Chewing action recognition method, eating behavior monitoring earphone and storage medium

Abstract

The application relates to the technical field of diet monitoring, in particular to a chewing action recognition method, a diet behavior monitoring earphone and a storage medium. The chewing action recognition method comprises the following steps: receiving a target sensing signal from a diet behavior monitoring earphone in a target terminal; wherein, the target behavior feature is obtained through the following steps in the diet behavior monitoring earphone: performing vibration sensing on a target object to obtain sensing data flow, screening out a noise frequency band from the sensing data flow to obtain the target sensing signal; performing feature extraction on the target sensing signal to obtain chewing rhythm features, chewing strength features and chewing time value features of the target object; and performing chewing behavior analysis on the chewing rhythm features, the chewing strength features and the chewing time value features to determine a chewing analysis result of the target object. In this way, the chewing action of the monitoring object can be accurately recognized.

Description

Technical Field

[0001] This application relates to the field of dietary monitoring technology, and in particular to a chewing action recognition method, a dietary behavior monitoring headset, and a storage medium. Background Technology

[0002] Existing dietary monitoring technologies have some limitations in recognizing chewing movements. The commonly used amplitude thresholding method combined with general frequency domain analysis, while simple to implement, struggles to reliably distinguish chewing from other similar movements, resulting in low recognition accuracy and frequent false triggers. For example, it may misinterpret loud talking as chewing or miss genuine eating behavior.

[0003] Traditional fixed threshold methods lack flexibility. Some existing technologies rely on absolute numerical judgments, which struggle to adapt to individual differences among users and in different eating scenarios, resulting in poor adaptability and insufficient robustness. Currently, how to accurately identify the chewing movements of monitored subjects is a problem that urgently needs to be solved in the industry. Summary of the Invention

[0004] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application proposes a chewing action recognition method, an earphone for monitoring eating behavior, and a storage medium, which can accurately identify the chewing actions of the monitored subject.

[0005] The chewing action recognition method according to the first aspect of this application, applied to a target terminal, includes:

[0006] Target perception signal is received from the diet behavior monitoring earphone; wherein, the target perception signal is obtained in the diet behavior monitoring earphone through the following steps: vibration sensing is performed on the target object to obtain a perception data stream, and noise frequency band is filtered out from the perception data stream to obtain the target perception signal;

[0007] Feature extraction is performed on the target perception signal to obtain the chewing rhythm feature, chewing force feature, and chewing duration feature of the target object;

[0008] Chewing behavior analysis is performed on the chewing rhythm characteristics, chewing force characteristics, and chewing duration characteristics to determine the chewing analysis results of the target object.

[0009] According to some embodiments of this application, the step of extracting features from the target sensing signal to obtain the chewing rhythm features, chewing force features, and chewing duration features of the target object includes:

[0010] Based on the target sensing signal, a target vibration sensing spectrum is generated;

[0011] Spectral dynamic analysis is performed on the target vibration sensing spectrum to determine the chewing rhythm characteristics of the target sensing signal;

[0012] Spectral amplitude analysis is performed on the target vibration sensing spectrum to determine the chewing force characteristics of the target sensing signal;

[0013] Spectral time value analysis is performed on the target vibration sensing spectrum to determine the chewing time value characteristics of the target sensing signal.

[0014] According to some embodiments of this application, the step of performing spectral dynamic analysis on the target vibration sensing spectrum to determine the chewing rhythm characteristics of the target sensing signal includes:

[0015] The vibration sensing spectrum is divided based on a predetermined first division interval to obtain multiple dynamic analysis frames;

[0016] Extract the corresponding vibration main frequency data from each of the multiple dynamic analysis frames;

[0017] Dynamic feature analysis is performed on the vibration main frequency data of multiple consecutive dynamic analysis frames to obtain the dynamic features of the main frequency data as the chewing rhythm features.

[0018] According to some embodiments of this application, the step of performing spectral amplitude analysis on the target vibration sensing spectrum to determine the chewing force characteristics of the target sensing signal includes:

[0019] Select a target spectrum segment from the target vibration sensing spectrum;

[0020] The target spectrum segment is divided based on a predetermined second division interval to obtain multiple spectrum sub-segments;

[0021] Amplitude analysis is performed on each of the aforementioned spectral sub-segments to obtain the corresponding local amplitude;

[0022] Amplitude variation calculations are performed on the local amplitudes of multiple consecutive spectral sub-segments to obtain amplitude variation coefficients;

[0023] The chewing force feature is generated based on the amplitude variation coefficient and the local amplitude of each of the spectral sub-segments.

[0024] According to some embodiments of this application, the step of performing spectral time-value analysis on the target vibration sensing spectrum to determine the chewing time-value characteristics of the target sensing signal includes:

[0025] Select a target spectrum segment from the target vibration sensing spectrum;

[0026] Based on preset chewing spectrum discrimination conditions, the target spectrum segment is divided into a chewing occurrence sub-segment and a chewing cessation sub-segment;

[0027] Time-value analysis is performed based on the distribution characteristics of the chewing occurrence sub-segment and the chewing cessation sub-segment in the target spectral segment to determine the chewing time-value characteristics of the target sensing signal.

[0028] According to some embodiments of this application, the step of performing chewing behavior analysis on the chewing rhythm features, the chewing force features, and the chewing duration features to determine the chewing analysis results of the target object includes:

[0029] Based on the chewing rhythm characteristics, rhythm index analysis is performed to obtain the average chewing frequency and chewing frequency stability.

[0030] Based on the chewing force characteristics, the force index is analyzed to obtain the average chewing energy intensity and amplitude variation coefficient.

[0031] Based on the chewing time value characteristics, time value index analysis is performed to obtain the total chewing time, the percentage of effective chewing time, and the number of chewing interruptions.

[0032] The average chewing frequency, the stability of the chewing frequency, the average chewing energy intensity, the amplitude variation coefficient, the total chewing time, the proportion of effective chewing time, and the number of chewing interruptions are subjected to feature fusion processing to obtain the target chewing feature vector.

[0033] Chewing pattern recognition is performed based on the target chewing feature vector to determine the chewing analysis results of the target object.

[0034] According to some embodiments of this application, the step of performing chewing pattern recognition based on the target chewing feature vector to determine the chewing analysis result of the target object includes:

[0035] Obtain a dietary behavior pattern library; wherein, the dietary behavior pattern library stores corresponding chewing pattern representation vectors for various dietary behavior patterns;

[0036] The target chewing feature vector is compared with each chewing pattern representation vector to select the hit pattern representation vector from the dietary behavior pattern library.

[0037] The chewing analysis results are generated based on the dietary behavior pattern corresponding to the hit pattern representation vector.

[0038] The chewing action recognition method according to a second aspect embodiment of this application, applied to an earphone for monitoring eating behavior, includes:

[0039] Vibration sensing is performed on the target object to obtain a sensing data stream;

[0040] The target sensing signal is obtained by filtering out noise frequency bands from the sensing data stream.

[0041] Feature extraction is performed on the target perception signal to obtain the target behavior features of the target object;

[0042] The target behavioral features are sent to the target terminal, which then extracts features from the target perception signal to obtain the chewing rhythm features, chewing force features, and chewing duration features of the target object. Chewing behavior analysis is then performed based on the chewing rhythm features, chewing force features, and chewing duration features to determine the chewing analysis results of the target object.

[0043] Thirdly, embodiments of this application provide an earphone for monitoring dietary behavior, including: a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the chewing action recognition method as described in any one of the embodiments of the first aspect of this application.

[0044] Fourthly, embodiments of this application provide a computer-readable storage medium storing a program that is executed by a processor to implement the chewing action recognition method as described in any one of the embodiments of the first aspect of this application.

[0045] The chewing action recognition method, dietary behavior monitoring earphone, and storage medium according to the embodiments of this application have at least the following beneficial effects:

[0046] The chewing action recognition method of this application requires receiving a target perception signal from a diet behavior monitoring headset in the target terminal. The target behavior characteristics are obtained in the diet behavior monitoring headset through the following steps: vibration sensing is performed on the target object to obtain a perception data stream; noise frequency bands are filtered out from the perception data stream to obtain the target perception signal; feature extraction is performed on the target perception signal to obtain the chewing rhythm characteristics, chewing force characteristics, and chewing duration characteristics of the target object; chewing behavior analysis is performed on the chewing rhythm characteristics, chewing force characteristics, and chewing duration characteristics to determine the chewing analysis results of the target object. In this way, the chewing actions of the monitored object can be identified relatively accurately.

[0047] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0048] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:

[0049] Figure 1 A schematic flowchart of a chewing action recognition method provided in an embodiment of this application;

[0050] Figure 2 Another flowchart illustrating the chewing action recognition method provided in this application embodiment;

[0051] Figure 3 Another flowchart illustrating the chewing action recognition method provided in this application embodiment;

[0052] Figure 4A The changes in angular velocity data across all axes of the target sensing data over time are shown.

[0053] Figure 4B This shows how the angular velocity data of the target sensing data changes over time along the X-axis;

[0054] Figure 4C This shows how the angular velocity data of the target sensing data on the Y-axis changes over time;

[0055] Figure 4D This shows how the angular velocity data of the target sensing data on the Z-axis changes over time;

[0056] Figure 4E Another flowchart illustrating the chewing action recognition method provided in this application embodiment;

[0057] Figure 5 Another flowchart illustrating the chewing action recognition method provided in this application embodiment;

[0058] Figure 6 Another flowchart illustrating the chewing action recognition method provided in this application embodiment;

[0059] Figure 7 Another flowchart illustrating the chewing action recognition method provided in this application embodiment;

[0060] Figure 8 This is a schematic diagram of the hardware structure of the eating behavior monitoring earphone provided in an embodiment of this application;

[0061] Figure 9 This is a schematic diagram of another hardware structure of the eating behavior monitoring earphone provided in this application embodiment. Detailed Implementation

[0062] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0063] In the description of this application, "several" means one or more, "more than" means two or more, "greater than," "less than," and "exceeding" are understood to exclude the stated number, while "above," "below," and "within" are understood to include the stated number. The use of "first" and "second" in the description is merely for distinguishing technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of indicated technical features, or implicitly indicating the order of the indicated technical features.

[0064] In the description of this application, it should be understood that the orientation descriptions, such as up, down, left, right, front, and back, are based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not 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 this application.

[0065] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0066] In the description of this application, it should be noted that, unless otherwise explicitly defined, terms such as "setting," "installation," and "connection" should be interpreted broadly. Those skilled in the art can reasonably determine the specific meaning of the above terms in this application based on the specific content of the technical solution. Furthermore, the identification of specific steps in the following text does not imply a limitation on the order of steps or execution logic. The execution order and logic between each step should be understood and inferred from the content described in the embodiments.

[0067] Existing dietary monitoring technologies have limitations in recognizing chewing movements. Traditional methods primarily rely on inertial sensors to monitor large-amplitude body movements, and their algorithms are designed for movements with strong signals and simple patterns, such as walking and running. However, chewing involves minute head movements with drastically different signal characteristics. This technological mismatch makes it difficult for existing solutions to accurately capture chewing signals, and even more difficult to effectively handle interference from everyday head movements such as speaking, nodding, and shaking.

[0068] The commonly used amplitude thresholding method combined with general frequency domain analysis, while simple to implement, struggles to reliably distinguish chewing from other similar actions, resulting in low recognition accuracy and frequent false triggers. For example, it may misinterpret loud talking as chewing or miss genuine eating behavior. Traditional fixed thresholding methods lack flexibility. Some existing technologies use absolute numerical judgments, which are ill-suited to individual differences among users and in different eating scenarios, leading to poor adaptability and insufficient robustness.

[0069] Currently, how to accurately identify the chewing movements of monitored subjects is a problem that urgently needs to be solved in the industry.

[0070] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application proposes a chewing action recognition method, an earphone for monitoring eating behavior, and a storage medium, which can accurately identify the chewing actions of the monitored subject.

[0071] Reference Figure 1 The chewing action recognition method according to the embodiments of this application, applied to an earphone for monitoring eating behavior, may include:

[0072] Step S101: Vibration sensing is performed on the target object to obtain a sensing data stream;

[0073] Step S102: Noise frequency bands are filtered out from the sensing data stream to obtain the target sensing signal;

[0074] Step S103: Extract features from the target perception signal to obtain the target behavior features of the target object;

[0075] Step S104: The target behavior features are sent to the target terminal so that the target terminal can extract features from the target perception signal to obtain the chewing rhythm features, chewing force features and chewing duration features of the target object. Then, chewing behavior analysis is performed on the chewing rhythm features, chewing force features and chewing duration features to determine the chewing analysis results of the target object.

[0076] The chewing action recognition method according to the embodiments of this application, applied to a target terminal, may include:

[0077] Step S105: Receive target perception signal from the diet behavior monitoring earphone, and extract features from the target perception signal to obtain the chewing rhythm features, chewing force features, and chewing time value features of the target object.

[0078] Step S106: Perform chewing behavior analysis on chewing rhythm characteristics, chewing force characteristics, and chewing duration characteristics to determine the chewing analysis results of the target object.

[0079] The chewing action recognition method of this application aims to achieve accurate recognition of chewing actions through dietary behavior monitoring headphones, thereby solving the problems of low recognition accuracy and insufficient adaptability in the prior art.

[0080] In step S101 of some embodiments, vibration sensing is performed on the target object to obtain a sensing data stream;

[0081] It's important to note that the dietary behavior monitoring headphones first detect vibrations in the target subject (the user), acquiring raw sensory data. This process utilizes the headphones' built-in high-sensitivity sensors to capture minute vibration signals in the ear area, including mechanical vibrations from chewing movements. Because the headphones are in close contact with the body, they effectively filter out environmental noise, initially ensuring signal purity and stability. This initial step lays the foundation for subsequent signal processing, ensuring the quality of the input data.

[0082] In step S102 of some embodiments, noise bands are filtered out for the sensing data stream to obtain the target sensing signal;

[0083] It should be noted that, in the following embodiments, the sensing data stream undergoes noise band filtering to obtain the target sensing signal. This step is crucial for solving the noise interference problem in traditional technologies. By analyzing the spectral characteristics of the sensing data stream, frequency bands unrelated to chewing actions are identified and removed, such as high-frequency electronic noise or low-frequency background vibrations. This frequency band filtering method effectively reduces the false trigger rate, avoiding misinterpreting vibrations generated by non-chewing actions such as loud talking as chewing signals, thereby improving signal quality and reliability. This process not only improves the signal-to-noise ratio but also provides a clearer signal foundation for subsequent feature extraction.

[0084] In step S103 of some embodiments, feature extraction is performed on the target perception signal to obtain the target behavior features of the target object;

[0085] It's important to note that feature extraction is performed on the target perception signal to obtain the target object's behavioral characteristics. This step is the core of the entire method, providing crucial information for subsequent analysis by extracting features related to chewing behavior from the signal. The feature extraction process includes not only simple amplitude analysis but also more complex frequency and temporal characteristic analysis, such as chewing rhythm, chewing force, and chewing duration. These features comprehensively reflect the essential characteristics of chewing actions, laying the foundation for accurate identification. For example, chewing rhythm features can be determined by analyzing the periodic changes in the signal, chewing force features are measured by changes in signal amplitude, and chewing duration features are evaluated by the duration of the signal.

[0086] In some embodiments, step S104 involves sending the target behavior features to the target terminal so that the target terminal can extract features from the target perception signal to obtain the chewing rhythm features, chewing force features, and chewing duration features of the target object. Then, chewing behavior analysis is performed based on the chewing rhythm features, chewing force features, and chewing duration features to determine the chewing analysis results of the target object.

[0087] It should be noted that the extracted target behavioral features are sent to the target terminal. After receiving these features, the target terminal further analyzes the chewing behavior to determine the chewing analysis results of the target object. The terminal device has stronger computing and data analysis capabilities, enabling in-depth processing of the extracted features, including but not limited to pattern recognition, classification analysis, and health assessment. Through the analysis of the terminal device, this embodiment of the application can accurately determine whether the user's chewing behavior is normal and whether there are any bad habits, such as chewing too quickly, not chewing enough, or not eating continuously. This process not only improves the accuracy of identification but also allows for adaptive adjustments based on different users and scenarios, providing strong support for dietary behavior monitoring and health management.

[0088] Reference Figure 2 According to some embodiments of this application, step S104, which involves analyzing chewing behavior based on chewing rhythm characteristics, chewing force characteristics, and chewing duration characteristics to determine the chewing analysis results of the target object, may include:

[0089] Step S201: Based on the chewing rhythm characteristics, analyze the rhythm index to obtain the average chewing frequency and chewing frequency stability.

[0090] Step S202: Analyze the force index based on the chewing force characteristics to obtain the average chewing energy intensity and amplitude variation coefficient;

[0091] Step S203: Analyze the time value indicators based on the chewing time value characteristics to obtain the total chewing time, the percentage of effective chewing time, and the number of chewing interruptions;

[0092] Step S204: Perform feature fusion processing on the average chewing frequency, chewing frequency stability, average chewing energy intensity, amplitude variation coefficient, total chewing time, effective chewing time percentage and number of chewing interruptions to obtain the target chewing feature vector.

[0093] Step S205: Perform chewing pattern recognition based on the target chewing feature vector to determine the chewing analysis results of the target object.

[0094] In the embodiments of this application, the core step of chewing behavior analysis is to perform in-depth analysis and fusion processing on the extracted chewing rhythm features, chewing force features, and chewing time value features, thereby determining the chewing analysis results of the target object.

[0095] In some embodiments, step S201 involves analyzing rhythmic indicators based on chewing rhythm characteristics to obtain the average chewing frequency and chewing frequency stability.

[0096] It should be noted that, starting with the analysis of rhythm indicators, this application's embodiments calculate the average chewing frequency and chewing frequency stability based on chewing rhythm characteristics. The average chewing frequency reflects the number of chews per minute and is a key indicator for measuring eating speed; while chewing frequency stability describes the regularity of chewing movements, with a stable frequency generally indicating a more even chewing process. These two indicators provide important information about chewing rhythm for subsequent behavioral analysis.

[0097] In step S202 of some embodiments, the force index is analyzed based on the chewing force characteristics to obtain the average chewing energy intensity and amplitude variation coefficient;

[0098] It should be noted that the embodiments of this application analyze the chewing force characteristics to obtain the average chewing energy intensity and amplitude variation coefficient. The average chewing energy intensity quantifies the degree of force exerted by the user during chewing, reflecting the intensity of the chewing action; the amplitude variation coefficient measures the uniformity of chewing force, and a lower variation coefficient indicates that the user exerts relatively consistent force during chewing. These two indicators can reveal whether the user exerts excessive or insufficient force during chewing, and whether there is a problem of uneven force exertion, providing strong support for assessing the adequacy and health of chewing.

[0099] In step S203 of some embodiments, time value index analysis is performed based on chewing time value characteristics to obtain total chewing time, effective chewing time percentage and number of chewing interruptions;

[0100] It should be noted that this application's embodiments analyze chewing time characteristics to obtain total chewing time, effective chewing time percentage, and number of chewing interruptions. Total chewing time directly reflects the total time required for a user to complete one meal; the effective chewing time percentage shows the proportion of time the user actually spends chewing, and a higher percentage usually indicates greater focus during eating; the number of chewing interruptions records the number of pauses caused by drinking water, talking, or distraction during eating. These time-value indicators comprehensively reflect the user's eating continuity and focus, providing important evidence for identifying poor eating habits.

[0101] In step S204 of some embodiments, feature fusion processing is performed on the average chewing frequency, chewing frequency stability, average chewing energy intensity, amplitude variation coefficient, total chewing time, effective chewing time percentage and number of chewing interruptions to obtain the target chewing feature vector.

[0102] It should be noted that after obtaining the aforementioned key indicators, this embodiment performs feature fusion processing, integrating the average chewing frequency, chewing frequency stability, average chewing energy intensity, amplitude variation coefficient, total chewing time, effective chewing time percentage, and number of chewing interruptions into a target chewing feature vector. This fusion process is not a simple data accumulation, but rather uses specific algorithms or models to weight and normalize features from different dimensions, generating a feature vector that comprehensively describes the user's chewing behavior. This vector integrates information from three aspects: rhythm, intensity, and duration, providing a unified input format for subsequent chewing pattern recognition.

[0103] In some embodiments, step S205 involves performing chewing pattern recognition based on the target chewing feature vector to determine the chewing analysis results of the target object.

[0104] It should be noted that, based on the target chewing feature vector, this embodiment of the application performs chewing pattern recognition by mapping the feature vector to specific chewing behavior patterns through a preset classification model or rule engine. This process can identify whether a user has poor chewing habits, such as eating too quickly, insufficient chewing, uneven force, or discontinuous eating. Through pattern recognition, this embodiment of the application can provide users with accurate analysis results and personalized improvement suggestions, helping users optimize their eating behavior and improve their health. The entire analysis process, from the parsing and fusion of multi-dimensional features to pattern recognition, forms a complete technical chain, ensuring the accuracy and reliability of the analysis results.

[0105] Reference Figure 3 According to some embodiments of this application, step S205, which involves performing chewing pattern recognition based on the target chewing feature vector to determine the chewing analysis result of the target object, may include:

[0106] Step S301: Obtain the dietary behavior pattern library; wherein, the dietary behavior pattern library stores corresponding chewing pattern representation vectors for various dietary behavior patterns.

[0107] Step S302: Based on the comparison between the target chewing feature vector and the representation vectors of each chewing pattern, the matching pattern representation vector is selected from the dietary behavior pattern library.

[0108] Step S303: Generate chewing analysis results based on the dietary behavior patterns corresponding to the hit pattern representation vectors.

[0109] In some embodiments of this application, chewing pattern recognition is based on the target chewing feature vector, which is a key step in determining the chewing analysis results of the target object.

[0110] In some embodiments, step S301 involves obtaining a dietary behavior pattern library; wherein the dietary behavior pattern library stores corresponding chewing pattern representation vectors for various dietary behavior patterns.

[0111] It should be noted that this application requires obtaining a dietary behavior pattern database, which forms the basis of the entire recognition process. The dietary behavior pattern database stores chewing pattern representation vectors corresponding to various dietary behavior patterns. These representation vectors are obtained through the analysis and summarization of a large amount of sample data, and they accurately reflect the chewing characteristics under different dietary behavior patterns. For example, the representation vector of a rapid eating pattern may have a high average chewing frequency and a low proportion of effective chewing time; while the representation vector of an insufficiently chewed pattern may exhibit a high coefficient of variation and a short total chewing time. These representation vectors provide standards and references for subsequent pattern comparison.

[0112] In some embodiments, step S302 involves comparing the target chewing feature vector with each chewing pattern representation vector to select the hit pattern representation vector from the dietary behavior pattern library.

[0113] It should be noted that after acquiring the dietary behavior pattern library, this embodiment compares the target chewing feature vector with each chewing pattern representation vector in the library. This comparison process is the core of the identification; it determines the best-matching pattern by calculating the similarity or distance between the target feature vector and each representation vector in the library. For example, Euclidean distance, cosine similarity, or other statistical methods can be used to quantify the difference between two vectors. This embodiment compares the target vector with each representation vector in the library one by one, searching for the vector with the smallest difference or the highest similarity. This process not only needs to consider the matching degree of individual features but also needs to comprehensively evaluate the overall consistency among multiple features. For example, even if a feature is very close to a pattern in the library, if other features have significant differences from the representation vector of that pattern, then this pattern may not be selected as the hit pattern. Through this comprehensive comparison, this embodiment can select the hit pattern representation vector that best matches the target chewing feature vector from the dietary behavior pattern library.

[0114] In some embodiments, step S303 generates chewing analysis results based on the dietary behavior patterns corresponding to the hit pattern representation vectors.

[0115] It should be noted that, based on the dietary behavior patterns corresponding to the hit pattern representation vectors, this embodiment generates chewing analysis results. This process transforms the identified patterns into specific analytical conclusions, providing users with detailed information about their chewing behavior. For example, if the dietary behavior pattern corresponding to the hit pattern representation vector is "eating quickly and not chewing sufficiently," then the analysis results will clearly indicate that the user has problems with eating quickly and not chewing sufficiently. This result not only includes a description of the current chewing behavior but can also provide further health advice, such as suggesting that the user slow down their eating speed and increase the number of chews. Through this process from feature vector to pattern comparison and then to result generation, this embodiment can provide users with accurate and instructive chewing analysis results, helping users understand their eating habits and take corresponding improvement measures.

[0116] It should be understood that the entire chewing pattern recognition process in this application is a systematic analysis chain. From acquiring the dietary behavior pattern library to feature vector comparison, and finally to the generation of analysis results, each step is closely linked and interdependent. This process not only solves the problem of low recognition accuracy in existing technologies, but also provides users with personalized dietary behavior assessments and suggestions through refined feature analysis and pattern matching, thereby achieving technological progress in the field of dietary monitoring.

[0117] In some embodiments, step S105 involves receiving a target perception signal from the diet behavior monitoring headphones and extracting features from the target perception signal to obtain the chewing rhythm features, chewing force features, and chewing time value features of the target object.

[0118] It should be noted that in the target terminal, target perception signals are received from the eating behavior monitoring earphones, and feature extraction is performed on these signals. Unlike the initial processing at the earphone end, the terminal device can perform a more complex feature extraction process to obtain more accurate chewing rhythm features, chewing force features, and chewing duration features. These features not only reflect the physical characteristics of chewing actions but also reveal the user's eating habits and health status. For example, by analyzing chewing rhythm and force, embodiments of this application can determine whether the user has insufficient chewing; by analyzing chewing duration and the number of interruptions, embodiments of this application can assess the user's eating continuity and focus.

[0119] It is worth noting that feature extraction from the target perception signal to obtain chewing rhythm features, chewing force features, and chewing duration features is a crucial step in the entire chewing action recognition and analysis process. This process not only provides a precise data foundation for subsequent behavior analysis but also improves the accuracy and adaptability of the recognition system, enabling it to effectively cope with complex and ever-changing real-world usage scenarios.

[0120] First, extracting chewing rhythm features can reflect the frequency and regularity of a user's chewing actions. By analyzing periodic changes in perceived signals, embodiments of this application can determine the number of chews per minute and the stability of chewing actions. For example, a stable chewing rhythm usually indicates that the user is focused and chews thoroughly while eating, while frequent rhythm changes may suggest that the user is distracted or chewing unevenly during eating. This extraction of rhythm features provides direct evidence for identifying bad habits such as eating too quickly and not chewing thoroughly, and also lays the foundation for subsequent behavioral analysis and health assessment.

[0121] Secondly, extracting chewing force characteristics can reveal the force exerted by the user during chewing. By analyzing the amplitude changes of the perceived signal, this embodiment can quantify the magnitude of each chew and the uniformity of force changes. The strength of chewing force directly affects the grinding effect of food, thereby affecting the burden on the digestive system. For example, insufficient force may lead to incomplete grinding of food, increasing the difficulty of gastrointestinal digestion; while excessive force may put unnecessary pressure on the teeth and jaw joints. By extracting chewing force characteristics, this embodiment can identify potential problems such as uneven force or excessive force, and provide users with targeted improvement suggestions to help them establish healthy chewing habits.

[0122] Finally, the extraction of chewing time value features can comprehensively reflect the temporal characteristics of the user's eating process. By analyzing the duration of perceived signals, this embodiment can determine the total duration of a single meal, the proportion of effective chewing time, and the number of interruptions during the eating process. These parameters can intuitively reflect the user's eating speed, eating continuity, and eating focus. For example, a shorter total duration may mean eating too quickly, while frequent interruptions may indicate that the user is distracted or drinking water frequently during the eating process. By analyzing these time value features, this embodiment can identify bad habits such as discontinuous eating and eating too quickly, and provide users with personalized intervention measures to help them adjust their eating rhythm and improve eating quality.

[0123] In summary, the extraction of chewing rhythm features, chewing force features, and chewing duration features provides multi-dimensional data support for the identification and analysis of chewing actions. These features can not only accurately describe a user's chewing behavior but also reveal potential unhealthy habits and their impact on health. Through comprehensive analysis of these features, the embodiments of this application can provide users with personalized dietary advice and health interventions, thereby effectively promoting the establishment of healthy eating habits and improving overall health. This process fully demonstrates the important value of refined feature extraction in complex behavior recognition and provides technical support for the development of dietary behavior monitoring technology.

[0124] Reference Figure 4A This shows all axes of the target perception data. Figure 4A This reflects the change of the gyroscope's three-axis angular velocity data over time. This data can be used to analyze chewing rhythm characteristics, chewing force characteristics, and chewing duration characteristics.

[0125] Analysis of the chewing rhythm characteristics reveals periodic fluctuations in the Z-axis angular velocity during chewing. The frequency of these fluctuations reflects the chewing rhythm, i.e., the number of chews per unit time. By calculating the average period of these fluctuations, the average chewing frequency can be derived, a key indicator for measuring eating speed. Furthermore, analyzing the consistency of these fluctuations allows for the assessment of the stability of the chewing frequency, i.e., the uniformity of the chewing motion.

[0126] Regarding chewing force characteristics, the amplitude variation of the Z-axis angular velocity in the figure provides relevant information. During chewing, the magnitude of the angular velocity amplitude reflects the chewing force; a larger amplitude generally indicates greater chewing force, while a smaller amplitude may indicate less chewing force. By calculating the average value and coefficient of variation of these amplitudes, the average level and consistency of chewing force can be assessed, thereby identifying whether the chewing force is uniform and whether there are problems with excessive or insufficient force.

[0127] As for the characteristics of chewing duration, this can be determined by analyzing the time periods during which chewing occurs and stops in the graph. By identifying the distribution of chewing actions on the time axis, parameters such as the total chewing duration, the percentage of effective chewing time, and the number of chewing interruptions can be calculated. These parameters reflect the continuity and focus of eating; for example, the total duration reflects the duration of a single meal, the percentage of effective chewing time reflects the proportion of chewing during the eating process, and the number of chewing interruptions reveals the frequency of pauses during eating.

[0128] Reference Figure 4B The graph specifically illustrates the variation of target perception data's angular velocity along the X-axis over time. X-axis angular velocity data analysis: The graph shows that the angular velocity fluctuates relatively little throughout the time period, with minimal amplitude changes. This may indicate minimal or stable head movement in the X-axis direction. If the X-axis represents horizontal movement, this might suggest minimal horizontal head movement during eating, which could be unrelated to chewing actions or indicate that the head remains relatively stable during eating.

[0129] Reference Figure 4C The diagram specifically illustrates the variation of target perception data's angular velocity along the Y-axis over time. Y-axis angular velocity data analysis: Similar to the X-axis, the angular velocity variation along the Y-axis is relatively stable with small fluctuations. This may indicate less motion in the Y-axis direction, which could be either vertical or horizontal. This stability may help in more accurate analysis of chewing-related Z-axis data, as it reduces potential interference from motion in other directions.

[0130] Reference Figure 4D The diagram specifically illustrates the variation of target perception data's angular velocity along the Z-axis over time. Z-axis angular velocity data analysis: These data exhibit significant fluctuations during chewing, which is crucial for analyzing chewing characteristics. The Z-axis fluctuations are substantial, with distinct peaks appearing at certain time intervals; these peaks likely correspond to chewing movements. By analyzing the frequency of these peaks, the rhythmic characteristics of chewing can be determined, such as the number of chews per minute. Simultaneously, the magnitude of the peaks can reflect the force characteristics of chewing, i.e., the degree of effort exerted during chewing. Furthermore, by observing the duration and intervals of the Z-axis angular velocity, the temporal characteristics of chewing can be analyzed, such as the total chewing duration and the number of interruptions.

[0131] It is worth noting that the significant fluctuations in the Z-axis data indicate a close correlation between movement along this axis and chewing behavior, making Z-axis data particularly important for analyzing chewing rhythm, force, and duration characteristics. Similarly, the X-axis and Y-axis data provide valuable information and are crucial for a comprehensive understanding of head movements and accurate assessment of eating behavior. While the fluctuations in X-axis and Y-axis data are smaller, they provide information about head movements in the horizontal and possibly vertical directions. This data helps in analyzing the comprehensiveness of head movements, ensuring that all relevant motion dimensions are considered when assessing chewing behavior. When analyzing Z-axis data, X-axis and Y-axis data can help identify and eliminate non-chewing-related motion interferences. For example, if the X-axis or Y-axis data shows abnormal fluctuations within a certain time period, it may indicate that the Z-axis fluctuations during that period are not entirely caused by chewing but are influenced by other head movements. Furthermore, X-axis and Y-axis data can aid in identifying specific behavioral patterns. For example, slight head shaking or tilting may be associated with certain specific actions during eating, such as tilting the head to better chew food. Identifying these patterns can deepen the understanding of eating behavior. Therefore, the completeness of multi-axis data is crucial when performing any type of motion analysis. Considering data from all axes further enhances the accuracy and reliability of the analysis results.

[0132] It should be understood that combining data from all three axes allows for a more accurate identification and analysis of chewing behavior. The stability of the X and Y axes helps confirm that fluctuations on the Z-axis are primarily caused by chewing. Detailed analysis of the Z-axis data allows for the extraction of chewing rhythm, intensity, and duration characteristics, which are crucial for understanding eating behavior and identifying potential unhealthy eating habits. For example, frequent high-intensity fluctuations in the Z-axis data may indicate a rapid and forceful chewing habit, which can put stress on the digestive system. Conversely, smoother fluctuations with longer intervals may indicate slower and more thorough chewing, which is generally considered a healthier eating habit.

[0133] Reference Figure 4E According to some embodiments of this application, step S105 extracts features from the target perception signal to obtain the chewing rhythm features, chewing force features, and chewing time value features of the target object, which may include:

[0134] Step S401: Generate the target vibration sensing spectrum based on the target sensing signal;

[0135] Step S402: Perform dynamic spectral analysis on the target vibration sensing spectrum to determine the chewing rhythm characteristics of the target sensing signal;

[0136] Step S403: Perform spectral amplitude analysis on the target vibration sensing spectrum to determine the chewing force characteristics of the target sensing signal;

[0137] Step S404: Perform spectral time value analysis on the target vibration sensing spectrum to determine the chewing time value characteristics of the target sensing signal.

[0138] In some embodiments of this application, the process of feature extraction of target perception signals is a key step in chewing action recognition. Through this process, the chewing rhythm features, chewing force features, and chewing duration features of the target object can be obtained.

[0139] In some embodiments, step S401 involves generating a target vibration sensing spectrum based on the target sensing signal.

[0140] It should be noted that this embodiment of the application begins with generating the target vibration sensing spectrum, which is the result of frequency domain transformation of the original sensing signal. By converting the time-domain sensing signal into a spectrum, this embodiment of the application can more clearly identify the different frequency components in the signal and their corresponding energy distributions. This spectrum conversion provides the basis for subsequent feature extraction, enabling the system to separate specific frequency features related to chewing actions from complex signals.

[0141] In some embodiments, step S402 involves performing spectral dynamic analysis on the target vibration sensing spectrum to determine the chewing rhythm characteristics of the target sensing signal.

[0142] It should be noted that this application embodiment performs spectral dynamic analysis on the target vibration sensing spectrum to determine the chewing rhythm characteristics of the target sensing signal. This analysis process mainly focuses on the changes in energy over time in the spectrum, especially those dynamic changes related to the periodicity of chewing movements. For example, chewing movements produce vibrations with a certain frequency, which corresponds to the opening and closing speed of the jaw. By analyzing the spectral dynamics, this application embodiment can identify such periodic changes and quantify them as chewing rhythm characteristics, such as the average chewing frequency and the stability of the chewing frequency. These characteristics reflect the speed of the user's chewing and the regularity of the chewing movements, and are important indicators for assessing the health of chewing behavior.

[0143] Reference Figure 5 According to some embodiments of this application, step S402, which involves performing spectral dynamic analysis on the target vibration sensing spectrum to determine the chewing rhythm characteristics of the target sensing signal, may include:

[0144] Step S501: Divide the vibration sensing spectrum based on a predetermined first division interval to obtain multiple dynamic analysis frames;

[0145] Step S502: Extract the corresponding vibration main frequency data from multiple dynamic analysis frames respectively;

[0146] Step S503: Dynamic feature analysis is performed based on the vibration main frequency data of multiple continuous dynamic analysis frames to obtain the dynamic features of the main frequency data as chewing rhythm features.

[0147] In some embodiments of this application, the process of performing spectral dynamic analysis on the target vibration sensing spectrum to determine chewing rhythm characteristics is a refined signal processing step.

[0148] In some embodiments, step S501 involves dividing the vibration sensing spectrum based on a predetermined first division interval to obtain multiple dynamic analysis frames.

[0149] It should be noted that the vibration sensing spectrum is divided based on a predetermined first division interval. This division method involves segmenting continuous spectral data into multiple dynamic analysis frames with fixed time lengths. The choice of the first division interval is crucial because it directly determines the level of signal detail that each analysis frame can capture. If the interval is too long, rapid changes in chewing motion may be missed; while if the interval is too short, excessive noise will be introduced, increasing computational complexity. Therefore, the determination of the first division interval needs to match the typical frequency characteristics of chewing motion so that each dynamic analysis frame can contain sufficient information to reflect the periodic changes in chewing motion.

[0150] In step S502 of some embodiments, the corresponding vibration main frequency data are extracted from multiple dynamic analysis frames respectively;

[0151] It should be noted that after dividing the vibration sensing spectrum into multiple dynamic analysis frames, this embodiment extracts the corresponding dominant vibration frequency data from each dynamic analysis frame. The dominant frequency data refers to the frequency component that dominates each analysis frame; it typically corresponds to the main vibration frequency generated by the chewing action. This frequency component directly reflects the chewing action, as the opening and closing motion of the jaw during chewing generates a vibration signal with a certain frequency. By extracting the dominant frequency data from each dynamic analysis frame, this embodiment can capture the frequency characteristics of the chewing action at different time points, thereby providing basic data for further dynamic feature analysis.

[0152] In step S503 of some embodiments, dynamic feature analysis is performed based on the vibration main frequency data of multiple continuous dynamic analysis frames to obtain the dynamic features of the main frequency data as chewing rhythm features.

[0153] It should be noted that the embodiments of this application perform dynamic feature analysis based on the vibration dominant frequency data of multiple continuous dynamic analysis frames. This process involves performing time series analysis on the extracted dominant frequency data to identify the variation pattern of the dominant frequency data over time. This variation pattern directly reflects the rhythmic characteristics of chewing actions, such as the speed and regularity of chewing. By analyzing the dynamic changes of the dominant frequency data, the embodiments of this application can calculate the average chewing frequency, i.e., the number of chewing actions per unit time, and the stability of the chewing frequency, i.e., whether the chewing actions are uniform and consistent. These dynamic features, as chewing rhythm features, can provide key information for subsequent behavioral analysis, helping to identify whether users have bad habits such as eating too quickly or chewing irregularly.

[0154] It should be understood that the entire spectrum dynamic analysis process is a systematic signal processing chain. From spectrum division to extraction of dominant frequency data, and then to the analysis of dynamic features, each step is closely linked and interdependent. This process not only solves the problem of the difficulty in stably distinguishing chewing actions in existing technologies, but also provides users with accurate chewing behavior assessments and health recommendations through refined feature extraction. Through this step-by-step analysis from spectrum to features, the embodiments of this application can effectively identify the rhythmic characteristics of chewing actions, thereby achieving technological progress in the field of dietary behavior monitoring.

[0155] In some embodiments, step S403 involves performing spectral amplitude analysis on the target vibration sensing spectrum to determine the chewing force characteristics of the target sensing signal.

[0156] It should be noted that this application embodiment performs spectral amplitude analysis on the target vibration sensing spectrum to determine the chewing force characteristics of the target sensing signal. This analysis focuses on the energy intensity of different frequency components in the spectrum, particularly those frequency ranges related to chewing force. For example, greater chewing force produces higher energy vibration signals, while less force corresponds to lower energy. By analyzing the spectral amplitude, this application embodiment can quantify chewing force, such as average chewing energy intensity and amplitude variation coefficient. These characteristics not only reflect the degree of force exerted by the user during chewing but also reveal whether the force is uniform, which is crucial for assessing the adequacy and health of chewing.

[0157] Reference Figure 6 According to some embodiments of this application, step S403, which involves performing spectral amplitude analysis on the target vibration sensing spectrum to determine the chewing force characteristics of the target sensing signal, may include:

[0158] Step S601: Select a target spectrum segment from the target vibration sensing spectrum;

[0159] Step S602: Divide the target spectrum segment based on the predetermined second division interval to obtain multiple spectrum sub-segments;

[0160] Step S603: Perform amplitude analysis on each spectral sub-segment to obtain the corresponding local amplitude;

[0161] Step S604: Calculate the amplitude variation for the local amplitudes of multiple continuous spectral sub-segments to obtain the amplitude variation coefficient;

[0162] Step S605: Generate chewing force features based on amplitude variation coefficient and local amplitude of each spectral sub-segment.

[0163] In some embodiments of this application, the process of performing spectral amplitude analysis on the target vibration sensing spectrum to determine the chewing force characteristics is a progressively refined signal processing flow.

[0164] In some embodiments, step S601 involves selecting a target spectrum segment from the target vibration sensing spectrum;

[0165] It should be noted that the target spectral segment is selected from the target vibration sensing spectrum. Since the vibration signal generated by chewing has a specific frequency range, this embodiment of the application needs to filter out the spectral segment directly related to chewing from the entire spectrum. This segment selection is based on prior knowledge of the frequency characteristics of chewing actions, ensuring that subsequent analysis focuses on the frequency range most relevant to chewing behavior, thereby improving the accuracy and efficiency of the analysis.

[0166] In step S602 of some embodiments, the target spectrum segment is divided based on a predetermined second division interval to obtain multiple spectrum sub-segments;

[0167] It should be noted that, in this embodiment, the segment is divided based on a predetermined second division interval to obtain multiple spectral sub-segments. This division method further subdivides the target spectral segment into smaller time windows, allowing for more detailed analysis of the signal within each sub-segment. The selection of the second division interval is equally crucial; it needs to be determined based on the typical duration of the chewing action to ensure that each sub-segment can capture the signal characteristics of the complete cycle or a portion of the chewing action. This division method enables the system to monitor changes in the force of the chewing action more precisely.

[0168] In some embodiments, step S603 involves performing amplitude analysis on each spectral sub-segment to obtain the corresponding local amplitude.

[0169] It should be noted that amplitude analysis is performed for each spectral sub-segment, and the embodiments of this application calculate the local amplitude within each sub-segment. The local amplitude reflects the signal strength within each sub-segment and is a key indicator for measuring chewing force. By analyzing the amplitude of each sub-segment, the embodiments of this application can capture the changes in the force of chewing action at different time points. For example, a larger local amplitude may correspond to a more forceful chewing action, while a smaller amplitude may indicate a lighter chewing force. This process provides basic data for subsequent amplitude variation calculations, enabling the system to further quantify the characteristics of chewing force variation.

[0170] In step S604 of some embodiments, amplitude variation calculation is performed on the local amplitudes of multiple continuous spectral sub-segments to obtain amplitude variation coefficients;

[0171] It should be noted that, in this embodiment, amplitude variation calculations are performed on the local amplitudes of multiple continuous spectral sub-segments to obtain the amplitude variation coefficient. The amplitude variation coefficient is an important indicator for measuring the uniformity of chewing force variation. By calculating the ratio of the standard deviation to the mean of the local amplitude of each sub-segment, this embodiment can quantify the degree of fluctuation of chewing force over time. A lower amplitude variation coefficient indicates more uniform chewing force, while a higher variation coefficient may suggest significant variations in chewing force, such as uneven force application or intermittent excessive force during chewing. This indicator provides an important basis for assessing the health of chewing behavior.

[0172] In some embodiments, step S605 generates chewing force features based on the amplitude variation coefficient and the local amplitude of each spectral sub-segment.

[0173] It should be noted that, based on the amplitude variation coefficient and the local amplitude of each spectral sub-segment, the embodiments of this application generate chewing force characteristics. This characteristic integrates the absolute intensity of chewing force (reflected by local amplitude) and the uniformity of force variation (reflected by amplitude variation coefficient). The chewing force characteristic not only quantifies the degree of force exerted by the user during chewing but also reveals whether the force is uniform, thus providing comprehensive force-related information for subsequent behavioral analysis and health assessment. For example, if the local amplitude is high but the amplitude variation coefficient is low, it may indicate that the user is using relatively strong but uniform chewing force; conversely, if the local amplitude is low but the amplitude variation coefficient is high, it may suggest that the user is using relatively weak and uneven chewing force. Through this comprehensive analysis, the embodiments of this application can provide users with a detailed assessment of chewing force, helping them optimize their chewing habits and improve their dietary health.

[0174] It should be understood that the entire spectral amplitude analysis process is a systematic signal processing chain. From the selection of the target spectral segment to the division of sub-segments, and then to the calculation of local amplitudes and the generation of amplitude variation coefficients, each step is closely linked and interdependent. This process not only solves the problem of accurately quantifying chewing force in existing technologies, but also provides users with precise chewing behavior assessments and health recommendations through refined feature extraction. Through this step-by-step analysis from spectrum to features, the embodiments of this application can effectively identify the force characteristics of chewing actions, thereby achieving technological progress in the field of dietary behavior monitoring.

[0175] In some embodiments, step S404 involves performing spectral time value analysis on the target vibration sensing spectrum to determine the chewing time value characteristics of the target sensing signal.

[0176] It should be noted that this application embodiment performs spectral time value analysis on the target vibration sensing spectrum to determine the chewing time value characteristics of the target sensing signal. This analysis process mainly focuses on the temporal characteristics of energy duration in the spectrum, especially the duration information related to chewing actions. For example, by analyzing the time periods of energy duration in the spectrum, this application embodiment can determine the total duration of chewing actions, the proportion of effective chewing time, and the number of chewing interruptions. These time value characteristics reflect the continuity and focus of the user's eating, and are important criteria for assessing whether eating habits are healthy.

[0177] Reference Figure 7 According to some embodiments of this application, step S404, which involves performing spectral time value analysis on the target vibration sensing spectrum to determine the chewing time value characteristics of the target sensing signal, may include:

[0178] Step S701: Select a target spectrum segment from the target vibration sensing spectrum;

[0179] Step S702: Based on the preset chewing spectrum discrimination conditions, the target spectrum segment is divided into a chewing occurrence sub-segment and a chewing cessation sub-segment;

[0180] Step S703: Perform time value analysis based on the distribution characteristics of the chewing occurrence sub-segment and chewing cessation sub-segment in the target spectral segment to determine the chewing time value characteristics of the target sensing signal.

[0181] In some embodiments of this application, the process of performing spectral time value analysis on the target vibration sensing spectrum is an important step in chewing action recognition. Its purpose is to determine the chewing time value characteristics of the target sensing signal by analyzing the spectral data.

[0182] In some embodiments, step S701 involves selecting a target spectral segment from the target vibration sensing spectrum;

[0183] It should be noted that the target spectral segment is selected from the target vibration sensing spectrum. Since the vibration signal generated by chewing has a specific frequency range and energy distribution characteristics in the spectrum, it is necessary to filter out the portion relevant to chewing behavior from the entire spectrum. This selection process is based on prior knowledge of the frequency characteristics of chewing behavior, ensuring that subsequent analysis focuses on the spectral interval most relevant to chewing behavior, thereby improving the accuracy and efficiency of the analysis.

[0184] In step S702 of some embodiments, the target spectrum segment is divided into a chewing occurrence sub-segment and a chewing cessation sub-segment based on a preset chewing spectrum discrimination condition.

[0185] It should be noted that after selecting the target spectral segment, this embodiment divides the target spectral segment into chewing occurrence sub-segments and chewing cessation sub-segments based on preset chewing spectrum discrimination conditions. This division process is achieved by analyzing the energy changes and frequency distribution in the spectral segment. Chewing action is typically manifested in the spectrum as an increase in energy within a specific frequency range, while the energy decreases significantly when chewing stops. Therefore, this embodiment identifies which sub-segments correspond to the occurrence of chewing action and which sub-segments correspond to the cessation of chewing action through preset discrimination conditions, such as energy thresholds or frequency characteristics. This division not only needs to consider the characteristics of individual sub-segments but also the continuity and correlation between sub-segments to ensure the accuracy and reliability of the division results.

[0186] In some embodiments, step S703 involves performing time value analysis based on the distribution characteristics of the chewing occurrence sub-segment and chewing cessation sub-segment in the target spectral segment to determine the chewing time value characteristics of the target sensing signal.

[0187] It should be noted that this application embodiment performs time-value analysis based on the distribution characteristics of chewing occurrence and chewing cessation sub-segments in the target spectral segment. This analysis process mainly determines the chewing time-value characteristics of the target perceived signal by statistically analyzing the duration of chewing occurrence sub-segments and the interval time of chewing cessation sub-segments. Specifically, the total duration of chewing occurrence sub-segments reflects the total duration of the chewing action, while the interval time of chewing cessation sub-segments reflects the number of interruptions and the duration of these interruptions. Through this statistical information, this application embodiment can calculate key parameters such as the total chewing duration, the proportion of effective chewing time, and the number of chewing interruptions. These parameters not only directly reflect the user's eating speed and eating continuity but also reveal the user's behavioral patterns and potential problems during the eating process.

[0188] It should be understood that the entire spectral timing analysis process is a systematic signal processing chain. From the selection of the target spectral segment to the division of sub-segments, and then to the calculation of timing features, each step is closely linked and interdependent. This process not only solves the problem of accurately quantifying chewing timing features in existing technologies, but also provides users with precise chewing behavior assessments and health recommendations through refined feature extraction. Through this step-by-step analysis from spectrum to features, the embodiments of this application can effectively identify the timing features of chewing actions, thereby achieving technological progress in the field of dietary behavior monitoring. This analysis method provides important temporal dimension information for subsequent behavioral analysis and health assessment, helping users better understand their eating habits and take corresponding improvement measures.

[0189] Through the steps described in the embodiments of this application, comprehensive and accurate chewing features can be extracted from the target perception signal. These features not only cover the rhythm, force, and duration of chewing actions, but also reveal the user's behavioral patterns and potential problems during eating. The extraction of these features provides a solid data foundation for subsequent chewing behavior analysis and health assessment, enabling the system to provide users with personalized dietary recommendations and health interventions.

[0190] In some embodiments, step S106 involves analyzing chewing behavior based on chewing rhythm characteristics, chewing force characteristics, and chewing duration characteristics to determine the chewing analysis results of the target object.

[0191] It should be noted that this application embodiment performs behavioral analysis on the extracted chewing features to determine the chewing analysis results of the target object. This analysis process combines multiple features, and by comprehensively evaluating the user's chewing rhythm, force, and duration, this application embodiment can comprehensively understand the user's eating behavior. For example, by analyzing chewing rhythm and force, this application embodiment can determine whether the user has insufficient chewing; by analyzing chewing duration and the number of interruptions, this application embodiment can assess the user's eating continuity and focus. These analysis results not only provide users with personalized dietary suggestions but also provide a basis for health monitoring and behavioral intervention. Through this phased and multi-level processing approach, the method of this application can effectively solve the problems existing in the prior art and provide a more accurate, flexible, and adaptable chewing action recognition solution.

[0192] Reference Figure 8 , Figure 8 This illustration shows an ear-hook structure of a headphone for monitoring eating behavior according to an embodiment of this application, designed to monitor and analyze the user's chewing behavior. Figure 8The earphone shell is visible, providing not only physical protection but also housing all the internal components. Inside the earphones are several key electronic modules, each with its specific function, working together to achieve the earphones' intelligent monitoring capabilities.

[0193] Figure 8 The location of the vibration sensor is marked on the image. This sensor is the core component for monitoring chewing movements; it captures the minute vibrations generated when the user's jaw moves. These vibration signals are then sent to a processor for further analysis and processing. The precise placement and high sensitivity of the sensor are crucial for capturing the details of chewing movements, thus ensuring the accuracy of the monitoring data.

[0194] Secondly, the power module provides the necessary power to the headphones. This typically includes a small battery and possibly charging circuitry, ensuring the headphones can operate continuously for extended periods. The design of the power module needs to consider the size limitations of the headphones and the user's wearing comfort, while also ensuring a stable power supply to support all headphone functions.

[0195] The processor is the brain of the headphones; it's responsible for processing the data collected by the vibration sensors. The processor executes complex algorithms to analyze characteristics such as chewing rhythm, force, and timing, and transforms this information into useful monitoring results. The processor's performance directly impacts the speed and accuracy of data processing, therefore a balance between computing power and energy consumption must be struck during the design phase.

[0196] The memory is used to store data analyzed by the processor, as well as the headset's firmware or software programs. This can include temporary storage to buffer data streams and long-term storage to store the user's historical chewing data. The size and read / write speed of the memory affect the headset's data storage capacity and response speed.

[0197] The communication module is responsible for transmitting the data collected and processed by the earphones to the target terminal, such as a smartphone or tablet. Data transmission is achieved wirelessly, such as via Bluetooth. The design of the communication module needs to ensure the stability and security of data transmission, while also considering power consumption and compatibility issues.

[0198] Reference Figure 9 This illustration shows another in-ear structure of the dietary behavior monitoring earphone according to an embodiment of this application, including a shell, a power module, a processor, a memory, a vibration sensor, and a communication module. The power module provides power to the earphone, the processor processes chewing motion data captured by the vibration sensor, the memory stores data and programs, the vibration sensor is specifically used to monitor vibrations during chewing, and the communication module is responsible for wirelessly transmitting the processed data to an external device. These components work together to enable the earphone to monitor and analyze the user's dietary behavior in real time.

[0199] In conclusion, Figure 8 and Figure 9 The presentation showcases the key components and functions within the dietary behavior monitoring headset. The sophisticated design and collaborative work of these components enable the headset to accurately monitor the user's eating habits and effectively transmit the data to the target terminal for further analysis.

[0200] This application also provides a computer program product, which includes a computer program. The processor of a computer device reads and executes the computer program, causing the computer device to perform the chewing action recognition method described above.

[0201] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in this disclosure and the foregoing drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “including,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatuses.

[0202] It should be understood that in this disclosure, "at least one item" means one or more, and "more than one" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0203] It should be understood that in the description of the embodiments of this application, "multiple" means two or more, "greater than", "less than", "exceeding" etc. are understood to exclude the number itself, and "above", "below", "within" etc. are understood to include the number itself.

[0204] In the several embodiments provided in this disclosure, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection between apparatuses or units, and may be electrical, mechanical, or other forms.

[0205] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0206] Furthermore, the functional units in the various embodiments of this disclosure can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0207] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this disclosure, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this disclosure. The aforementioned storage medium may include: a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and other media capable of storing program code.

[0208] It should also be understood that the various implementation methods provided in this application can be combined arbitrarily to achieve different technical effects.

[0209] The above is a detailed description of the embodiments of this disclosure. However, this disclosure is not limited to the above embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of this disclosure. All such equivalent modifications or substitutions are included within the scope defined by the claims of this disclosure.

Claims

1. A method for recognizing chewing actions, characterized in that, Applied to target terminals, including: Target perception signal is received from the diet behavior monitoring earphone; wherein, the target perception signal is obtained in the diet behavior monitoring earphone through the following steps: vibration sensing is performed on the target object to obtain a perception data stream, and noise frequency band is filtered out from the perception data stream to obtain the target perception signal; Feature extraction is performed on the target perception signal to obtain the chewing rhythm feature, chewing force feature, and chewing duration feature of the target object; Chewing behavior analysis is performed on the chewing rhythm characteristics, chewing force characteristics, and chewing duration characteristics to determine the chewing analysis results of the target object. The step of performing chewing behavior analysis on the chewing rhythm features, chewing force features, and chewing duration features to determine the chewing analysis results of the target object includes: Based on the chewing rhythm characteristics, rhythm index analysis is performed to obtain the average chewing frequency and chewing frequency stability. Based on the chewing force characteristics, the force index is analyzed to obtain the average chewing energy intensity and amplitude variation coefficient. Based on the chewing time value characteristics, time value index analysis is performed to obtain the total chewing time, the percentage of effective chewing time, and the number of chewing interruptions. The average chewing frequency, the stability of the chewing frequency, the average chewing energy intensity, the amplitude variation coefficient, the total chewing time, the proportion of effective chewing time, and the number of chewing interruptions are subjected to feature fusion processing to obtain the target chewing feature vector. Obtain a dietary behavior pattern library; wherein, the dietary behavior pattern library stores corresponding chewing pattern representation vectors for various dietary behavior patterns; The similarity or distance between the target chewing feature vector and each chewing pattern representation vector is compared to select the hit pattern representation vector from the dietary behavior pattern library. Based on the dietary behavior pattern corresponding to the hit pattern representation vector, the chewing analysis result is generated; wherein, the dietary behavior pattern includes a behavior pattern of eating quickly and not chewing sufficiently, and the chewing analysis result includes the conclusion that the user has problems of eating quickly and not chewing sufficiently during the eating process.

2. The method according to claim 1, characterized in that, The step of extracting features from the target sensing signal to obtain the chewing rhythm features, chewing force features, and chewing duration features of the target object includes: Based on the target sensing signal, a target vibration sensing spectrum is generated; Spectral dynamic analysis is performed on the target vibration sensing spectrum to determine the chewing rhythm characteristics of the target sensing signal; Spectral amplitude analysis is performed on the target vibration sensing spectrum to determine the chewing force characteristics of the target sensing signal; Spectral time value analysis is performed on the target vibration sensing spectrum to determine the chewing time value characteristics of the target sensing signal.

3. The method according to claim 2, characterized in that, The step of performing spectral dynamic analysis on the target vibration sensing spectrum to determine the chewing rhythm characteristics of the target sensing signal includes: The vibration sensing spectrum is divided based on a predetermined first division interval to obtain multiple dynamic analysis frames; Extract the corresponding vibration main frequency data from each of the multiple dynamic analysis frames; Dynamic feature analysis is performed on the vibration main frequency data of multiple consecutive dynamic analysis frames to obtain the dynamic features of the main frequency data as the chewing rhythm features.

4. The method according to claim 2, characterized in that, The step of performing spectral amplitude analysis on the target vibration sensing spectrum to determine the chewing force characteristics of the target sensing signal includes: Select a target spectrum segment from the target vibration sensing spectrum; The target spectrum segment is divided based on a predetermined second division interval to obtain multiple spectrum sub-segments; Amplitude analysis is performed on each of the aforementioned spectral sub-segments to obtain the corresponding local amplitude; Amplitude variation calculations are performed on the local amplitudes of multiple consecutive spectral sub-segments to obtain amplitude variation coefficients; The chewing force feature is generated based on the amplitude variation coefficient and the local amplitude of each of the spectral sub-segments.

5. The method according to claim 2, characterized in that, The step of performing spectral time-value analysis on the target vibration sensing spectrum to determine the chewing time-value characteristics of the target sensing signal includes: Select a target spectrum segment from the target vibration sensing spectrum; Based on preset chewing spectrum discrimination conditions, the target spectrum segment is divided into a chewing occurrence sub-segment and a chewing cessation sub-segment; Time-value analysis is performed based on the distribution characteristics of the chewing occurrence sub-segment and the chewing cessation sub-segment in the target spectral segment to determine the chewing time-value characteristics of the target sensing signal.

6. A computer-readable storage medium, characterized in that, The storage medium stores a program, which is executed by a processor to implement the chewing action recognition method as described in any one of claims 1 to 5.

Images