period may include 1 minute, 2 minutes, 3 minutes, 5 minutes, 10 minutes, etc., and the specific value is not limited.

[0113] Optionally, the motion data can be collected locally by the wearable device (e.g., a watch), or by a device such as a ring, and after the motion data is collected, it is sent to the wearable device (e.g., a watch) for motion analysis.

[0114] Step 202-d: Determine if the eating hand is the wearing hand.

[0115] Step 202-e: Determine that the feeding hand is the hand other than the wearing hand.

[0116] Understandably, if the action of the wearing hand is an eating action, it means the user is using the wearing hand to eat, and thus the eating hand is identified as the wearing hand. If the action of the wearing hand is not an eating action, it means the user is using the non-wearing hand (i.e., the hand other than the wearing hand) to eat, and thus the eating hand is identified as the hand other than the wearing hand.

[0117] The above describes method D, which involves the device automatically detecting the dominant hand used for eating. This automatic detection method is imperceptible to the user, requires no user cooperation, and causes minimal disturbance.

[0118] Method D can be combined with either Method A or Method B mentioned earlier. Taking the combination of Method D and Method B as an example, where both the wearing hand and the dominant eating hand are automatically detected, the processing flow of the wearable device could include: Figure 7In (a), the wearable device displays an interface including a "Join Research" button 106. After receiving an operation on the "Join Research" button 106, the wearable device automatically detects the wearing hand and the dominant eating hand in the background. To provide feedback to the user, the wearable device may display something like... Figure 7 The interface in (b) displays the message "Your watch will study your daily eating behavior, please wait," and may also display a "Complete" button. After the wearable device receives an operation to press the "Complete" button, it exits this interface. Therefore, Figure 7 In this approach, users do not need to actively input their wearing hand and dominant eating hand; the wearable device can automatically detect these, requiring no user cooperation and providing a better user experience. For example, combining methods D and A, the wearable device can obtain user input of their wearing hand through a questionnaire, and then automatically detect their dominant eating hand.

[0119] In this embodiment, the wearable device can automatically detect the dominant hand used for eating through an algorithm model, which can be called a "dominant hand detection model." For example, the input parameters of the dominant hand detection model may include at least one of the following: current geographical location, surrounding environmental data, and the user's driving path; the output parameter is the dominant hand used for eating. The principle behind the dominant hand detection model detecting the dominant hand based on at least one of the following: current geographical location, surrounding environmental data, and the user's driving path, has been explained previously. Figure 6 As already described, this will not be repeated. Optionally, after the wearable device detects the dominant eating hand using the dominant eating hand detection model, it can display a prompt message to ask the user to confirm the accuracy of the dominant eating hand detection. For example, if the wearable device detects that the dominant eating hand is the right hand using the dominant eating hand detection model, it will display the prompt message "Your dominant eating hand is detected to be the right hand, correct it?" If the user corrects their dominant eating hand to the left hand, then the dominant eating hand is confirmed to be the left hand. If the wearable device determines that the model's detection result (i.e., right hand) is inconsistent with the user's correction result (i.e., left hand), it can retrain the dominant eating hand detection model to improve accuracy. Optionally, the dominant eating hand detection model can be pre-installed in the wearable device, for example, the wearable device is configured with the dominant eating hand detection model at the factory. Alternatively, the dominant eating hand detection model can be pre-installed in the application package of the first application. After the wearable device downloads the application package of the first application, the application package includes the dominant eating hand detection model.

[0120] In summary, the detection of the wearing hand can be achieved using a wearing hand detection model, and the detection of the dominant eating hand can be achieved using a dominant eating hand detection model. Furthermore, the wearable device can determine whether these two models need to be retrained based on user confirmation. For example, ... Figure 8A(a) The wearable device displays the detection result "Your wearing hand is your left hand, and your eating hand is your right hand." Optionally, it can also prompt "Correct?" and provide "Yes" and "No" buttons. If the wearable device receives an action on the "Yes" button, it will display as follows: Figure 8A (b) The user's questionnaire interface for the wearer's hand is shown here. After receiving the set operation, the wearable device displays the following: Figure 8A (c) The user's preferred eating hand questionnaire interface allows the user to input their preferred eating hand. Optionally, the setting operation can be an up swipe, a left swipe, etc.; alternatively, no setting operation is required, for example... Figure 8A In (b), if the wearable device detects that the user has selected to wear a wrist, it will automatically enter... Figure 8A (c) is the interface. It should be noted that... Figure 8A In this example, the wearable device displays the questionnaire on the wearing hand first, followed by the questionnaire on the dominant eating hand. Alternatively, the questionnaire on the dominant eating hand can be displayed first, followed by the questionnaire on the wearing hand; or the questionnaires on the wearing hand and the dominant eating hand can be displayed on the same screen, and so on. The wearable device can determine whether the model needs to be retrained based on the user's confirmation. For example, if the dominant eating hand output by the dominant eating hand detection model is inconsistent with the user's confirmed dominant eating hand, the dominant eating hand detection model will be retrained. Similarly, if the wearing hand detection model outputs the wearing hand inconsistent with the user's confirmed wearing hand, the wearing hand detection model will be retrained.

[0121] It should be noted that, Figure 8A The example provided illustrates how wearable devices output prompts to allow users to confirm the detection results of the wearing hand and the dominant eating hand. Optionally, if the wearable device is connected to other devices such as mobile phones or tablets, the prompts can be output through these other devices to remind the user to confirm the detection results of the wearing hand and the dominant eating hand. For example, regarding the connection between the wearable device and a mobile phone, please refer to [link to relevant documentation]. Figure 8B (a) The phone displays an interface, which may be the interface of an application on the phone (e.g., the Huawei Health app). This interface includes a message identifier to be processed. After receiving an operation targeting this identifier, the phone displays the following: Figure 8B (b) The interface includes questionnaires on the wearing hand and the dominant eating hand. If the user enters the wearing hand and dominant eating hand on this interface, the mobile phone sends the user's information to the wearable device. Continuing as... Figure 8B (a) The interface also includes a prompt message 800: "The above content is generated based on your personal information about the hand you wear and the hand you are dominant when eating. There may be errors. You can correct them."

[0122] Optionally, the wearable device or other devices connected to it (e.g., mobile phones, tablets, etc.) may display information about the wearing hand detection model and / or the eating dominant hand detection model. The information about the wearing hand detection model may include: the name and type of the wearing hand detection model (e.g., decision tree, random forest, etc.). The information about the eating dominant hand detection model may include the name and type of the eating dominant hand detection model (e.g., decision tree, random forest, etc.). For example, Figure 8B The prompt message 800 in (a) can be replaced with "The above content is generated based on the wearing hand detection model and the eating dominant hand detection model, and may contain errors. You can correct them."

[0123] Step 203: Determine whether the hand being worn is the same as the hand used for eating. If they are different, proceed to step S204; if they are the same, proceed to step S205.

[0124] Step 204: Use the first detection method to detect the user's eating behavior.

[0125] The first detection method may include: the wearable device detecting the user's eating behavior based on the user's physiological data and / or current scene data. Therefore, the first detection method may include Method 1 and / or Method 2. Method 1 involves the wearable device detecting the user's eating behavior based on the user's physiological data; Method 2 involves the wearable device detecting the user's eating behavior based on current scene data. These will be explained separately below.

[0126] Method 1: Wearable devices detect users' eating behavior based on their physiological data.

[0127] Optionally, the user's physiological data may include at least one of heart rate, energy, blood oxygen, body temperature, blood pressure, respiration, pulse, and blood glucose. Heart rate can be detected by a sensor in the wearable device, which may be an optical sensor, such as a photoplethysmography (PPG) sensor. Energy can be detected by a motion sensor in the wearable device, such as an inertial measurement unit (IMU) sensor. For example, an IMU sensor may include an accelerometer (ACC) or a gyroscope. The user's body temperature can be detected by a sensor in the wearable device, such as a temperature sensor, such as a negative temperature coefficient (NTC) thermistor. Furthermore, one or more of blood oxygen, blood pressure, respiration, and pulse may be detected by an optical sensor in the wearable device, such as a PPG sensor. Blood glucose can be detected by a blood glucose meter, which will be described in detail later.

[0128] In this embodiment, the wearable device detects the user's eating behavior based on the user's physiological data, which may include: repeatedly collecting the user's physiological data, including at least one of heart rate, energy, blood oxygen, body temperature, blood pressure, respiration, pulse, and blood glucose; determining the changes in the user's physiological data based on the repeatedly collected physiological data; and detecting the user's eating behavior based on the changes in physiological data. For example, if the changes in physiological data are consistent with preset changes, then it is determined that eating behavior exists, and the preset changes are used to describe the changes in the user's physiological data before and after eating.

[0129] Optionally, before determining the changes in a user's physiological data based on multiple collected physiological data, the wearable device may preprocess the multiple collected physiological data. The preprocessing may include one or more of the following: data segmentation, data filtering, and data quality assessment.

[0130] Taking data segmentation as an example, for instance, the multiple data collections could be collected once per minute. In this case, the wearable device can segment the collected data according to the collection time, such as the data collected in the i-th minute, the data collected in the (i+1)-th minute, and so on. The data segmented according to the collection time can be used to determine the changes in the user's physiological data.

[0131] Taking data filtering as an example, considering that the data collected by sensors in wearable devices may be subject to interference, the wearable device can perform interference removal processing on the collected data, for example, by using filtering algorithms (e.g., inertial filtering algorithms). Optionally, after data segmentation, interference removal can be performed separately on the segmented data, such as data collected every minute.

[0132] Taking data quality assessment as an example, in some cases, the data collected by wearable devices may be of poor quality. Wearable devices can perform quality assessments on the collected data and delete poor-quality data. For example, when a wearable device detects heart rate via a PPG sensor, during wear, the device may separate from the wrist skin or not fit the skin well. In such cases, the PPG sensor may not detect the heart rate or the detected heart rate may be inaccurate. Therefore, the wearable device can perform a quality assessment on the detected heart rate, for example, determining whether the heart rate is within a preset range. If it is not within the preset range, the detected heart rate data is determined to be of poor quality and can be deleted. Optionally, the preset range can be a heart rate range pre-set in the wearable device. It should be noted that this example uses the wearable device to assess the quality of collected heart rate data; quality assessments can also be performed on other physiological data, with the same principle, and will not be repeated here.

[0133] The following text uses four physiological data points—energy, heart rate, body temperature, and blood sugar—as examples to illustrate how wearable devices detect a user's eating behavior based on their physiological data.

[0134] Example 1

[0135] This example uses user physiological data, including energy, as an example. For instance, a wearable device collects energy multiple times, determines energy changes based on these multiple collections, and then detects the user's eating behavior based on these energy changes. For example, if the energy changes match a preset energy change pattern, then eating behavior is determined to have occurred. The preset energy change pattern describes the changes in the user's energy levels before and after eating.

[0136] To facilitate understanding, the meaning of energy in the embodiments of this application will be explained first. Energy is used to reflect the magnitude of a user's exercise and is directly proportional to the user's exercise. When a user exercises while wearing a wearable device (e.g., walking, picking up objects), the wearable device can determine the user's energy. For example, the wearable device collects motion data through sensors (e.g., accelerometers, gyroscopes, etc.) and determines the user's energy based on the motion data. Taking an accelerometer as an example, when a user exercises while wearing a wearable device, the accelerometer will collect three-axis acceleration data, namely: X-axis acceleration data, Y-axis acceleration data, and Z-axis acceleration data. Energy can be obtained from the three-axis acceleration data. For example, energy can be the sum, sum of squares, average, weighted average, variance, covariance, etc. of the three-axis acceleration data. Taking energy as the sum of squares of the three-axis acceleration data as an example, it can be expressed by the following formula:

[0137] W=X 2 +Y 2 +Z 2

[0138] Where W represents energy, X represents the X-axis acceleration data collected by the accelerometer, Y represents the Y-axis acceleration data collected by the accelerometer, and Z represents the Z-axis acceleration data collected by the accelerometer.

[0139] The meaning of energy has been explained above. As mentioned earlier, in this embodiment, the wearable device can detect a user's eating behavior based on their physiological data. Energy is a type of user physiological data; therefore, the process of the wearable device detecting a user's eating behavior based on this physiological data will be described below. It is worth emphasizing that, as described in the background art, in the prior art, detecting eating behavior through a wearable device requires the user to wear the device on the wrist of their dominant eating hand. The reason for this requirement is that when a user eats, their eating hand will make eating movements. Since the eating hand is the same as the wearing hand, the wearable device can capture the movements of the wearing hand (e.g., determine the movements of the wearing hand through motion data collected by an IMU sensor) and determine whether the user is eating based on the movements of the wearing hand. Unlike the prior art, in this embodiment, when the wearing hand and the dominant eating hand are different, the wearable device does not detect eating behavior by capturing the movements of the wearing hand, but rather by detecting eating behavior through the user's physiological data. The user's physiological data can be detected by sensors; for example, energy can be determined by motion data collected by an IMU sensor. In summary, in existing technologies, wearable devices collect data via IMU sensors, but this data is used to analyze the movements of the wearer's hand, which are then used to detect whether eating behavior is present. In this embodiment, the wearable device can also collect data via IMU sensors, but this data is used to determine the user's physiological data (e.g., energy levels), and the presence of eating behavior is detected based on this physiological data.

[0140] In real-world scenarios, a user's energy levels change before, during, and after eating. The following example illustrates these energy changes in a real-world context.

[0141] For example, when a user's physical activity is relatively low (e.g., sitting at an office), the triaxial acceleration data collected by the accelerometer is relatively small, meaning the energy level is low. Before eating, the user needs to move from their work location to the dining area (e.g., a restaurant or cafeteria), and then perform a series of actions such as paying and picking up the food. During this process, the user's physical activity increases, meaning the energy level increases. Afterward, the user sits down to eat, and the physical activity decreases, meaning the energy level decreases. After eating, the user needs to leave the dining area, and the energy level increases again. Therefore, the energy level changes before, during, and after eating according to the pattern of increase -> decrease -> increase again. For a more intuitive understanding, please refer to [link to relevant documentation]. Figure 9A(a) shows the energy change curves of a user before, during, and after eating. These curves are located on a coordinate system where the horizontal axis represents time and the vertical axis represents energy. The energy change curves show that in real-world scenarios, the user's energy changes before, during, and after eating follow a pattern of increase → decrease → increase again.

[0142] Therefore, based on the above principles, wearable devices can collect energy multiple times. Based on the collected energy, the energy changes can be determined. If the energy changes match a preset energy change pattern, then feeding behavior is confirmed. The preset energy change pattern can be the previously mentioned pattern of energy increasing -> decreasing -> increasing again.

[0143] For example, wearable devices collect energy multiple times, and based on the collected energy, generate energy change curves, such as... Figure 9A (b) As shown in the energy change curve, the wearable device determines that there is a curve segment on the energy change curve. The energy change of this curve segment is consistent with the preset energy change, that is, the energy changes in the pattern of rising → falling → rising again. Then it is determined that there is eating behavior.

[0144] Optionally, the wearable device can also determine the feeding time based on energy changes, which can include the start and / or end times of feeding. For example, such as Figure 9A (b) The feeding begins at time T1, where time T1 can be any time within time period A, such as... Figure 9A (b) Time period A is the time period corresponding to the falling edge of the energy. For example... Figure 9A (b) The end time of eating is time T2, where time T2 can be any time within time period B, such as... Figure 9A (b) Time period B is the time period corresponding to the rising edge of the energy. Optionally, the wearable device can also verify the accuracy of the detection of eating behavior by measuring the eating time. For example, if the eating time falls within a preset eating time range, then eating behavior is confirmed. The preset eating time range can be a preset breakfast time range (e.g., 6:00-9:00), a lunch time range (e.g., 11:00-13:00), a dinner time range (e.g., 17:00-19:00), etc.

[0145] Example 2

[0146] This example uses user physiological data, including heart rate, as an example. For instance, a wearable device collects heart rate data multiple times. Based on these multiple heart rate data collections, the changes in heart rate are determined, and the user's eating behavior is detected based on these changes. For example, if the heart rate changes match a preset heart rate change pattern, then eating behavior is determined to have occurred. The preset heart rate change pattern is used to describe the changes in the user's heart rate before and after eating.

[0147] In real-world scenarios, a user's heart rate changes before, during, and after eating. The following example illustrates these heart rate changes in a real-world context.

[0148] For example, before eating, a person needs to move to the dining area and perform various actions such as paying and picking up food. During this process, the amount of physical activity increases, so the heart rate rises. During eating, as food digestion promotes blood flow, the heart rate does not decrease but remains at a higher level. After eating, the heart rate gradually returns to normal. Therefore, the changes in heart rate before, during, and after eating follow a pattern of: increase -> remain stable -> decrease. For a more intuitive understanding, please refer to [link to relevant documentation]. Figure 9B (a) shows the heart rate changes of a user before, during, and after eating. This curve is located on a coordinate system where the horizontal axis is time and the vertical axis is heart rate. The heart rate change curve shows that in real-world scenarios, the user's heart rate changes in the following pattern: increase → remain stable → decrease before, during, and after eating.

[0149] Therefore, based on the above principle, wearable devices can collect heart rate data multiple times. Based on these multiple heart rate data collections, the heart rate changes are determined. If the heart rate changes match a preset heart rate pattern, then eating behavior is confirmed. The preset heart rate change pattern can be the aforementioned pattern of heart rate increase -> maintenance -> decrease.

[0150] For example, a wearable device collects heart rate data multiple times, generates a heart rate variation curve based on the collected heart rate data, and then uses this heart rate variation curve as an example... Figure 9B (b) As shown in the heart rate change curve, if the wearable device determines that there is a curve segment on the heart rate change curve, and the heart rate change of this curve segment is consistent with the preset heart rate change, that is, the heart rate changes in the pattern of rising → maintaining → falling, then it is determined that there is eating behavior.

[0151] Optionally, the wearable device can also determine the eating time based on heart rate changes, including the start and / or end times of eating. For example, such as... Figure 9B (b) The feeding begins at time T1, where time T1 can be any time within time period A, such as... Figure 9B (b) Time period A may include the time period corresponding to the rising edge of the heart rate. For example... Figure 9B (b) The end time of eating is time T2, where time T2 can be any time within time period B, such as... Figure 9B(b) Time period B may include the time period corresponding to the decrease in heart rate. Optionally, the wearable device can also verify the accuracy of the detection of eating behavior by measuring the eating time. For example, if the eating time falls within a preset eating time range, then eating behavior is confirmed. The preset eating time range may be a preset breakfast time range (e.g., 6:00-9:00), a lunch time range (e.g., 11:00-13:00), a dinner time range (e.g., 17:00-19:00), etc.

[0152] Example 3

[0153] This example uses user physiological data, including body temperature. For instance, a wearable device collects body temperature multiple times. Based on these multiple temperature readings, the changes in body temperature are determined, and the user's eating behavior is detected based on these changes. For example, if the changes in body temperature match a preset set of temperature changes, then eating behavior is determined. The preset set of temperature changes describes the changes in the user's body temperature before and after eating.

[0154] In real-world scenarios, a user's body temperature changes before, during, and after eating. The following example illustrates these changes in body temperature during and after eating in a real-world context.

[0155] For example, before eating, a person needs to move to the dining area and perform various actions such as paying and picking up food. During this process, the amount of physical activity increases, causing body temperature to rise. During eating, as food digestion promotes increased blood flow, body temperature does not decrease but remains at a higher level. After eating, body temperature gradually returns to normal. Therefore, the changes in body temperature before, during, and after eating follow a pattern of rising -> maintaining -> falling. For a more intuitive understanding, please refer to [link to relevant documentation]. Figure 9C (a) shows the body temperature change curves of a user before, during, and after eating. These curves are located on a coordinate system where the horizontal axis represents time and the vertical axis represents body temperature. The body temperature change curves show that, in real-world scenarios, the user's body temperature changes in the following pattern: rising → remaining stable → falling before, during, and after eating.

[0156] Therefore, based on the above principle, wearable devices can collect body temperature multiple times. Based on these multiple temperature readings, the changes in body temperature can be determined. If the changes in body temperature match a preset pattern, then eating behavior is confirmed. The preset pattern of body temperature changes can be the aforementioned pattern of rising -> remaining constant -> falling.

[0157] For example, wearable devices collect body temperature multiple times, and based on these multiple temperature readings, generate a body temperature change curve. Figure 9C(b) shows the body temperature change curve. If the wearable device determines that there is a segment of the body temperature change curve where the body temperature change is consistent with the preset body temperature change, i.e., the body temperature changes in the pattern of rising → holding → falling, then it is determined that there is eating behavior.

[0158] Optionally, the wearable device can also determine the feeding time based on changes in body temperature, including the start and / or end times of feeding. For example, such as... Figure 9C (b) The feeding begins at time T1, where time T1 can be any time within time period A, such as... Figure 9C (b) Time period A includes the time period corresponding to the rise in body temperature. The end of eating is time T2, which can be any time within time period B, such as... Figure 9C (b) Time period B includes the time period corresponding to the decrease in body temperature.

[0159] Figure 9B and Figure 9C The principles of detecting a user's eating behavior through heart rate and body temperature have been explained. If blood pressure, respiration, pulse, etc. are used to detect a user's eating behavior, the principle is the same as that of heart rate or body temperature, so it will not be repeated.

[0160] Example 4

[0161] This example uses user physiological data, including blood glucose, as an example. For instance, a wearable device collects blood glucose data multiple times. Based on these multiple collections, the changes in blood glucose levels are determined, and the user's eating behavior is detected based on these changes. For example, if the changes in blood glucose levels match a preset pattern, then eating behavior is determined. The preset pattern describes the changes in blood glucose levels before and after the user eats.

[0162] In real-world scenarios, a user's blood sugar levels fluctuate before, during, and after eating. For example, before eating, blood sugar is low due to hunger; during eating, blood sugar gradually rises as food is digested; and after eating, blood sugar gradually returns to normal. Therefore, the changes in blood sugar before, during, and after eating follow a pattern of low -> gradually rising -> gradually falling. For a more intuitive understanding, please refer to [link to relevant documentation]. Figure 9D (a) shows the blood glucose change curves before, during, and after a user's meal. These curves are located on a coordinate system where time is on the horizontal axis and blood glucose is on the vertical axis. The blood glucose change curves show that, in real-world scenarios, the user's blood glucose levels follow a pattern of low -> gradually rising -> gradually falling before, during, and after a meal.

[0163] Therefore, based on the above principles, wearable devices can collect blood glucose levels multiple times. Based on these multiple collections, the changes in blood glucose levels can be determined. If the changes match a preset pattern, then eating behavior is confirmed. This preset pattern could be the previously mentioned pattern of low blood glucose levels followed by a gradual increase followed by a gradual decrease.

[0164] For example, wearable devices collect blood glucose levels multiple times, and based on these multiple collections, a blood glucose change curve is generated. Figure 9D (b) shows the blood glucose change curve. If the wearable device determines that there is a segment on the blood glucose change curve where the blood glucose change is consistent with the preset blood glucose change, i.e., the blood glucose shows a pattern of low -> gradually rising -> gradually falling, then it is determined that there is eating behavior.

[0165] Optionally, the wearable device can also determine the timing of meals based on blood glucose levels, including the start and / or end times of meals. For example, such as... Figure 9D (b) The feeding begins at time T1, where time T1 can be any time within time period A, such as... Figure 9D (b) Time period A includes the time period corresponding to the peak of blood glucose rise. The end of eating is time T2, which can be any time within time period B, such as... Figure 9D (b) Time period B includes the time period corresponding to the blood glucose decrease.

[0166] As mentioned above, wearable devices need to monitor blood glucose levels multiple times. One possible approach is for the wearable device to connect to a blood glucose meter, which collects the blood glucose data and then sends it to the wearable device. Optionally, the blood glucose meter can be an upper arm-type meter, a wristband-type meter, etc., without limitation. For example, the blood glucose meter could be a continuous glucose monitor (CGM). In this way, the wearable device can obtain multiple blood glucose readings, thus obtaining blood glucose fluctuations, and detecting eating behavior based on these fluctuations. Alternatively, the blood glucose meter can also generate its own blood glucose fluctuation data and detect eating behavior based on these fluctuations, then send the eating behavior detection results to the wearable device. The eating behavior detection results can include whether eating behavior occurred. Optionally, if eating behavior occurred, the blood glucose meter can also determine the eating time (the start and / or end time of eating) and then send the eating time to the wearable device. It is understood that if future wearable devices (e.g., watches) can integrate a blood glucose monitoring module, the wearable device can independently perform blood glucose monitoring without needing a separate blood glucose meter.

[0167] The above embodiments illustrate the principle by which wearable devices detect a user's eating behavior based on energy, heart rate, body temperature, blood glucose, etc. Energy, heart rate, body temperature, and blood glucose can be used individually or in combination. Taking the combination of energy and heart rate as an example, after the wearable device obtains an energy change curve, if there is a curve segment 1 on this curve, and the energy change of curve segment 1 is consistent with a preset energy change, it indicates that eating behavior is suspected to have occurred within the time period corresponding to curve segment 1. The wearable device can determine a curve segment 2 on the heart rate change curve based on the time corresponding to curve segment 1, and the time corresponding to curve segment 2 overlaps with (e.g., is the same as) the time corresponding to curve segment 1. If the heart rate change of curve segment 2 is consistent with a preset heart rate change, it indicates that eating behavior did indeed occur within the stated time period. This method improves the accuracy of eating behavior detection.

[0168] The above embodiments illustrate Method 1, which involves a wearable device detecting a user's eating behavior based on their physiological data. In the embodiments of this application, Method 1 can be implemented by an algorithm model. For example, please refer to... Figure 10 The wearable device includes algorithm model 1. The input parameters of algorithm model 1 include the user's physiological data. The wearable device runs algorithm module 1 and obtains an output result, which includes a confidence level of 1. Confidence level 1 indicates the probability of eating behavior. If the wearable device determines that the value of confidence level 1 is greater than or equal to threshold 1, it determines that eating behavior exists; otherwise, it determines that eating behavior does not exist. The principle by which algorithm model 1 detects user eating behavior based on the user's physiological data is explained in Method 1 of the first detection method described above (i.e., the wearable device detects user eating behavior based on the user's physiological data), and will not be elaborated further. Optionally, the wearable device can also determine the start and / or end time of eating based on the confidence level 1 output by algorithm model 1. For example, the input parameter of algorithm model 1, i.e., the user's physiological data, is a real-time data stream, and the output result of algorithm model 1, i.e., confidence level 1, is also a real-time data stream. In other words, the value of confidence level 1 changes dynamically over time. Specifically, when the confidence level 1 reaches the threshold 1, it is determined that there is eating behavior, and the moment when the confidence level 1 reaches the threshold 1 is recorded as the start time of eating; when the confidence level 1 is lower than the threshold 1, it is determined that there is no eating behavior, and the moment when the confidence level 1 is lower than the threshold 1 is recorded as the end time of eating.

[0169] Method 2: Wearable devices detect the user's eating behavior based on current scene data.

[0170] Optionally, the current scene data may include at least one of the following: current geographical location, surrounding environment data, and user driving path.

[0171] For example, see Figure 11This is a schematic flowchart of an embodiment of the eating behavior detection method provided in this application. Figure 11 The process includes the following steps:

[0172] Step 1101: Determine the current scenario as a eating scenario.

[0173] Optionally, step 1101 may include: the wearable device determining that the current scenario is a eating scenario based on at least one of the following: current geographical location, surrounding environmental data, and the user's driving path. It should be noted that the implementation principle of step 1101 is the same as described above. Figure 6 The implementation principle of step 202-a is the same, so it will not be repeated.

[0174] Step 1102: The user is detected to have sat down.

[0175] For example, wearable devices can collect motion data through sensors such as accelerometers and gyroscopes, and use the motion data to determine whether the user's posture is sitting.

[0176] Step 1103: The amount of movement of the wearing hand is less than the preset amount of movement (e.g., it is in a static state).

[0177] For example, wearable devices can collect motion data through sensors such as accelerometers and gyroscopes, and use the motion data to detect whether the amount of movement of the wearing hand is less than a preset amount of movement, such as whether the wearing hand is in a stationary state.

[0178] It should be noted that steps 1102 and 1103 may or may not be executed, therefore Figure 11 These two steps are represented by dashed lines. If steps 1102 and 1103 are not executed, that is, after step 1101 is executed, step 1104 is executed directly.

[0179] Step 1104: Confirm the start of feeding.

[0180] Step 1105: Record the start time of eating.

[0181] Optionally, the start time of eating can be at least one of the following:

[0182] (a) The moment of entering the eating scenario. For example, if the wearable device determines that the current scenario is a eating scenario based on the current geographical location, then the moment of entering the eating scenario is determined as the start time of eating.

[0183] (b) The moment the user sits down. For example, if a wearable device collects motion data through sensors such as accelerometers and gyroscopes, and determines that the user's posture has been adjusted to a sitting position based on the motion data, then the moment when the posture has been adjusted to a sitting position (e.g., the moment when the motion data was collected) is the moment when eating begins.

[0184] (c) The moment when the amount of motion of the wearing hand changes from greater than or equal to a preset amount of motion to less than a preset amount of motion. It is understood that when the wearing hand (e.g., left hand) is different from the user's dominant eating hand (e.g., right hand), the user's movements are smaller when using the dominant eating hand, and increase after the user finishes eating. Therefore, in one possible implementation, the wearable device can acquire the amount of motion of the wearing hand, compare the amount of motion of the wearing hand with the preset amount of motion, and thus determine the eating time. For example, if it is detected that the amount of motion of the wearing hand changes from greater than or equal to the preset amount of motion to less than the preset amount of motion at a first moment, then the first moment is determined as the start of eating.

[0185] Step 1106: The amount of movement of the wearing hand is detected to reach the preset amount of movement (e.g., in a non-static state).

[0186] For example, wearable devices can collect motion data using sensors such as accelerometers and gyroscopes. This motion data can then be used to detect whether the wearer's hand has moved to a preset range of motion, such as changing from a stationary to a non-stationary state. Understandably, when the wearer's hand (e.g., left hand) differs from their dominant eating hand (e.g., right hand), the wearer's hand movements are smaller when eating with their dominant eating hand, and more pronounced after eating, such as changing from a stationary to a non-stationary state.

[0187] Step 1107: The user was detected to have gotten up.

[0188] For example, wearable devices can collect motion data through sensors such as accelerometers and gyroscopes, and use the motion data to determine whether the user's posture changes from sitting to standing, i.e., the user gets up.

[0189] It should be noted that steps 1106 and 1107 may or may not be executed, therefore Figure 11 These two steps are represented by dashed lines. If steps 1106 and 1107 are not executed, that is, after step 1105 is executed, step 1108 is executed directly.

[0190] Step 1108: Determine if the user leaves the eating scene.

[0191] Optionally, step 1108 may include: the wearable device determining that the user has left the eating scene based on at least one of the following: current geographical location, surrounding environment data, and user driving path.

[0192] Take, for example, a wearable device determining whether a user has left a dining environment based on their current geographical location. The wearable device detects its current geographical location through a positioning module (e.g., GPS or BDS). If it determines that the current geographical location has moved away from the dining location (e.g., restaurant, canteen, convenience store, fast food restaurant, staff canteen, etc.), then it is determined that the user has left the dining environment.

[0193] Take, for example, a wearable device determining whether a user has left a dining scene based on surrounding environmental data. For instance, the wearable device collects environmental data from its surroundings using sensors. This environmental data can include image data and / or sound data. Based on this data, it determines whether the user has left the dining scene. For example, the wearable device collects image data in real time, forming an image stream. If the earlier images in the image stream contain food, tableware, restaurants, or other food-related image features, it indicates the user has entered the dining scene. If later images in the image stream do not contain these food-related features, it indicates the user has left the dining scene. As another example, the wearable device collects sound data in real time, forming a sound stream. If the earlier sounds in the sound stream contain chewing sounds, shouts, or announcements of dishes, it indicates the user has entered the dining scene. If later sounds in the sound stream do not contain these food-related features, it indicates the user has left the dining scene.

[0194] Take, for example, a wearable device determining whether a user has left a dining area based on their travel path. The wearable device records the user's travel path; if this path matches a second preset travel path, it determines that the user has left the dining area. This second preset travel path could be the path the user takes after finishing a meal in daily life. For instance, in daily life, a user might move from their office to a dining area such as a cafeteria or break room to eat. After eating, the user might return from the cafeteria or break room to their office or move to a rest area. Therefore, the second preset path could be the user's travel path from the cafeteria or break room back to their office or to a rest area.

[0195] Step 1109: Confirm the end of eating.

[0196] Step 1110: Record the time when the meal ends.

[0197] Optionally, the end of the meal can be at least one of the following:

[0198] (a) The moment of leaving the eating scene. For example, if the wearable device determines that the user has left the eating scene based on the current geographical location, then the moment of leaving the eating scene is determined as the end of the eating.

[0199] (b) The moment the user stands up. For example, if the wearable device collects motion data through sensors such as accelerometers and gyroscopes, and determines that the user's posture has changed from sitting to standing based on the motion data, then the moment of adjustment to standing (e.g., the moment of motion data collection) is determined as the moment of the end of eating.

[0200] (c) The moment when the amount of motion of the wearing hand changes from less than a preset amount to greater than or equal to a preset amount. As mentioned above, when the wearing hand (e.g., the left hand) is different from the eating-dominant hand (e.g., the right hand), the wearing hand's movements are smaller when the user eats with their eating-dominant hand, and increase after the user finishes eating. Therefore, if it is detected that the amount of motion of the wearing hand changes from less than a preset amount to greater than or equal to a preset amount at a second moment, then the second moment is determined as the end of eating. Optionally, the preset amount of motion can be an empirical value preset in the wearable device.

[0201] The above embodiments illustrate Method Two, which involves a wearable device detecting a user's eating behavior based on current scene data. In the embodiments of this application, Method Two can be implemented by an algorithm model. For example, please refer to... Figure 12 The wearable device includes algorithm model 2, whose input parameters include current scene data. The wearable device runs algorithm module 2 and obtains an output result, which includes a confidence level of 2, indicating the probability of eating behavior. If the wearable device determines that the confidence level of 2 is greater than or equal to a threshold of 2, it determines that eating behavior exists; otherwise, it determines that eating behavior does not exist. The principle by which algorithm model 2 detects user eating behavior based on current scene data is explained in Method 2 of the first detection method above (i.e., the wearable device detects user eating behavior based on current scene data), and will not be elaborated further. Optionally, the wearable device can also determine the start and / or end time of eating based on the confidence level of 2 output by algorithm model 2. For example, the input parameter of algorithm model 2, i.e., the current scene data, is a real-time data stream, and the output result of algorithm model 2, i.e., the confidence level of 2, is also a real-time data stream. In other words, the value of confidence level 2 changes dynamically over time. Specifically, when the confidence level 2 reaches the threshold 2, it is determined that there is eating behavior, and the moment when the confidence level 2 reaches the threshold 2 is recorded as the start time of eating; when the confidence level 2 is lower than the threshold 2, it is determined that there is no eating behavior, and the moment when the confidence level 2 is lower than the threshold 2 is recorded as the end time of eating.

[0202] The above embodiments provide a detailed description of the first detection method. The first detection method includes Method 1 (detecting eating behavior based on the user's physiological data) and Method 2 (detecting eating behavior based on current scene data). Method 1 and Method 2 can be used individually or in combination, without limitation. Taking the combination of Method 1 and Method 2 as an example, one possible combination is to obtain confidence level 1 through Method 1 (algorithm model 1) and confidence level 2 through Method 2 (algorithm model 2); then, based on confidence level 1 and confidence level 2, a comprehensive judgment is made on whether eating behavior exists. For example, if the value of confidence level 1 is greater than or equal to threshold 1 and the value of confidence level 2 is greater than or equal to threshold 2, then eating behavior is determined to exist; or, confidence level a is obtained based on confidence level 1 and confidence level 2, where confidence level a can be the sum, average, or weighted average of confidence levels 1 and 2. If the value of confidence level a is greater than or equal to threshold a, then eating behavior is determined to exist. Taking confidence level 'a' obtained from confidence levels 1 and 2 as an example, the wearable device can optionally determine the start and / or end time of eating based on confidence level 'a'. For example, both confidence levels 1 and 2 are real-time data streams, meaning confidence level 'a' is also a real-time data stream. Specifically, when the value of confidence level 'a' reaches a threshold 'a', it is determined that eating behavior exists, and the moment when the value of confidence level 'a' reaches the threshold 'a' is recorded as the start time of eating; when the value of confidence level 'a' is lower than the threshold 'a', it is determined that eating behavior does not exist, and the moment when the value of confidence level 'a' is lower than the threshold 'a' is recorded as the end time of eating.

[0203] In summary, the algorithm model corresponding to the first detection method (hereinafter referred to as the first eating behavior detection model) includes algorithm model 1 and / or algorithm model 2. Optionally, after the wearable device detects eating behavior using the first eating behavior detection model, it can confirm whether the detection of eating behavior is accurate (confirmation methods include the first and second confirmation methods described below). If the detection of eating behavior is confirmed to be inaccurate, the first eating behavior detection model is retrained to improve accuracy.

[0204] Step 105: Use the second detection method to detect the user's eating behavior.

[0205] The second detection method may include detecting the user's eating behavior based on the movement of the wearing hand.

[0206] For example, see Figure 13 This is a schematic diagram of another process for a method for detecting eating behavior provided in an embodiment of this application. Figure 13 The process includes:

[0207] Step 1301: Collect data using sensors.

[0208] For example, the sensor can be an accelerometer, a gyroscope, etc.

[0209] Step 1302: Determine the action of the wearing hand based on the data collected by the sensor.

[0210] Step 1303: Determine that the action of wearing the hand is a eating action.

[0211] For example, based on motion data, the wearable device determines that the wearer's hand has lifted and lowered itself, and if the number of times this lifting and lowering action occurs within a certain period of time is higher than a preset number, then the wearer determines that the action of the wearer's hand is an eating action. The certain period of time may include 1 minute, 2 minutes, 3 minutes, 5 minutes, 10 minutes, etc., and the specific value is not limited.

[0212] Step 1304: Record the start time of eating.

[0213] Optionally, the feeding start time can be the moment when the action of wearing the hand is determined to be the feeding action.

[0214] Step 1305: Confirm that the eating action of wearing the hand has ended.

[0215] For example, if the wearable device determines that the wearing hand has finished lifting and lowering its body based on motion data, and no more frequent lifting and lowering movements occur within a preset time, then the eating action of the wearing hand is determined to be over.

[0216] Step 1306: Record the time when the meal ends.

[0217] Optionally, the end of eating can be the moment when the eating action of the wearing hand ends.

[0218] In this embodiment of the application, the second detection method can be implemented through an algorithmic model (hereinafter referred to as the second eating behavior detection model). For example, please refer to... Figure 14 The wearable device includes algorithm model 3, which is used to implement the second detection method. For example, the input parameters of algorithm model 3 include the action of the wearing hand. The wearable device runs algorithm module 3 and obtains the output result, which includes a confidence level of 3, indicating the probability of eating behavior. If the wearable device determines that the confidence level 3 is greater than or equal to the threshold 3, it determines that eating behavior exists; otherwise, it determines that eating behavior does not exist. The principle by which algorithm model 3 detects user eating behavior based on the action of the wearing hand can be found in the previous text. Figure 13Steps 1303 to 1306 are not repeated here. Optionally, the wearable device can also determine the feeding start time and / or feeding end time based on the confidence level 3 output by algorithm model 3. For example, the input parameter of algorithm model 3, i.e., the action data of the wearing hand, is a real-time data stream, and the output result of algorithm model 3, i.e., the confidence level 3, is also a real-time data stream. In other words, the value of confidence level 3 changes dynamically over time. Specifically, when the value of confidence level 3 reaches threshold 3, it is determined that there is feeding behavior, and the moment when the value of confidence level 3 reaches threshold 3 is recorded as the feeding start time; when the value of confidence level 3 is lower than threshold 3, it is determined that there is no feeding behavior, and the moment when the value of confidence level 3 is lower than threshold 3 is recorded as the feeding end time.

[0219] Optionally, after the wearable device detects eating behavior using the second eating behavior detection model, it can confirm whether the detection of eating behavior is accurate (confirmation methods include the first and second confirmation methods described below). If the detection of eating behavior is confirmed to be inaccurate, the second eating behavior detection model can be retrained to improve accuracy.

[0220] In summary, in this embodiment, when the hand worn is different from the hand used for eating, the user's eating behavior is detected using a first detection method. This first detection method includes Method 1 and / or Method 2. Method 1 can be implemented using algorithm model 1, and Method 2 can be implemented using algorithm model 2. That is, the algorithm model corresponding to the first detection method (hereinafter referred to as the first eating behavior detection model) includes algorithm model 1 and / or algorithm model 2. When the hand worn is the same as the hand used for eating, the user's eating behavior is detected using a second detection method. This second detection method can be implemented using algorithm model 3. That is, the algorithm model corresponding to the second detection method (hereinafter referred to as the second eating behavior detection model) includes algorithm model 3.

[0221] In this embodiment of the application, after the wearable device detects eating behavior, it can confirm / verify / verify the eating behavior to determine whether the detection of the eating behavior is accurate. Optionally, the confirmation method may include any one of the following two methods.

[0222] The first confirmation method involves the wearable device detecting a user's eating behavior and then displaying a prompt message to ask the user to confirm whether the detection of the eating behavior is correct.

[0223] For example, see Figure 15A (a) The wearable device displays "Your eating behavior has been detected. Do you want to view this?"; it can also display a view button and an ignore button. When the wearable device detects an operation on the ignore button, it cancels the display of the prompt message, for example, it can return to the wearable device's main interface. When the wearable device detects an operation on the view button, it displays as follows: Figure 15A(b) The interface includes information about the user's eating behavior detected by the wearable device, which may include the start and / or end times of eating. Figure 15A (b) If the user determines that the start time of eating is incorrect, the start time can be adjusted. For example, the "+" button can be used to move the start time forward, and the "-" button can be used to move the start time forward. If the user determines that the end time of eating is incorrect, the end time can be adjusted. For example, the "+" button can be used to move the end time forward, and the "-" button can be used to move the end time forward. Figure 15A (b) The interface also includes a confirmation button. After the wearable device receives the operation of the confirmation button, it determines the start time in the interface as the start time of the current eating behavior and determines the end time in the interface as the end time of the current eating behavior. Figure 15A (b) The interface also includes a "no food" button. Upon receiving an operation on this button, the wearable device deletes the relevant information regarding the current eating behavior. Optionally, Figure 15A (b) The interface may also include a food addition button (not shown in the figure). When an operation is received on the button, the type of food to be eaten is added, such as rice, noodles, various vegetables, etc. Optionally, the amount of food can also be added, such as the weight of the food, such as how many grams of rice, how many grams of noodles, etc.

[0224] Figure 15A In one embodiment, after detecting a user's eating behavior, the wearable device notifies the user of the eating behavior but does not determine whether it is breakfast, lunch, or dinner. In other embodiments, after detecting the user's eating behavior, the wearable device can also determine whether the eating behavior is breakfast, lunch, or dinner, for example, based on the time the eating behavior occurred. Afterward, the wearable device prompts the user to confirm whether the detection of breakfast / lunch / dinner is correct. For example, please see... Figure 15B (a) The wearable device displays "Detected you have breakfast, do you want to check?"; it can also display a "Check" button and an "Ignore" button. When the wearable device detects an operation on the "Ignore" button, it cancels the display of the prompt message, for example, by returning to the wearable device's main interface. When the wearable device detects an operation on the "Check" button, it displays as follows: Figure 15B (b) The interface includes information about the user's breakfast detected by the wearable device, which may include the start and end times of the breakfast. If the user determines that the start time is incorrect, they can adjust it, for example, by using the "+" button to move the start time backward and the "-" button to move the start time forward. If the user determines that the end time is incorrect, they can adjust it, for example, by using the "+" button to move the end time backward and the "-" button to move the end time forward. Figure 15B(b) The interface also includes a confirmation button. After the wearable device receives the operation of the confirmation button, it determines the start time in the interface as the start time of breakfast and the end time in the interface as the end time of breakfast. Figure 15B (b) The interface also includes a "Breakfast Not Eaten" button. Upon receiving an action on this button, the wearable device deletes any breakfast-related information. Optionally, Figure 15B (b) The interface may also include a food addition button (not shown in the figure). When an operation is received on the button, the type of food to be eaten is added, such as rice, noodles, various vegetables, etc. Optionally, the amount of food can also be added, such as the weight of the food, such as how many grams of rice, how many grams of noodles, etc.

[0225] In some embodiments, the wearable device prompts the user to view the eating behavior each time it detects it. Considering that too many prompts might be disruptive, in other embodiments, the wearable device may periodically remind the user to view the eating behavior. The period can be daily, every three days, weekly, monthly, etc., and is not limited to this. For example, regarding daily reminders, please refer to [link to relevant documentation]. Figure 15C (a) The wearable device displays the message "You have eaten today. Please check and verify," along with an ignore button and a view button. When the wearable device receives an operation to press the ignore button, it cancels the display of the message, for example, by returning to the wearable device's main interface. When the wearable device receives an operation to press the view button, it displays the following: Figure 15C (b) The interface includes information about the day's breakfast recorded by the wearable device, such as the start and / or end times of breakfast. Users can also adjust the start and / or end times. Figure 15C (b) The interface also includes an "OK" button and an "I haven't eaten breakfast" button. For information on the functions of these two buttons, please refer to [link to relevant documentation]. Figure 15B The explanation will not be repeated here. Figure 15C In (b), after receiving the set operation, the wearable device can display, as shown below. Figure 15C (c) This interface displays information about the day's lunch recorded by the wearable device. Users can confirm or correct this information, following the same principle. Optionally, the setting operation can be an up swipe, a left swipe, etc.; alternatively, no setting operation may be required, for example... Figure 15C In (b), the wearable device detects that the user has clicked the confirmation button, and then automatically enters... Figure 15C (c) is the interface. Similarly, Figure 15C In (c), after receiving the setting operation, the wearable device can display, as shown below. Figure 15CThe interface in (d) displays information about the day's dinner recorded by the wearable device. Users can confirm or correct this information, following the same principle. Optionally, the setting operation can be an up swipe, a left swipe, etc.; alternatively, no setting operation may be required, for example... Figure 15C In (c), if the wearable device detects that the user has clicked the confirmation button, it will automatically enter... Figure 15C (d) interface.

[0226] Continue with Figure 15C For example, assuming the wearable device did not detect the user's dinner behavior, the wearable device could display... Figure 15C (e) The interface displays "No dinner information detected," and can also display "OK" and "Add Dinner" buttons. Upon receiving an action to the "OK" button, the wearable device confirms that no dinner has been prepared. Upon detecting an action to the "Add Dinner" button, the wearable device displays the following: Figure 15C (f) is the interface used to set the start and end times of dinner. Optionally, you can add the types of food to be eaten, such as rice, noodles, various vegetables, etc. Optionally, you can also add the quantity of food, such as the weight in grams, like how many grams of rice, how many grams of noodles, etc. Continue as follows... Figure 15C (f) The interface includes an "Add" button. Upon receiving an operation to press the "Add" button, the wearable device adds the dinner activity, determining the start time displayed on the interface as the start time of dinner and the end time as the end time of dinner. After successful addition, the wearable device can also display something like... Figure 15C The message "Added successfully" shown in (g)

[0227] In the previous text Figures 15A to 15C In the wearable device, the interface provides "+" and "-" buttons for adjusting the start and / or end times of eating. In other embodiments, the start and / or end times of eating can also be adjusted using methods other than the "+" and "-" buttons. For example, please refer to... Figure 15D (a) The wearable device displays "Your eating behavior has been detected. Do you want to view this?"; it can also display a view button and an ignore button. When the wearable device detects an action on the view button, it displays as follows: Figure 15D (b) The interface includes breakfast-related information, such as the start and / or end times of the meal, and may also display icons 1500 and 1501. Icon 1500 corresponds to the start time of the meal, and icon 1501 corresponds to the end time. After receiving an operation on icon 1500, the wearable device enters edit mode for the start time of the meal, allowing the user to adjust it. For example, the wearable device displays... Figure 15DThe interface shown in (c) is in edit mode for the start time of feeding (e.g., the background color of the start time of feeding is black and the font color is white). Figure 15D (c) The user changes the start time of the meal to 7:10. After the wearable device receives an operation to exit the editing state (e.g., a click operation elsewhere on the screen), the meal start time exits the editing state, for example, displaying... Figure 15D The interface shown in (d) exits the editing mode for the food start time. Optionally, to let the user know that they have confirmed the food start time, such as... Figure 15D (d) Display a confirmation icon, such as "√", near the start time of the meal to indicate that the user has confirmed the start time of the meal.

[0228] As mentioned in the above embodiments, the wearable device provides a daily reminder to prompt the user to confirm their eating behavior for the day. Optionally, the wearable device can output the reminder at the first time of each day, which can be a fixed time, such as 10 PM, 11 PM, or midnight, etc., without limitation. Considering that daily reminders may also disturb the user and cause annoyance, in other embodiments, the user can set the wearable device not to prompt the user to confirm their eating behavior for a certain period of time. There are various ways to set this. Figure 15E Provide an example. For example... Figure 15E The wearable device displays a message saying "You have eaten today. Please check and verify." It can also display a "Ignore for 3 days" button and a "View" button. After receiving a request to press the "View" button, the wearable device can display... Figure 15C (b) Interface. After the wearable device receives an operation on the "Ignore 3 Days" button, it will not prompt the user to confirm the eating behavior for 3 days (including the day of the operation). On the 4th day, it will prompt the user to confirm the eating behavior again, that is, it will output again on the 4th day. Figure 15E The displayed message is as follows. It should be noted that... Figure 15E Taking the button "Ignore 3 days" as an example, it can be understood that it could also be "Ignore 5 days", "Ignore 1 week", "Ignore 1 month", "Ignore 3 months", etc., without limitation.

[0229] If a user wants to view their eating behavior on a specific historical day, they can select the corresponding date to view the eating behavior for that day. For example, please see... Figure 15F (a) The watch displays "You have recently eaten. Please check and verify," and also shows an ignore button and a view button. After detecting an action on the view button, the wearable device displays the following: Figure 15F (b) Date information: After the wearable device determines the date selected by the user, it displays the date information as shown in the image. Figure 15F(c) The interface includes information about eating behaviors detected within the date selected by the user. For example, Figure 15F In (c), after the wearable device detects the swipe-up operation, it can also display information such as lunch and dinner in addition to breakfast. Figure 15F (c) The interface also includes a "Replenish Food" button. After receiving an action on this button, food can be added. Optionally, when adding food, the type of food can be selected, such as breakfast, lunch, or dinner. Taking adding dinner as an example, the principle of adding dinner is the same as described above. Figure 15C The principle of adding in (f) is the same and will not be repeated. Optionally, you can also add the types of food to be eaten, such as rice, noodles, various vegetables, etc. Optionally, you can also add the amount of food, such as the weight of the food, such as how many grams of rice, how many grams of noodles, etc.

[0230] It should be noted that the above embodiment uses the example of a user confirming eating behavior on a wearable device. Optionally, if the wearable device is connected to a mobile phone, tablet, or other device, the eating behavior can also be confirmed on the mobile phone, tablet, or other device. For example, please refer to... Figure 15G (a) The phone displays an interface, which can be the interface of an application on the phone (e.g., the Huawei Health app). This interface includes a prompt message "We have detected recent eating behavior. Please click to view and verify," and also displays a date button. After receiving an operation on the date button, the phone displays as follows: Figure 15G (b) is the interface used to select a date. After selecting a date, the following is displayed: Figure 15G (c) The interface includes information about eating behaviors detected on the date selected by the user, such as the start and / or end time of breakfast, the start and / or end time of lunch, etc. Figure 15G (c) The interface also includes a "feed replenishment" button. After receiving an operation on this button, food can be added. The principle of adding food is the same as described above. Figure 15C The principle for adding elements is the same, so I won't repeat it here.

[0231] The second confirmation method involves the wearable device itself verifying the accuracy of the eating behavior detection. Considering that if a user does eat, their blood sugar will be higher, the wearable device can confirm the accuracy of the detection based on the user's blood sugar level. For example, after detecting eating, the wearable device collects the user's blood sugar; if the user's blood sugar is higher than a preset level, the detection of the eating behavior is considered accurate. In this method, the wearable device needs to collect the user's blood sugar. One possible approach is for the wearable device to connect to a blood glucose meter, which collects the user's blood sugar and sends the data to the wearable device. For more information on blood glucose meters, please refer to Example 4 above. In another possible implementation, the blood glucose meter can directly determine whether the user has engaged in eating behavior and send this result to the wearable device. The wearable device can then combine the result from the blood glucose meter to further confirm the accuracy of the eating behavior detection.

[0232] Therefore, the accuracy of the eating behavior detection can be confirmed through the first or second confirmation method described above. For example, if the user changes the start and / or end time of eating, or determines that no food was eaten, it indicates that the eating behavior detection is inaccurate, and the model can be retrained. This part has been described above and will not be repeated.

[0233] After detecting a user's eating behavior, the wearable device can display an eating behavior icon to indicate the presence of eating. If multiple eating behaviors are detected, multiple eating behavior icons will be displayed. Therefore, users can know whether their eating is regular, such as whether they eat three meals a day, by the number of eating behavior icons. The eating behavior icon can be displayed on any interface of the wearable device, such as the main interface. Optionally, if the wearable device is connected to a mobile phone, tablet, or other device, it can also be displayed on any interface (e.g., the main interface) of that device. For example, if the wearable device is a watch, the eating behavior icon can be displayed on the watch face.

[0234] Optionally, after displaying the eating behavior identifier, the wearable device can also display the time information of the eating behavior indicated by the identifier. For example, if the eating behavior occurs between 7:00 and 7:20, the time information 7:00-7:20 will be displayed. Alternatively, it can only display the start time of eating at 7:00 or only the end time of eating at 7:20. If multiple eating behavior identifiers are displayed, the time information of the corresponding eating behavior can be displayed for each identifier.

[0235] The following example uses a watch to illustrate an instance of a watch displaying an indicator of eating behavior.

[0236] For example, see Figure 16The watch face includes time markings to indicate different times of day. For example, as... Figure 16 The watch face includes 24 time markers, each corresponding to one hour, for a total of 24 hours. After detecting eating behavior, the watch can display an eating behavior indicator on the watch face, positioned close to the time marker corresponding to the time of the eating behavior. For example, if the watch detects the user eating and determines the start time to be 8:00 and the end time to be 8:20, then the eating behavior indicator will be displayed near the 8:00 time marker on the watch face. Figure 16 The watch can display an eating behavior indicator (e.g., a picture of a bowl of rice) 1600. For example, if the watch detects a user eating and determines the start time of the eating behavior is 12:00 and the end time is 12:20, then the eating behavior indicator will be displayed near the 12:00 mark on the watch face. Figure 16 The device includes an eating behavior icon 1601 (e.g., a picture of a bowl of rice). Therefore, users can determine whether their eating habits are regular, such as whether they eat three meals a day, by using the eating behavior icon on the watch face.

[0237] Optionally, users can view, correct, and confirm eating behaviors recorded by the wearable device using eating behavior identifiers. For example, please see... Figure 17 In (a), the watch face includes eating behavior identifiers 1600 and 1601. After receiving an operation on eating behavior identifier 1600, the watch displays relevant information about the eating behavior corresponding to that identifier, such as... Figure 17 The interface in (b) displays the start and / or end times of the feeding behavior corresponding to feeding behavior identifier 1600. The function of this interface is explained in the preceding text. Figure 15C This has already been explained in detail and will not be repeated here. Optional, Figure 17 The interface shown in (b) can also include other meal supplement buttons. After receiving an operation on this button, the wearable device can add food. Optionally, when adding food, the user can select the type of food, such as breakfast, lunch, or dinner. Taking adding dinner as an example, the principle of adding dinner is the same as described above. Figure 15C The principle for adding in (f) is the same and will not be repeated. Optionally, you can also add the type of food (e.g., rice, noodles) and the weight of the food. Figure 17 In (a), after receiving an operation on the eating behavior identifier 1601, the watch displays relevant information about the eating behavior corresponding to the eating behavior identifier 1601, such as... Figure 17The interface in (c) displays the start and / or end times of the feeding behavior corresponding to feeding behavior identifier 1601. The function of this interface is explained in the preceding text. Figure 15C This has already been explained in detail and will not be repeated here. Optional, Figure 17 The interface shown in (c) can also include other meal supplement buttons. After receiving an operation on this button, the wearable device can add food. Optionally, when adding food, the user can choose the type of food, such as breakfast, lunch, or dinner. Taking adding dinner as an example, the principle of adding dinner is the same as described above. Figure 15C The principle for adding in (f) is the same and will not be repeated. Optionally, you can also add the type of food (e.g., rice, noodles) and the weight of the food.

[0238] Figure 17 This explains that users can view the eating behavior process through the eating behavior icon. If the wearable device detects that the user has modified the eating behavior, such as changing the start and / or end times, then the wearable device will adjust the display position of the eating behavior icon based on the modification result. For example, Figure 17 In (a), the eating behavior identifier 1600 is displayed near the 8:00 time scale (because the starting time of the eating behavior corresponding to the eating behavior identifier 1600 is 8:00 and the ending time is 8:20). If the user changes the starting time to 9:00 and the ending time to 9:30, the display position of the eating behavior identifier 1600 will be adjusted to be near the 9:00 time scale.

[0239] In some embodiments, before the user confirms the eating behavior, the eating behavior indicator may be displayed in a first display mode, which indicates that the eating behavior has not been confirmed by the user. For example, the first display mode may be a first pattern, with the first pattern having a first color and a first transparency. Continuing... Figure 17 Taking (a) as an example, both eating behavior identifiers 1600 and 1601 are displayed in a first display mode (e.g., a picture of a bowl of rice, and the picture is white), indicating that the eating behavior has not been confirmed by the user. After the user confirms the eating behavior, the eating behavior identifier adopts a second display mode, which is different from the first display mode. The second display mode is used to indicate that the eating behavior has been confirmed by the user. For example, the second display mode can be a second picture, with a second color and a second transparency. The second picture and the first picture can be the same picture or different pictures. If they are the same picture, for example, both are pictures of a bowl of rice, then the second color and the first color can be different, and / or, the second transparency and the first transparency can be different.

[0240] For example, such as Figure 18In (a), the eating behavior indicator 1600 is displayed as a bowl of rice in white, indicating that the user has not confirmed the action. After the watch detects that the user clicked on the eating behavior indicator 1600, it displays... Figure 18 In interface (b), after the watch detects the operation on the confirm button, it returns as shown below. Figure 18 In the interface (d), the eating behavior indicator 1600 is displayed as a bowl of rice in black, indicating that the user has confirmed it. Figure 18 In the interface shown in (c), the eating behavior indicator 1601 is displayed as a bowl of rice in white, indicating that the user has not confirmed the action. After the watch detects that the user has clicked on the eating behavior indicator 1601, it displays... Figure 18 In the interface shown in (d), after the watch detects the operation on the confirm button, it returns as follows: Figure 18 In the interface (e), the eating behavior indicator 1601 is displayed as a bowl of rice in black, indicating that the user has confirmed it.

[0241] Optionally, the wearable device can also detect the user's exercise behavior, such as running, skipping rope, brisk walking, gymnastics, and other various exercise behaviors. After detecting the user's exercise behavior, the wearable device can display an exercise behavior icon to indicate the presence of exercise behavior. If multiple exercise behaviors are detected, multiple exercise behavior icons will be displayed. The exercise behavior icon can be displayed on any interface of the wearable device, such as the main interface of the wearable device. Optionally, if the wearable device is connected to a mobile phone, tablet, or other device, it can also be displayed on any interface of the mobile phone, tablet, or other device (e.g., the main interface). For example, if the wearable device is a watch, the exercise behavior icon can be displayed on the watch face.

[0242] Optionally, after displaying the exercise behavior identifier, the wearable device can also display the time information of the exercise behavior indicated by the identifier. For example, if the exercise behavior occurs between 17:00 and 17:20, the time information 17:00-17:20 will be displayed. Alternatively, only the start time of the exercise (17:00) or only the end time of the exercise (17:20) can be displayed. If multiple exercise behavior identifiers are displayed, the time information of the corresponding exercise behavior can be displayed for each identifier.

[0243] The following example uses a watch to illustrate an instance of a watch displaying activity indicators.

[0244] For example, after the watch detects movement, it displays a movement indicator on the watch face, and the indicator is positioned close to the time scale corresponding to the time of the movement. Continuing... Figure 16For example, if the watch detects movement and determines that the start time of the movement is 17:00 and the end time is 17:20, then a movement indicator will be displayed near the 17:00 time mark on the watch face. Figure 16 The motion behavior identifier 1602 (e.g., human figure).

[0245] Optionally, users can view, correct, and confirm the exercise behavior recorded by the wearable device using exercise behavior tags. For example, please see... Figure 19 In (a), the watch face includes a motion behavior identifier 1602. After receiving an operation on the motion behavior identifier 1602, the watch displays relevant information about the motion behavior corresponding to the identifier 1602, such as... Figure 19 The interface in (b) displays the start and / or end times of the motion behavior corresponding to motion behavior identifier 1602. Users can adjust the start and / or end times of the motion. Optionally, Figure 19 The interface shown in (b) also includes other sports supplement buttons. After the wearable device receives an operation on the button, it can add sports activities. For example, it can set the start time and / or end time of the sports activity. Optionally, it can also add sports events, such as running, swimming, skipping rope, etc.

[0246] Figure 19 This describes the process by which users view their activity through activity icons. If the wearable device detects that the user has corrected the activity, such as changing the start and / or end times, the device will adjust the display position of the activity icons based on the correction. For example, Figure 19 In (a), the motion behavior identifier 1602 is displayed near the 17:00 time scale (because the start time of the motion behavior corresponding to the motion behavior identifier 1602 is 17:00). If the user corrects the start time to 18:00, the display position of the motion behavior identifier 1602 will be adjusted to be near the 18:00 time scale.

[0247] In some embodiments, before the user confirms the exercise behavior, the exercise behavior identifier may be displayed in a third display mode. This third display mode indicates that the exercise behavior has not been confirmed by the user. For example, the third display mode may be a third pattern, with the third pattern having a third color and a third transparency. Continuing... Figure 19Taking (a) as an example, the motion behavior identifier 1602 is displayed in a first display mode (e.g., a human figure in white), indicating that the motion behavior has not been confirmed by the user. After the user confirms the motion behavior, the motion behavior identifier is displayed in a fourth display mode, which is different from the third display mode. The fourth display mode is used to indicate that the motion behavior has been confirmed by the user. For example, the fourth display mode can be a fourth pattern, with a fourth color and a fourth transparency. The fourth pattern and the third pattern can be the same pattern or different patterns. If they are the same pattern, for example, both are human figures, then the fourth color and the third color can be different, and / or the fourth transparency and the third transparency can be different. For example, as shown in... Figure 19 In (b) of the above, after the watch detects the operation on the confirm button, it returns to the state shown in the image. Figure 19 The interface in (c) shows the motion behavior indicator 1602 in a fourth display mode (e.g., a human figure in black), indicating that the user has confirmed it.

[0248] Optionally, the wearable device can also detect the user's sleep behavior, such as sleep onset time and / or sleep wake-up time. After detecting the user's sleep behavior, the wearable device can display a sleep behavior icon to indicate the presence of sleep behavior. If multiple sleep behaviors are detected, multiple sleep behavior icons are displayed. The sleep behavior icon can be displayed on any interface of the wearable device, such as the main interface. Optionally, if the wearable device is connected to a mobile phone, tablet, or other device, it can also be displayed on any interface (e.g., the main interface) of the mobile phone, tablet, or other device. For example, if the wearable device is a watch, the sleep behavior icon can be displayed on the watch face.

[0249] Optionally, after displaying a sleep behavior identifier, the wearable device can also display the time information of the occurrence of the sleep behavior indicated by the identifier. For example, if the sleep behavior occurs between 21:00 and 6:00, the time information 21:00-6:00 will be displayed. Alternatively, it can only display the sleep start time of 21:00 or only the sleep end time of 6:00. If multiple sleep behavior identifiers are displayed, the time information of the occurrence of the corresponding sleep behavior can be displayed for each identifier.

[0250] The following example uses a watch to illustrate how a watch can display sleep behavior indicators.

[0251] For example, after detecting sleep behavior, the watch displays a sleep behavior indicator on the watch face, and the indicator is positioned close to the time scale corresponding to the occurrence of the sleep behavior. Continuing... Figure 16For example, if the watch detects a user's sleep behavior and determines that the sleep behavior started at 21:00 and ended at 6:00, then a sleep behavior indicator will be displayed near the 21:00 time mark on the watch face. Figure 16 Sleep behavior markers 1603 (e.g., moon pattern).

[0252] Optionally, users can view, correct, and confirm sleep behaviors recorded by wearable devices using sleep behavior tags. For example, please see... Figure 20 In (a), the watch face displays sleep behavior indicator 1603. After receiving an operation related to sleep behavior indicator 1603, the watch displays relevant information about the sleep behavior corresponding to sleep behavior indicator 1603, such as... Figure 20 The interface in (b) displays the start and / or end times of the sleep behavior corresponding to sleep behavior identifier 1603, and the user can adjust the sleep start and / or sleep end times. Optionally, Figure 20 The interface shown in (b) may also include other sleep supplement buttons. After the wearable device receives an operation on the button, it can add sleep behaviors, such as setting the start time and / or end time of the sleep behavior.

[0253] Figure 20 This describes how users view their sleep behavior through sleep behavior icons. If the wearable device detects that the user has modified the sleep behavior, such as changing the start and / or end times, the device will adjust the display position of the sleep behavior icons based on the modification. For example, Figure 20 In (a), the sleep behavior indicator 1603 is displayed near the 21:00 time scale (because the start time of the sleep behavior corresponding to the sleep behavior indicator 1603 is 21:00). If the user changes the start time to 22:00, the display position of the sleep behavior indicator 1603 is adjusted to be near the 22:00 time scale.

[0254] In some embodiments, before the user confirms the sleep behavior, the sleep behavior indicator may be displayed in a fifth display mode, which indicates that the sleep behavior has not been confirmed by the user. For example, the fifth display mode may be a fifth pattern, with the fifth color and the fifth transparency. Continuing... Figure 20Taking (a) as an example, the sleep behavior identifier 1603 is displayed in the fifth display mode (e.g., a moon pattern in white), indicating that the sleep behavior has not been confirmed by the user. After the user confirms the sleep behavior, the sleep behavior identifier is displayed in the sixth display mode, which is different from the fifth display mode. The sixth display mode is used to indicate that the sleep behavior has been confirmed by the user. For example, the sixth display mode can be a sixth pattern, with the sixth color and the sixth transparency. The sixth pattern and the fifth pattern can be the same pattern or different patterns. If they are the same pattern, such as both being moon patterns, then the sixth color and the fifth color can be different, and / or, the sixth transparency and the fifth transparency can be different. For example, as shown in... Figure 20 In (b) of the above, after the watch detects the operation on the confirm button, it returns to the state shown in the image. Figure 20 The interface in (c) shows the sleep behavior indicator 1603 in a sixth display mode (e.g., a moon pattern in black), indicating that the user has confirmed it.

[0255] After a wearable device detects multiple user eating behaviors, it can generate statistical results of eating behavior based on these multiple behaviors. Optionally, the multiple eating behaviors can be multiple eating behaviors collected within a statistical period. For example, a statistical period can be one day, three days, one week, one month, one year, etc., without limitation. Taking a statistical period of one week as an example, the statistical results can include one or more of the following within a week: daily eating frequency, average daily eating frequency, daily eating frequency, average eating speed per meal, and average food intake per meal. Among them, daily eating frequency is, for example, the number of times eating each day from Monday to Sunday within a week. Average daily eating frequency is, for example, the average number of times eating each day within a week. Daily eating frequency is used to indicate the time interval between two meals within a day, for example, the time interval between breakfast and lunch, and the time interval between lunch and dinner. Average daily eating frequency is used to indicate the average daily eating frequency within a week, for example, the average time interval between breakfast and lunch within a week, and the average time interval between lunch and dinner within a week. Average eating speed per meal indicates the average eating speed per meal over a week. Eating speed per meal can be the duration from the start to the end of each meal. Average food intake per meal indicates the average food intake per meal over a week. Food intake per meal can be determined based on the types and weight of food consumed at each meal. These factors can be manually added by the user, as described above (e.g., ...). Figure 15A and Figure 15B As described in [the document], I will not repeat it here.

[0256] Optionally, the eating behavior statistics may also include post-eating activity and / or post-eating sleep patterns. For example, post-eating activity includes the average time interval between the end of eating and the start of activity. Post-eating sleep includes the average time interval between the end of eating and the start of sleep. It should be noted that the eating behavior statistics can be displayed on a wearable device, or on other devices such as mobile phones or tablets connected to the wearable device. For example, the eating behavior statistics can be displayed on a mobile phone.

[0257] For example, see Figure 21 In (a) of the example, the mobile phone displays a statistical result of the daily number of meals. This result is located in a coordinate system, with the horizontal axis representing the date and the vertical axis representing the number of meals per day. Optionally, the mobile phone can also display a baseline in the coordinate system. For example, the vertical axis of this baseline might be 3 times, which represents the number of meals per day that meets the health standard. Therefore, users can quickly determine whether their meal frequency meets the health standard using this baseline. For example, Figure 21 In (a) of the data, points above the baseline represent excessive eating, while points below the baseline represent insufficient eating.

[0258] For example, continue as Figure 21 In (a) of the example, the phone also displays statistics on the average eating speed per meal, displayed on a coordinate system where the horizontal axis represents the date and the vertical axis represents the average eating speed per meal. Optionally, the phone can also display a baseline on the coordinate system; for example, the vertical axis of this baseline might correspond to 20 minutes, representing a healthy eating speed. Therefore, users can quickly determine whether their eating speed meets healthy standards using this baseline. For example, Figure 21 In (a) of the data, points above the baseline represent eating too fast, and points below the baseline represent eating too slow.

[0259] For example, see Figure 21 In (b) of the example, the mobile phone displays statistical results of the average time interval between the end of eating and exercise. These results are displayed on a coordinate system where the horizontal axis represents the date and the vertical axis represents the average time interval between the end of eating and exercise. Optionally, the wearable device can also display a baseline on the coordinate system. For example, the vertical axis of this baseline might be 30 minutes, representing the time interval between the end of eating and exercise that meets health standards. Therefore, users can quickly determine whether their post-meal exercise behavior meets health standards using this baseline. For example, Figure 21 In (b) of the table, points above the baseline indicate that exercise occurred more than 30 minutes after eating, which is more in line with health standards. Points below the baseline indicate that exercise occurred less than 30 minutes after eating, which is not in line with health standards.

[0260] For example, continue as Figure 21 In (b) of the example, the phone also displays statistics on the average time interval between the end of eating and sleep. These statistics are displayed on a coordinate system, with the horizontal axis representing the date and the vertical axis representing the average time interval between the end of eating and sleep. Optionally, the wearable device can also display a baseline on this coordinate system. For example, the vertical axis of this baseline might be 180 minutes, representing the time interval between the end of eating and sleep that meets health standards. Therefore, users can use this baseline to quickly determine whether their sleep behavior after eating meets health standards. For example, Figure 21 In (b) of the table, points above the baseline indicate that it takes more than 180 minutes after eating before going to sleep, which is more in line with health standards. Points below the baseline indicate that it takes less than 180 minutes after eating before going to sleep, which does not meet health standards.

[0261] Wearable devices can generate user eating profiles based on statistical results of user eating behavior. These profiles include an overview of the user's eating habits. For example, based on the statistical results of the number of meals eaten per day, the wearable device can determine the regularity of eating. For instance, if the statistical results of the number of meals eaten per day indicate that the user frequently eats only two meals a day, then the user's meal times are considered irregular. For example, based on the statistical results of the average eating speed per meal, the wearable device can determine the average eating speed (also known as the average eating time), and thus determine whether the user eats too quickly. For instance, if the statistical results of the average eating speed per meal indicate that the user's average eating speed is 5 minutes, then the user's eating speed is considered too fast. Furthermore, based on the statistical results of the average time interval between the end of eating and exercise, the wearable device can determine that the user frequently exercises immediately after eating. Alternatively, based on the statistical results of the average time interval between the end of eating and sleep, the wearable device can determine that the user frequently falls asleep immediately after eating. Optionally, after determining the user's eating profile, the wearable device can provide suggestions for improvement. For example, if the user's meal times are irregular, the wearable device can suggest that the user eat more regularly. For example, if a user eats too quickly, suggest they slow down. If a user exercises immediately after eating, suggest they wait half an hour. If a user goes to sleep immediately after eating, suggest they wait a while before sleeping.

[0262] After obtaining an eating profile, the wearable device can output an eating profile, and optionally, it can also output improvement suggestions based on the eating profile. For example, please see... Figure 22A (a) The wearable device displays the prompt message "Please view your eating profile," and can also display an ignore button and a view button. When the wearable device detects an action on the view button, it displays the following: Figure 22A(b) The interface includes a user's eating profile and improvement suggestions. For example, the interface may display: Your meals are irregular, you often skip breakfast, and we suggest you eat more regularly. Or, the interface may also display: Your average eating speed is 5 minutes, eating too fast is not good for your digestive health, and we suggest you slow down your eating speed. Figure 22A In (b), when the wearable device receives an up swipe operation, it displays as follows: Figure 22A (c) The interface displays: You often engage in high-intensity exercise immediately after eating. It is recommended to wait at least half an hour after eating before exercising to avoid affecting your gastrointestinal health. Alternatively, the interface displays: You often go to sleep immediately after eating. It is recommended to wait a while after eating before going to sleep.

[0263] Optionally, the eating profile can include weekday eating profiles and holiday eating profiles. Weekday eating profiles are derived from statistical results of eating behavior on weekdays, such as Monday to Friday. Holiday eating profiles are derived from statistical results of eating behavior during holidays (such as Saturday and Sunday, May Day, National Day, etc.). Optionally, the improvement suggestions for weekday eating profiles can differ from those for holiday eating profiles; or, improvement suggestions can be provided only for weekday eating profiles, without requiring suggestions for holiday eating profiles. For example, improvement suggestions for weekday eating profiles can be more stringent (e.g., suggestions are provided if meal times are irregular, such as eating breakfast or dinner too late), while suggestions for holiday eating profiles can be more lenient (e.g., no improvement suggestions are required if meal times are irregular, such as eating breakfast or dinner too late), which helps improve the user experience.

[0264] Optionally, the eating profile can also include the types of food the user eats, such as rice and noodles. For example, the eating profile may include prompts to remind the user of the types of food they frequently consume, such as "We have detected that your food intake is usually noodles." Optionally, suggestions may also be provided, such as "We suggest you supplement your diet with vegetables and protein."

[0265] Figure 22A Taking the output of an eating image from a wearable device as an example, optionally, if the wearable device is connected to other devices such as mobile phones or tablets, the eating image can also be output through those other devices. For example, please refer to... Figure 22B The phone displays an interface, which can be the interface of an application on the phone (such as the Huawei Health app). This interface includes the user's eating profile, and optionally, it can also include improvement suggestions based on the eating profile.

[0266] The above embodiments illustrate the process by which wearable devices determine a user's eating profile. In some embodiments, the wearable device can also conduct a questionnaire survey on the user's eating profile in advance, such as investigating whether the user's meals are regular, their eating speed, their post-meal exercise, and their post-meal sleep. The eating profile questionnaire survey can be conducted on the wearable device, or on a mobile phone or tablet connected to the wearable device. Taking a mobile phone as an example, please refer to [link to relevant documentation]. Figure 22C (a) The mobile phone displays an interface that includes a questionnaire about the user's eating profile, such as whether their meals are regular and their eating speed. After receiving an action (e.g., page turning), the mobile phone displays the following... Figure 22C (b) This interface includes information such as post-eating activity and post-eating sleep patterns. Users can fill in their own eating profile. It is understood that the user's input eating profile may or may not match the user's eating profile detected by the wearable device. If they match, the detected eating profile is output; if they do not match, eating behavior can be detected multiple times to increase the sample size of the statistical results, thereby improving the accuracy of the statistical results and ultimately the accuracy of the eating profile. Afterward, the detected eating profile can be output.

[0267] Please see Figure 23 This is a schematic flowchart illustrating another method for detecting eating behavior provided in an embodiment of this application. This method can be applied to wearable devices. Figure 23 The process includes the following steps:

[0268] Step 2301: Collect data.

[0269] For example, wearable devices include various sensors that collect data.

[0270] Step 2302: Use the collected data as input parameters for the hand-wearing detection model, and run the detection model to determine the hand being worn.

[0271] For example, the data collected by the wearable device includes the watch face orientation and / or a face image. The watch face orientation and / or face image serve as input parameters for a hand detection model, which is then run to identify the wearing hand. For an explanation of the principles behind the hand detection model, please refer to the preceding description.

[0272] Step 2303: Use the collected data as input parameters for the eating-dominant hand detection model, and run the detection model to determine the eating-dominant hand.

[0273] For example, the data collected by wearable devices includes at least one of the following: current geographical location, surrounding environment data, and the user's driving path. This data serves as input parameters for a dominant eating hand detection model, which is then run to determine the dominant eating hand. For an explanation of the principles behind the dominant eating hand detection model, please refer to the preceding description.

[0274] Step 2304: Output a prompt message to prompt the user to confirm the test results of the wearing hand and / or the dominant hand used for eating.

[0275] For example, step 2304 can be achieved through the preceding text. Figure 8A Implementation. (Details omitted.)

[0276] Step 2305: Determine whether the user's confirmation result is consistent with the model's output result. If they are consistent, proceed to step 2306. If they are inconsistent, retrain the wearing hand detection model and / or the eating dominant hand detection model (i.e., self-learning), and proceed to step 2306 based on the user's confirmed wearing hand and eating dominant hand.

[0277] Step 2306: Determine if the wearing hand and the eating hand are the same. If they are not the same, proceed to step 2307. If they are the same, proceed to step 2308.

[0278] Step 2307: Use the first eating behavior detection model to detect the user's eating behavior.

[0279] For details on the first type of eating behavior detection model, please refer to the previous description. For example, the first type of eating behavior detection model includes Algorithm Model 1 and / or Algorithm Model 2 mentioned above.

[0280] Step 2308: Use the second eating behavior detection model to detect the user's eating behavior.

[0281] For information on the second type of eating behavior detection model, please refer to the previous description. For example, the second type of eating behavior detection model includes algorithm model 3 mentioned above.

[0282] Step 2309: Output the confidence score, which is the probability that the eating behavior exists.

[0283] Step 2310: If the confidence level is higher than the threshold, then it is determined that there is eating behavior.

[0284] Step 2311: Confirm the eating behavior.

[0285] Optionally, the confirmation method can include two types, namely the first confirmation method and the second confirmation method mentioned above, which will not be repeated.

[0286] Step 2312: Determine whether the confirmation result is consistent with the model detection result. If they are consistent, proceed to step 2313. If they are inconsistent, retrain the corresponding eating behavior detection model (i.e., self-learning).

[0287] It should be noted that if the confirmation result is inconsistent with the model detection result, the confirmation result shall prevail. For example, the confirmation result can be included in the statistical results of eating behavior.

[0288] Step 2313: Generate eating behavior statistics.

[0289] For example, the statistical results may include one or more of the following: daily meal frequency, average daily meal frequency, average daily meal rate, average eating speed per meal, and average food intake per meal. Optionally, such as... Figure 23 In step 2313, the wearable device can further combine the user's exercise and / or sleep patterns to generate statistical results on eating behavior. For example, the wearable device can detect how long after a user exercises and / or falls asleep after eating, and generate statistical results based on the detection results. That is, the statistical results can also include post-eating exercise (e.g., the average time interval from the end of eating to exercise) and / or post-eating sleep (e.g., the average time interval from the end of eating to sleep). For more information on eating behavior statistical results, please refer to the preceding text. Figure 21 The relevant descriptions will not be repeated here.

[0290] Step 2314: Generate an eating profile based on the statistical results of eating behavior.

[0291] Optional, such as Figure 23 In step 2314, the wearable device can further generate an eating profile by combining the user's exercise and / or sleep data. One possibility is that the eating behavior statistics obtained in step 2313 do not incorporate the user's exercise and / or sleep data. In this case, in step 2314, the wearable device can generate an eating profile based on the eating statistics obtained in step 2313, combined with the user's exercise and / or sleep data. Therefore, the eating profile includes: post-eating exercise (e.g., how long after eating does one exercise) and / or post-eating sleep (e.g., how long after eating does one fall asleep). Another possibility is that the eating behavior statistics obtained in step 2313 already incorporate the user's exercise and / or sleep data. In this case, in step 2314, the wearable device does not need to detect the user's exercise and / or sleep data; it can directly generate the eating profile based on the statistical results obtained in step 2313. For more information on eating profiles, please refer to the previous text. Figure 22A or Figure 22B Related descriptions.

[0292] Next, the electronic devices involved in the embodiments of this application will be described.

[0293] For example, see Figure 24 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device can be a wearable device, such as the various types of wearable devices listed above. Figure 24 The electronic device may include: a data acquisition module, a module for detecting the wearing hand and the dominant eating hand, a module for detecting eating behavior, a user interaction module, a self-learning module, and an improvement suggestion module. It should be noted that the user interaction module, self-learning module, and improvement suggestion module are optional and not limited to any particular module. Figure 24 These three modules are represented by dashed lines.

[0294] The data acquisition module is used to collect data. For example, the data acquisition module includes various sensors, including but not limited to accelerometers, gyroscopes, image sensors, sound sensors, etc.

[0295] The module for detecting the wearing hand and the dominant eating hand is used to detect these two hands. For example, this module includes a model for detecting the wearing hand and a model for detecting the dominant eating hand. The detection principles of these models have been described previously and will not be repeated here. After detecting the wearing hand and / or the dominant eating hand, a prompt message can be output through the user interaction module for user confirmation. If the user's confirmation result is inconsistent with the detection result, the detection model can be retrained through the self-learning module.

[0296] The eating behavior detection module is used to detect user eating behavior. For example, the eating behavior detection module includes eating behavior detection models, including the first and second eating behavior detection models mentioned above. The eating behavior module can determine whether the hand wearing the device is the same as the dominant hand used for eating. If they are different, the first eating behavior detection model is used to detect the user's eating behavior; if they are the same, the second eating behavior detection model is used. For details on the first and second eating behavior detection models, please refer to the previous descriptions. Figure 24 The eating behavior detection module can also include an eating probability analysis module, used to analyze the confidence level output by the eating behavior detection model, such as determining whether the confidence level is higher than a threshold; if so, eating behavior is confirmed. After confirming the existence of eating behavior, a prompt message can be output through a user interaction module for user confirmation. If the user confirmation result is inconsistent with the detection result, the detection model can be retrained through a self-learning module. Figure 24 The eating behavior detection module also includes an eating behavior analysis module, which is used to perform statistical analysis on multiple eating behaviors when they are detected, and obtain statistical results of the eating behavior. For example... Figure 24 The eating behavior detection module also includes an eating profile module, which is used to generate eating profiles based on the statistical results of eating behavior.

[0297] The user interaction module outputs detection results of the wearing hand and / or the dominant hand for eating for user confirmation. Alternatively, after detecting eating behavior, it can output prompts for user confirmation. The user interaction module is connected to the sub-learning module; if the user confirmation result is inconsistent with the model detection result, the detection model is retrained.

[0298] The self-learning module is used to train various detection models to improve their accuracy.

[0299] The improvement suggestion module is used to generate improvement suggestions based on the eating profile and output the improvement suggestions through the user interaction module.

[0300] Please see Figure 25 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device can be a wearable device, such as the various types of wearable devices listed above. Figure 25 As shown, the electronic device may include a processor 110, an external memory interface 120, an internal memory 121, a universal serial bus (USB) interface 130, a charging management module 140, a power management module 141, a battery 142, an antenna 1, an antenna 2, a mobile communication module 150, a wireless communication module 160, an audio module 170, a speaker 170A, a receiver 170B, a microphone 170C, a headphone jack 170D, a sensor module 180, buttons 190, a motor 191, an indicator 192, a camera 193, a display screen 194, and a subscriber identification module (SIM) card interface 195, etc. The sensor module 180 may include a pressure sensor 180A, a gyroscope sensor 180B, a barometric pressure sensor 180C, a magnetic sensor 180D, an accelerometer sensor 180E, a distance sensor 180F, a proximity sensor 180G, a fingerprint sensor 180H, a temperature sensor 180J, a touch sensor 180K, an ambient light sensor 180L, a bone conduction sensor 180M, etc.

[0301] Processor 110 may include one or more processing units, such as application processors (APs), modem processors, graphics processing units (GPUs), image signal processors (ISPs), controllers, memory, video codecs, digital signal processors (DSPs), baseband processors, and / or neural network processing units (NPUs). Different processing units may be independent devices or integrated into one or more processors. The controller may serve as the nerve center and command center of the electronic device. The controller can generate operation control signals based on instruction opcodes and timing signals to control instruction fetching and execution. Processor 110 may also include memory for storing instructions and data. In some embodiments, the memory in processor 110 is a cache memory. This memory can store instructions or data that processor 110 has just used or is recurring. If processor 110 needs to reuse an instruction or data, it can directly retrieve it from the memory. This avoids repeated access, reduces the waiting time of processor 110, and thus improves system efficiency.

[0302] In some embodiments, the processor 110 may execute the eating behavior detection method provided in the embodiments of this application. For example, when the electronic device is a wearable device, the processor 110 may determine the hand on which the user is currently wearing the wearable device and determine the user's dominant hand for eating; if the wearing hand is different from the dominant hand for eating, the user's eating behavior is detected using a first detection method; if the wearing hand is the same as the dominant hand for eating, the user's eating behavior is detected using a second detection method; wherein the first detection method and the second detection method are different.

[0303] In some embodiments, the processor 110 may include one or more interfaces. Interfaces may include an inter-integrated circuit (I2C) interface, an inter-integrated circuit sound (I2S) interface, a pulse code modulation (PCM) interface, a universal asynchronous receiver / transmitter (UART) interface, a mobile industry processor interface (MIPI), a general-purpose input / output (GPIO) interface, a subscriber identity module (SIM) interface, and / or a universal serial bus (USB) interface, etc.

[0304] The I2C interface is a bidirectional synchronous serial bus, including a serial data line (SDA) and a serial clock line (SCL). In some embodiments, the processor 110 may include multiple I2C buses. The processor 110 can couple to the touch sensor 180K, charger, flash, camera 193, etc., through different I2C bus interfaces. For example, the processor 110 can couple to the touch sensor 180K through the I2C interface, enabling the processor 110 and the touch sensor 180K to communicate through the I2C bus interface, thereby realizing the touch function of the electronic device 100.

[0305] The I2S interface can be used for audio communication. In some embodiments, the processor 110 may include multiple I2S buses. The processor 110 can be coupled to the audio module 170 via the I2S bus to enable communication between the processor 110 and the audio module 170. In some embodiments, the audio module 170 can transmit audio signals to the wireless communication module 160 via the I2S interface to enable the function of answering phone calls through a Bluetooth headset.

[0306] The PCM interface can also be used for audio communication, sampling, quantizing, and encoding analog signals. In some embodiments, the audio module 170 and the wireless communication module 160 can be coupled via the PCM bus interface. In some embodiments, the audio module 170 can also transmit audio signals to the wireless communication module 160 via the PCM interface, enabling the function of answering phone calls through a Bluetooth headset. Both the I2S interface and the PCM interface can be used for audio communication.

[0307] The UART interface is a universal serial data bus used for asynchronous communication. This bus can be a bidirectional communication bus. It converts the data to be transmitted between serial and parallel communication. In some embodiments, the UART interface is typically used to connect the processor 110 and the wireless communication module 160. For example, the processor 110 communicates with the Bluetooth module in the wireless communication module 160 via the UART interface to implement Bluetooth functionality. In some embodiments, the audio module 170 can transmit audio signals to the wireless communication module 160 via the UART interface to enable music playback through Bluetooth headphones.

[0308] The MIPI interface can be used to connect the processor 110 to peripheral devices such as the display screen 194 and the camera 193. The MIPI interface includes a camera serial interface (CSI) and a display serial interface (DSI). In some embodiments, the processor 110 and the camera 193 communicate via the CSI interface to enable the electronic device 100 to capture images. The processor 110 and the display screen 194 communicate via the DSI interface to enable the electronic device 100 to display images.

[0309] The GPIO interface can be configured via software. It can be configured as a control signal or a data signal. In some embodiments, the GPIO interface can be used to connect the processor 110 to a camera 193, a display screen 194, a wireless communication module 160, an audio module 170, a sensor module 180, etc. The GPIO interface can also be configured as an I2C interface, an I2S interface, a UART interface, a MIPI interface, etc.

[0310] USB port 130 is a USB standard compliant interface, specifically a Mini USB port, Micro USB port, USB Type-C port, etc. USB port 130 can be used to connect a charger to charge electronic device 100, and can also be used for data transfer between electronic device 100 and peripheral devices. It can also be used to connect headphones for audio playback. This interface can also be used to connect other electronic devices, such as AR devices.

[0311] It is understood that the interface connection relationships between the modules illustrated in the embodiments of the present invention are merely illustrative and do not constitute a structural limitation on the electronic device 100. In other embodiments of this application, the electronic device 100 may also employ different interface connection methods or combinations of multiple interface connection methods as described in the above embodiments.

[0312] The wireless communication function of the electronic device can be implemented through antenna 1, antenna 2, mobile communication module 150, wireless communication module 160, modem processor, and baseband processor. Antenna 1 and antenna 2 are used to transmit and receive electromagnetic wave signals. Each antenna in the electronic device can be used to cover one or more communication frequency bands. Different antennas can also be reused to improve antenna utilization. For example, antenna 1 can be reused as a diversity antenna for a wireless local area network. In some other embodiments, the antenna can be used in conjunction with a tuning switch.

[0313] The mobile communication module 150 can provide solutions for wireless communication applications including 2G / 3G / 4G / 5G in electronic devices. The mobile communication module 150 may include at least one filter, switch, power amplifier, low noise amplifier (LNA), etc. The mobile communication module 150 can receive electromagnetic waves via antenna 1, and perform filtering, amplification, and other processing on the received electromagnetic waves before transmitting them to a modem processor for demodulation. The mobile communication module 150 can also amplify the signal modulated by the modem processor and convert it into electromagnetic waves for radiation via antenna 1. In some embodiments, at least some functional modules of the mobile communication module 150 may be housed in processor 110. In some embodiments, at least some functional modules of the mobile communication module 150 and at least some modules of the processor 110 may be housed in the same device.

[0314] The wireless communication module 160 can provide solutions for wireless communication applications in electronic devices, including wireless local area networks (WLANs) (such as wireless fidelity (Wi-Fi) networks), Bluetooth (BT), global navigation satellite system (GNSS), frequency modulation (FM), near field communication (NFC), and infrared (IR) technologies. The wireless communication module 160 can be one or more devices integrating at least one communication processing module. The wireless communication module 160 receives electromagnetic waves via antenna 2, performs frequency modulation and filtering of the electromagnetic wave signals, and sends the processed signal to processor 110. The wireless communication module 160 can also receive signals to be transmitted from processor 110, perform frequency modulation and amplification, and convert them into electromagnetic waves for radiation via antenna 2.

[0315] In some embodiments, antenna 1 of the electronic device is coupled to mobile communication module 150, and antenna 2 is coupled to wireless communication module 160, enabling the electronic device to communicate with networks and other devices via wireless communication technology.

[0316] The display screen 194 is used to display the application's interface, etc. The display screen 194 includes a display panel. In some embodiments, the electronic device may include one or N display screens 194, where N is a positive integer greater than 1.

[0317] The electronic device 100 can perform shooting functions through an ISP, a camera 193, a video codec, a GPU, a display 194, and an application processor. The ISP is used to process the data fed back by the camera 193.

[0318] Internal memory 121 can be used to store computer executable program code, which includes instructions. Processor 110 executes various functional applications and data processing of the electronic device by running the instructions stored in internal memory 121. Internal memory 121 may include a program storage area and a data storage area. The program storage area may store the operating system and software code for at least one application program. The data storage area may store data generated during the use of the electronic device (e.g., images, videos, etc.). Furthermore, internal memory 121 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, general-purpose flash memory, etc.

[0319] The external storage interface 120 can be used to connect an external memory card, such as a Micro SD card, to expand the storage capacity of the electronic device. The external memory card communicates with the processor 110 through the external storage interface 120 to perform data storage functions. For example, images, videos, and other files can be saved on the external memory card.

[0320] Electronic devices can implement audio functions such as music playback and recording through audio modules 170, speakers 170A, receivers 170B, microphones 170C, headphone jacks 170D, and application processors.

[0321] The audio module 170 is used to convert digital audio information into analog audio signals for output, and also to convert analog audio input into digital audio signals. The audio module 170 can also be used for encoding and decoding audio signals. In some embodiments, the audio module 170 may be located in the processor 110, or some functional modules of the audio module 170 may be located in the processor 110.

[0322] The speaker 170A, also known as a "loudspeaker," is used to convert audio electrical signals into sound signals. The electronic device 100 can listen to music or listen to hands-free calls and other external playback scenarios through one or more speakers 170A.

[0323] The receiver 170B, also known as a "handpiece," can be one or more, and is used to convert audio electrical signals into sound signals. When the electronic device 100 answers a telephone call or voice message, the receiver 170B can be brought close to the ear to listen to the voice.

[0324] The microphone 170C, also known as a "microphone" or "voice transducer," is used to convert sound signals into electrical signals.

[0325] The 170D headphone jack is used to connect wired headphones.

[0326] The pressure sensor 180A is used to sense pressure signals and can convert the pressure signals into electrical signals. In some embodiments, the pressure sensor 180A may be disposed on the display screen 194.

[0327] The gyroscope sensor 180B can be used to determine the motion attitude of an electronic device. In some embodiments, the gyroscope sensor 180B can determine the angular velocity of the electronic device about three axes (i.e., the x, y, and z axes). The gyroscope sensor 180B can be used for image stabilization.

[0328] The barometric pressure sensor 180C is used to measure air pressure. In some embodiments, the electronic device calculates altitude using the air pressure value measured by the barometric pressure sensor 180C to assist in positioning and navigation.

[0329] The magnetic sensor 180D includes a Hall effect sensor. Electronic devices can use the magnetic sensor 180D to detect the opening and closing of a flip cover.

[0330] The 180E accelerometer can detect the magnitude of acceleration in various directions (typically three axes) of electronic devices. When the electronic device is stationary, it can detect the magnitude and direction of gravity.

[0331] The 180F distance sensor is used to measure distance. Electronic devices can measure distance using infrared or laser.

[0332] The proximity sensor 180G may include, for example, a light-emitting diode (LED) and a light detector, such as a photodiode. The LED may be an infrared LED. The electronic device emits infrared light outward through the LED. The electronic device uses the photodiode to detect infrared reflected light from nearby objects. When sufficient reflected light is detected, it can be determined that an object is near the electronic device. When insufficient reflected light is detected, the electronic device can determine that no object is near the electronic device.

[0333] An ambient light sensor 180L is used to detect ambient light levels. Electronic devices can adaptively adjust the brightness of the display screen 194 based on the detected ambient light levels.

[0334] The fingerprint sensor 180H is used to collect fingerprints.

[0335] The 180J temperature sensor is used to detect temperature.

[0336] Touch sensor 180K, also known as a "touch panel," can be located on display screen 194. The touch sensor 180K and display screen 194 together form a touchscreen, also known as a "touch screen." Touch sensor 180K is used to detect touch operations applied to or near it. The touch sensor can then transmit the detected touch operation to the application processor to determine the type of touch event.

[0337] The bone conduction sensor 180M can acquire vibration signals. In some embodiments, the bone conduction sensor 180M can acquire vibration signals from the vibrating bone segments of the human vocal cords.

[0338] Buttons 190 include a power button, volume buttons, etc. Buttons 190 can be mechanical buttons or touch buttons. The electronic device can receive button inputs and generate key signal inputs related to user settings and function control. Motor 191 can generate vibration alerts. Motor 191 can be used for incoming call vibration alerts or for touch vibration feedback. Indicator 192 can be an indicator light, used to indicate charging status, battery level changes, messages, missed calls, notifications, etc. SIM card interface 195 is used to connect a SIM card. The SIM card can be inserted into or removed from the SIM card interface 195 to achieve contact and separation with the electronic device.

[0339] Understandable Figure 25 The components shown do not constitute a specific limitation on the electronic device. The electronic device in the embodiments of the present invention may include, but is not limited to, components that are more advanced than those shown. Figure 25 More or fewer parts. Furthermore, Figure 25 The combination / connection relationships between the components can also be adjusted and modified.

[0340] Figure 26 This is a schematic diagram of the structure of the electronic device 2600 provided in an embodiment of this application. The electronic device 2600 can be a wearable device as described above, such as a watch. Figure 26As shown, electronic device 2600 may include: one or more processors 2601; one or more memories 2602; a communication interface 2603; and one or more computer programs 2604. These components can be connected via one or more communication buses 2605. The one or more computer programs 2604 are stored in the memory 2602 and configured to be executed by the one or more processors 2601. The one or more computer programs 2604 include instructions. For example, when electronic device 2600 is a wearable device as described above, the instructions can be used to perform relevant steps of the wearable device as in the corresponding embodiments above. The communication interface 2603 is used to enable communication between electronic device 2600 and other devices; for example, the communication interface may be a transceiver.

[0341] Based on the above, this application also provides a computer-readable storage medium storing instructions that, when executed, cause the method provided in any of the above-described method embodiments to be implemented. The computer-readable storage medium may include various media capable of storing program code, such as a USB flash drive, portable hard drive, read-only memory, random access memory, magnetic disk, or optical disk.

[0342] Based on the above, this application also provides a computer program product, which includes: a computer program (also referred to as code or instructions), which, when run on a computer, causes the computer to perform the method provided in any of the above method embodiments. Optionally, the computer can be a terminal device, such as a wearable device.

[0343] In the embodiments provided above, the methods provided by the embodiments of this application are described from the perspective of an electronic device (e.g., a wearable device) as the executing entity. To implement the functions of the methods provided in the embodiments of this application above, the electronic device may include hardware structures and / or software modules, implementing the above functions in the form of hardware structures, software modules, or a combination of hardware structures and software modules. Whether a particular function is implemented in the form of hardware structures, software modules, or a combination of hardware structures and software modules depends on the specific application and design constraints of the technical solution.

[0344] The embodiments of this application involve at least one, including one or more; where "multiple" means two or more. Furthermore, it should be understood that in the description of this specification, terms such as "first," "second," and "third" are used only for descriptive purposes and should not be construed as indicating relative importance or order. For example, the first detection method and the second detection method do not represent their relative importance or order, but are merely for descriptive distinction. In the embodiments of this application, "and / or" merely describes the relationship, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0345] The directional terms mentioned in the embodiments of this application, such as "up", "down", "left", "right", "inner", and "outer", are only for reference to the directions in the accompanying drawings. Therefore, the directional terms used are for better and clearer explanation and understanding of the embodiments of this application, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the embodiments of this application.

[0346] References to "one embodiment," "in some examples," or "some embodiments" as described in the embodiments of this application mean that one or more embodiments of this specification include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in some examples," "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0347] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk (SSD)). Where there is no conflict, the solutions in the above embodiments can be combined.

[0348] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0349] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0350] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0351] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0352] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the scope and intent of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application is also intended to include such modifications and variations.

Claims

1. A method for detecting eating behavior, characterized in that, Applied to wearable devices, the method includes: Determine the hand on which the user is currently wearing the wearable device; Determine the user's dominant hand for eating; If the hand being worn is different from the hand that is the user's dominant hand for eating, the first detection method is used to detect the user's eating behavior. If the hand being worn is the same as the hand the user is accustomed to using for eating, the second detection method is used to detect the user's eating behavior. The first detection method differs from the second detection method.

2. The method according to claim 1, characterized in that, The method of detecting a user's eating behavior using the first detection method includes: Detect the user's eating behavior based on the user's physiological data and / or current scene data.

3. The method according to claim 2, characterized in that, The process of detecting a user's eating behavior based on their physiological data includes: The user's physiological data is collected multiple times, including at least one of heart rate, energy, blood oxygen, body temperature, blood pressure, respiration, pulse, and blood glucose. Based on the multiple collections of physiological data, determine the changes in the user's physiological data; The user's eating behavior is detected based on changes in the physiological data.

4. The method according to claim 3, characterized in that, The step of detecting the user's eating behavior based on the changes in the physiological data includes: If the changes in the physiological data are consistent with the preset changes, then it is determined that there is eating behavior. The preset changes are used to describe the changes in the physiological data before and after the user eats.

5. The method according to claim 3 or 4, characterized in that, The method further includes: Based on the changes in the physiological data, determine the start time and / or end time of the eating behavior.

6. The method according to claim 2, characterized in that, The step of detecting the user's eating behavior based on current scenario data includes: If the current scenario is determined to be a eating scenario based on at least one of the following: current geographical location, surrounding environmental data, and user driving path, then the existence of eating behavior is confirmed.

7. The method according to claim 6, characterized in that, Before determining that eating behavior has occurred, the method further includes: It is determined that the amount of movement of the wearing hand is less than the preset amount of movement.

8. The method according to claim 7, characterized in that, The method further includes: Determine the start and / or end times of eating; The start time of eating is the moment when the user enters the eating scene or the moment when the amount of movement of the wearing hand changes from greater than the preset amount of movement to less than the preset amount of movement. The end of the meal is the moment when the user leaves the meal scene or the moment when the amount of movement of the wearing hand changes from less than the preset amount of movement to more than the preset amount of movement.

9. The method according to any one of claims 1-8, characterized in that, After detecting the user's eating behavior, the method further includes: Output a first prompt message, which is used to prompt the verification of relevant information of the eating behavior, including the start time and / or end time of the eating behavior; If a user confirmation is received, then the relevant information regarding the eating behavior is confirmed; If a user correction request is received, the relevant information regarding the eating behavior will be corrected.

10. The method according to any one of claims 1-9, characterized in that, After detecting the user's eating behavior, the method further includes: Detect the user's blood glucose; The eating behavior was verified based on the blood glucose levels.

11. The method according to any one of claims 1-10, characterized in that, After detecting the user's eating behavior, the method further includes: An eating behavior identifier is displayed at a first location, the eating behavior identifier being used to indicate the presence of eating behavior, and the first location being used to indicate the time when the eating behavior occurred.

12. The method according to claim 11, characterized in that, The eating behavior identifier is displayed in a first display mode, which is used to indicate that the eating behavior corresponding to the eating behavior identifier has not been confirmed by the user.

13. The method according to claim 12, characterized in that, The method further includes: Receive an operation for the eating behavior identifier, and display relevant information about the eating behavior corresponding to the eating behavior identifier, including the start time and / or end time of the eating behavior; After receiving confirmation of information related to the eating behavior, the eating behavior identifier is displayed in a second display mode. The second display mode is used to indicate that the eating behavior corresponding to the first identifier has been confirmed by the user. The second display mode is different from the first display mode.

14. The method according to any one of claims 11-13, characterized in that, A motion behavior identifier is displayed at a second location, the motion behavior identifier being used to indicate the presence of motion behavior, and the second location being used to indicate the time when the motion behavior occurred; And / or, A sleep behavior identifier is displayed at a third location, which indicates the presence of sleep behavior, and the third location indicates the time when the sleep behavior occurred.

15. The method according to any one of claims 1-14, characterized in that, Determining the user's dominant hand for eating includes: After determining the wearing hand, if the current scene is detected as an eating scene, then it is determined whether the action of the wearing hand is an eating action; If the action of wearing the hand is an eating action, then the dominant hand for eating is identified as the wearing hand; If the action of the wearing hand is not an eating action, then the eating dominant hand is determined to be the other hand besides the wearing hand.

16. The method according to any one of claims 1-15, characterized in that, The method further includes: Based on the user's eating behavior within the statistical period, the user's eating behavior statistical results are generated. The eating behavior statistical results include at least one of the following: the user's average number of meals, average eating speed, post-meal exercise, and post-meal sleep within the statistical period. Output the statistical results of the eating behavior.

17. The method according to any one of claims 1-16, characterized in that, The first detection method detects the user's eating behavior, including: using a first detection model to detect the user's eating behavior, wherein the input parameters of the first detection model include: the user's physiological data and / or current scene data, and the output parameters of the first detection model include a first confidence level, which is used to indicate the probability of eating behavior. The second detection method detects the user's eating behavior, including: using a second detection model to detect the user's eating behavior. The input parameters of the second detection model include: action data of the wearing hand. The output parameters of the second detection model include a second confidence level, which is used to indicate the probability of eating behavior.

18. A wearable device, characterized in that, Used to perform the method as described in any one of claims 1-17.

19. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program that, when run on a computer, causes the computer to perform the method as described in any one of claims 1 to 17.

20. A computer program product, characterized in that, Includes a computer program that, when run on a computer, causes the computer to perform the method as described in any one of claims 1 to 17.

Images