CN118098501A - A calorie analysis and matching system based on IoT devices - Google Patents

Abstract

The present invention discloses a calorie analysis and matching system based on an Internet of Things device, comprising a database, an information collection unit, a processing and analysis unit, and an interactive planning unit. The information collection unit collects food image information and a user's body function parameters, firstly constructs a food comparison model to classify food in different scenes, generates corresponding food types, and then obtains corresponding nutrients and volume sizes to generate food calories; further, through analysis of body function parameters, a first body surface coefficient and a first motion coefficient are generated, and a four-dimensional monitoring vector is constructed and analyzed to judge the degree of exercise consumption, thereby generating consumed calories; finally, the food calories and the consumed calories are analyzed to realize health planning, and help users adjust their eating and exercise habits in a timely manner.

Description

A calorie analysis and matching system based on IoT devices

Technical Field

The present invention relates to the field of health management technology, and in particular to a calorie analysis and matching system based on Internet of Things devices.

Background technique

With the continuous development of society, people are gradually pursuing a goal of balanced diet and reasonable weight in health management. Calories are the unit for measuring the calorie consumption of human exercise. When users eat, they often need to manually calculate the calories they consume. For some foods such as dishes, they need to be identified and distinguished, which greatly increases the calculation time. Existing exercise reminder functions are often installed on wearable devices. Wearable devices only remind people to exercise according to existing model algorithms. When food calories are consumed too much or too little, the calorie differences caused by personal physique are not taken into account, and personalized diet and exercise plans cannot be provided, affecting the final health effect.

In view of the above technical defects, a solution is now proposed.

Summary of the invention

The purpose of the present invention is to propose a calorie analysis and matching system based on Internet of Things devices. By taking pictures of food, the food can be accurately identified, and the calories of the food can be calculated through an algorithm according to the type and volume of the food consumed and the corresponding nutritional composition table; at the same time, the user's intake and consumption can be monitored in real time, and real-time reminders and suggestions can be provided according to the set goals to help users adjust their eating and exercise habits in a timely manner.

In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: a calorie analysis and matching system based on Internet of Things devices, including a database, an information acquisition unit, a processing and analysis unit and an interactive planning unit; the information acquisition unit includes an image acquisition module and a data acquisition module; the image acquisition module acquires images of food, and the data acquisition module acquires the user's body function parameters, and both are sent to the database for storage; the processing and analysis unit is connected to the information acquisition unit, and is connected to the database of the system background and performs corresponding data transmission; it includes a food recognition processing module and a motion data processing module, the food recognition processing module acquires and analyzes the various parameter information of the target food stored in the database, establishes a food comparison model, and calculates the calories of the food; the motion data processing module acquires and analyzes the user's body function parameters stored in the database, establishes a motion consumption model, and calculates the calories consumed; the calories of the food and the calories consumed are also sent to the database for storage; the interactive planning unit is connected to the processing and analysis unit, and is connected to the database of the system background and performs corresponding data transmission, and realizes health planning by analyzing the calories of the food and the calories consumed.

Furthermore, the specific analysis process of the food identification processing module is as follows:

Step A1, first establish a food comparison model, and retrieve the food type determination table in the database, and then compare the similarity between the target food and the food type determination table of the target food, and the preset food type determination table data includes food type data in different scenes, and then obtain the food type of the target food according to the matching similarity in the picture comparison process;

Step A2, obtaining the food type of the target food, and then obtaining the corresponding nutritional components and volume according to the food type, and constructing an impact analysis model to analyze the nutritional components and volume, and obtain the impact coefficients respectively; wherein the nutritional components include energy (kJ), sodium (mg), protein (g), fat (g), carbohydrate (g), calcium (mg), iron (mg) and organic acid (mg) contained in each ingredient per 100 grams

Step A2-1, first build an impact analysis model:

The input information is marked as a parameter set U, which contains N element values. Any element value is marked as Xi, and then the N element values are respectively assigned corresponding weight coefficients φi to establish the influence coefficient Ue of the formula output parameter set U;

Step A2-2, then substitute the nutrients and volume into the impact analysis model in sequence to obtain the nutrient impact coefficient and volume impact coefficient, and mark them as U1 and U2 respectively;

Step A3, calculating the calories K0 of the target food according to the target food type, the volume corresponding to the target food type and the nutritional components.

Furthermore, the specific process of establishing the food comparison model is as follows:

Step A1-1, obtaining a comparison scene of the target food, framing the target food in an area, and then obtaining color features of the target food and the preset judgment food, the color features including the red element R, the green element L and the blue element B, establishing a spatial rectangular coordinate system with the same base point, and the horizontal axis of the spatial rectangular coordinate system is the color space of R/G (red element/green element), and the vertical axis is the color space of B/G (blue element/green element);

Step A1-2, obtaining the distribution quantity m1 and the distribution quantity m2 of the target food and the preset judgment food in the color space, and comparing the distribution quantity m1 with the distribution quantity m2:

When m1=m2, it means that the target food completely matches the preset judgment food, and the matching similarity is 100%;

When m1≠m2, it means that the target food does not completely match the preset judgment food, then the difference data between the distribution quantities is recorded, the model is trained again, and the set Q of the difference data is obtained, and the minimum value Qmin of the set Q is obtained accordingly, and the matching similarity at this time is calculated;

Step A1-3, setting a threshold for matching similarity, and comparing the matching similarity with the threshold preset in the database:

When the matching similarity is less than the set threshold, the network parameters are continuously changed and the model training is re-performed;

When the matching similarity is greater than the set threshold, the adopted model is saved and a food comparison model is generated.

Furthermore, the specific analysis process of the motion data processing module is as follows:

Step B1, first obtaining the user's body function parameters at the current time point, the body function parameters including the first function parameters and the second function parameters, and then sequentially processing and analyzing the first function parameters and the second function parameters;

Step B1-1, the first functional parameter includes the user's motion speed direction and motion speed value, and a motion posture analysis model is constructed to analyze the first functional parameter:

First, the direction of the motion speed is decomposed: taking the three-dimensional coordinate axis as the reference, the angles between the motion speed direction and the X-axis direction, Y-axis direction and Z-axis direction of the three-dimensional coordinate axis are marked as inclination angle θ1, inclination angle θ2 and inclination angle θ3 (θ1, θ2 and θ3 are all between 0° and 90°); then the motion speed value is mapped according to the motion speed direction to obtain the sub-speed value, and marked as speed value v1, speed value v2 and speed value v3 respectively, and the speed value is combined with the inclination angle to generate the first motion coefficient D1 at the current time point;

Step B1-2, the second functional parameters include the user's heart rate, temperature and blood oxygen concentration, which are marked as c1, c2 and c3 respectively, and then the weight coefficients corresponding to the heart rate, temperature and blood oxygen concentration are assigned to generate the first body surface coefficient D2;

Step B2, divide a day into multiple time periods, and obtain the first motion coefficient D1, the first body surface coefficient D2 and the motion category S to construct a four-dimensional monitoring vector <first motion coefficient D1, first body surface coefficient D2, motion category S, time Time>, count the average value and fluctuation value of the first motion coefficient D1 and the second motion coefficient D2 in each time period, for any time period Ta, mark the motion category of the time period Ta as Sa, the average values of the first motion coefficient D1 and the first body surface coefficient D2 are T _HD1 and T _HD2 respectively, and the fluctuation values are T _RD1 and T _RD2 respectively, and construct judgment vectors according to different combinations;

Step B3, establish a motion consumption model: mark the input information as vector W, vector W contains N0 component vectors, mark any component vector as Yj, and then assign corresponding consumption coefficients σj to each of the N0 component vectors, and establish a formula to output the judgment value Wd of the vector W;

Then, any judgment vector is input into the motion consumption model to obtain the corresponding judgment value, and a threshold of the judgment value is set, and the judgment value is compared and analyzed with the preset threshold to determine the degree of motion consumption;

Step B4, obtaining the user's weight, and setting a formula to calculate the consumed calories K1.

Furthermore, the specific generation process of the motion category is as follows:

Get the motion speed value and sub-speed value within the preset time, set the threshold vo1 and threshold vo2, compare and analyze the sub-speed value with the preset threshold, and generate the corresponding motion category S:

When v3>v1 and v3>v2, the movement speed value at this time is marked as s2, edited as a secondary character, and s2 and the secondary character are combined to generate a secondary movement category;

Otherwise, when v3>preset value vo1 and v1>preset value vo2, the motion speed value at this time is marked as s1, edited as a first-level character, and s1 and the first-level character are combined to generate a first-level motion category;

Otherwise, the motion speed value at this time is marked as s3, and edited as a third-level character, and s3 and the third-level character are combined to generate a third-level motion category.

Furthermore, the specific analysis process of health planning is as follows:

Step C1, using the calories of food K0 as the horizontal coordinate and the calories consumed K1 as the vertical coordinate, establish a calorie consumption change relationship diagram, and record the turning point moment and mark it as a change point; at the same time, preset an ideal threshold curve in the coordinate system;

Step C2, respectively obtaining the length of the line segment of the food calorie K0-consumed calorie K1 curve located above the preset threshold curve and the area enclosed by the upper line segment and the preset threshold curve, and the length of the line segment of the food calorie K0-consumed calorie K1 curve located below the preset threshold curve and the area enclosed by the lower line segment and the preset threshold curve;

Step C3, draw a circle with the change point as the center and r as the radius. If the enclosed area mj1 or area mj2 does not fall completely within the circle, a reminder signal is generated to carry out health planning.

In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

The present invention first collects food image information and the user's body function parameters through an information collection unit, and then constructs a food comparison model to classify food in different scenarios, generate corresponding food types, and then obtain corresponding nutrients and volume sizes to generate food calories; it also generates a first body surface coefficient and a first motion coefficient through analysis of body function parameters, and constructs and analyzes a four-dimensional monitoring vector to judge the degree of exercise consumption, and then generates consumed calories; finally, the calories of food and the calories consumed are analyzed to achieve health planning, and help users adjust their eating and exercise habits in time.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 shows a schematic structural diagram of the present invention;

FIG. 2 shows a system flow chart of the present invention.

Detailed ways

The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

Embodiment 1: As shown in FIG1-2, a calorie analysis and matching system based on an IoT device includes a database, an information acquisition unit, a processing and analysis unit, and an interactive planning unit; the information acquisition unit includes an image acquisition module and a data acquisition module; the image acquisition module acquires images of food, and the data acquisition module acquires body function parameters of the user, and both are sent to the database for storage;

The processing and analysis unit is connected to the information collection unit and is connected to the database of the system background to perform corresponding data transmission; it includes a food recognition processing module and a motion data processing module. The food recognition processing module acquires and analyzes the various parameter information of the target food stored in the database, establishes a food comparison model, and calculates the calories of the food; the motion data processing module acquires and analyzes the user's body function parameters stored in the database, establishes a motion consumption model, and calculates the calories consumed; and also sends the food calories and the consumed calories to the database for storage;

Among them, the specific analysis process of the food identification processing module is as follows:

Step A1, first establish a food comparison model, and retrieve the food type determination table in the database, and then compare the similarity between the target food and the food type determination table of the target food, and the preset food type determination table data includes food type data in different scenes, and then obtain the food type of the target food according to the matching similarity in the picture comparison process;

The process of establishing the food comparison model is as follows:

Step A1-1, obtaining a comparison scene of the target food, framing the target food in an area, and then obtaining color features of the target food and the preset judgment food, the color features including the red element R, the green element L and the blue element B, establishing a spatial rectangular coordinate system with the same base point, and the horizontal axis of the spatial rectangular coordinate system is the color space of R/G (red element/green element), and the vertical axis is the color space of B/G (blue element/green element);

Step A1-2, obtaining the distribution quantity m1 and the distribution quantity m2 of the target food and the preset judgment food in the color space, and comparing the distribution quantity m1 with the distribution quantity m2:

When m1=m2, it means that the target food completely matches the preset judgment food, and the matching similarity is 100%;

When m1≠m2, it means that the target food does not completely match the preset judgment food, then the difference data between the distribution quantities is recorded, the model is trained again, and the set Q of the difference data is obtained, and the minimum value Qmin of the set Q is obtained accordingly, and the matching similarity at this time is calculated;

Step A1-3, setting a threshold for matching similarity, and comparing the matching similarity with the preset threshold:

When the matching similarity is less than the set threshold, the network parameters are continuously changed, such as the kernel size, number of kernels, moving step and padding number of the convolution layer, and the model training is re-performed;

When the matching similarity is greater than the set threshold, the adopted model is saved and a food comparison model is generated;

It should be noted that the model can often achieve a high recognition accuracy on the training set when learning on the training set. Ideally, the model can converge well on a large amount of training data. If the accuracy on the training data is high, it can adapt well to the test set and have a good test accuracy. In fact, there is often a certain gap between the accuracy on the training set and the test set. This is because the model is overfitted to the training data. The learning of the training data is no longer a single category of pattern recognition, but a memorization of a single sample. The validation set can be used to set the early stopping mechanism in the model learning process, which can prevent the model from overfitting to the training data to a certain extent.

Step A2, obtaining the food type of the target food, and then obtaining the corresponding nutritional components and the volume occupied according to the food type, and constructing an impact analysis model to analyze the nutritional components and the volume, and respectively obtain the impact coefficients; wherein the nutritional components include energy (kJ), sodium (mg), protein (g), fat (g), carbohydrate (g), calcium (mg), iron (mg) and organic acid (mg) contained in each component per 100 grams, and are marked as a1, a2, a3, a4, a5, a6, a7 and a8 respectively;

Step A2-1, first build an impact analysis model:

The input information is marked as a parameter set U, which contains N element values. Any element value is marked as Xi, and then the N element values are respectively assigned corresponding weight coefficients φi, and the influence coefficient Ue of the formula output parameter set U is established. The formula is:

Step A2-2, then substitute the nutrients and volume into the impact analysis model in sequence to obtain the nutrient influence coefficient and volume influence coefficient, and mark them as U1 and U2 respectively;

A2-21, taking nutrients as input information, establish Ua set = {energy a1, sodium a2, protein a3, fat a4, carbohydrate a5, calcium a6, iron a7, organic acid a8}, substitute Ua set into the impact analysis model, obtain the nutrient impact coefficient U1, the formula is: U1 = a1*β1+a2*β2+a3*β3+a4*β4+a5*β5+a6*β6+a7*β7+a8*β8, where β1, β2, β3, β4, β5, β6, β7, β8 are the weight coefficients of energy, sodium, protein, fat, carbohydrate, calcium, iron, and organic acid per 100 grams of each ingredient, respectively, and β1, β2, β3, β4, β5, β6, β7, β8 are all greater than 0;

A2-22, take the corresponding volume of the same type of target food as input information, establish Ub set = {volume b1, volume b2, ..., volume bN}, where bN represents the volume of the Nth block of the food type, and N is an integer greater than 0; substitute Ub set into the influence analysis model to obtain the volume influence coefficient U2, the formula is: Where i is the number of volume blocks, i is a positive integer and i∈[1,N], χi is the weight coefficient corresponding to the i-th volume block, and χi is greater than 0;

Step A3, according to the target food type, the volume corresponding to the target food type and the nutrient content, the calorie K0 of the target food is calculated, and the formula is: K0 = ∑(U2*U1*Nu), where U1 is the nutrient content, U2 is the volume, and Nu is the food type;

Among them, the specific analysis process of the motion data processing module is:

Step B1, first obtaining the user's body function parameters at the current time point, the body function parameters including the first function parameters and the second function parameters, and then sequentially processing and analyzing the first function parameters and the second function parameters;

Step B1-1, the first functional parameter includes the user's motion speed direction and motion speed value, and a motion posture analysis model is constructed to analyze the first functional parameter:

First, the direction of the motion speed is decomposed: taking the three-dimensional coordinate axis as the reference, the angles between the motion speed direction and the X-axis direction, Y-axis direction and Z-axis direction of the three-dimensional coordinate axis are marked as inclination angle θ1, inclination angle θ2 and inclination angle θ3 (θ1, θ2 and θ3 are all between 0° and 90°); then the motion speed value is mapped according to the motion speed direction to obtain the sub-speed value, and marked as speed value v1, speed value v2 and speed value v3 respectively, and the speed value is combined with the inclination angle to generate the first motion coefficient D1 at the current time point; the formula is: D1=v1*cosθ1*η1+v2*cosθ2*η2+v3*cosθ3*η3, wherein η1, η2 and η3 are conversion coefficients respectively, and the conversion coefficients are preset through a large amount of data measurement, and η1, η2 and η3 are all greater than 0; when the speed value is higher, the inclination angle is larger, the corresponding cosine value is larger, and the first motion coefficient is larger;

Step B1-2, the second functional parameters include the user's heart rate, temperature and blood oxygen concentration, which are marked as c1, c2 and c3 respectively, and then the weight coefficients corresponding to the heart rate, temperature and blood oxygen concentration are assigned to generate the first body surface coefficient D2; the formula is: Among them, δ1, δ2 and δ3 are weight coefficients of heart rate, temperature and blood oxygen concentration, respectively, and δ1, δ2 and δ3 are all greater than 0; when the heart rate is higher, the temperature is higher, and the blood oxygen concentration is lower, the larger the first body surface coefficient is, indicating that the exercise at the current time point is more intense;

It should be noted that the ear cavity temperature value is detected by the ear cavity temperature and humidity sensor, and the user's motion state is detected by the three-axis acceleration sensor;

Step B2, dividing a day into multiple time periods, and obtaining the first motion coefficient D1, the first body surface coefficient D2 and the motion category S to construct a four-dimensional monitoring vector <first motion coefficient D1, first body surface coefficient D2, motion category S, time Time>, counting the average value and fluctuation value of the first motion coefficient D1 and the second motion coefficient D2 in each time period, for any time period Ta, marking the motion category of the time period Ta as Sa, the average values of the first motion coefficient D1 and the first body surface coefficient D2 are T _HD1 and T _HD2 respectively, and the fluctuation values are T _RD1 and T _RD2 respectively, and four judgment vectors are constructed according to different combinations;

<T _HD1 +T _HD2 , T _RD1 +T _RD2 , Sa>;

<T _HD1 +T _HD2 T _RD1 -T _RD2 , Sa>;

<T _HD1 -T _HD2 , T _RD1 +T _RD2 , Sa>;

<T _HD1 -T _HD2 , T _RD1 -T _RD2 , Sa>;

The specific generation process of the motion category is as follows:

Get the motion speed value and sub-speed value within the preset time, set the threshold vo1 and threshold vo2, compare and analyze the sub-speed value with the preset threshold, and generate the corresponding motion category S:

In this embodiment, taking upright walking as an example:

When v3>v1 and v3>v2, the movement speed value at this time is marked as s2, and edited as a secondary character at the same time, and s2 and the secondary character are combined to generate a secondary movement category, indicating stepping;

Otherwise, when v3>preset value vo1 and v1>preset value vo2, the movement speed value at this time is marked as s1, edited as a first-level character, and s1 and the first-level character are combined to generate a first-level movement category, indicating running;

Otherwise, the movement speed value at this time is marked as s3, and edited as a third-level character, and s3 and the third-level character are combined to generate a third-level movement category, indicating walking;

Step B3, establish a motion consumption model: mark the input information as vector W, vector W contains N0 sub-vectors, mark any sub-vector as Yj, and then assign corresponding consumption coefficients σj to each of the N0 sub-vectors, and establish a formula to output the judgment value Wd of the vector W, the formula is:

Then input any judgment vector into the motion consumption model to obtain the corresponding judgment value, set the threshold of the judgment value, and compare the judgment value with the preset threshold:

When the output judgment value is greater than the preset threshold, it indicates that the degree of exercise consumption is too great;

When the output judgment value is less than the preset threshold, it means that the exercise consumption is too small;

It should be noted that the degree of exercise consumption changes over time, so the exercise consumption period can be recalculated once a day. Too much or too little exercise consumption is not good for the human body.

Step B4, obtaining the user's weight, and setting the consumed calories K1 formula as: K1 = ∑ (weight*Jg);

The interactive planning unit is connected to the processing and analysis unit, and is connected to the database of the system background and performs corresponding data transmission, so as to realize health planning by analyzing the calories of food and the calories consumed;

Step C1, using the calories of food K0 as the horizontal coordinate and the calories consumed K1 as the vertical coordinate, establish a calorie consumption change relationship diagram, and record the turning point moment and mark it as a change point; at the same time, preset an ideal threshold curve in the coordinate system;

Step C2, respectively obtaining the length of the line segment above the preset threshold curve and the area enclosed by the upper line segment and the preset threshold curve of the food calories K0-consumed calories K1 curve, and marking them as xd1 and mj1 respectively, and the length of the line segment below the preset threshold curve and the area enclosed by the lower line segment and the preset threshold curve of the food calories K0-consumed calories K1 curve, and marking them as xd2 and mj2 respectively;

Step C3, draw a circle with the change point as the center and r as the radius. If the enclosed area mj1 or area mj2 does not completely fall within the circle, a reminder signal is generated to perform health planning:

When the enclosed area is in the mj1 area, the reminder signal text is edited and displayed, and the text content is: "It is necessary to reduce food intake or increase exercise consumption at this time"; when the enclosed area is in the mj2 area, the reminder signal text is edited and displayed, and the text content is: "It is necessary to increase food intake or reduce exercise consumption at this time";

The user's prompt information obtained through analysis is then fed back through a display terminal (such as the curved display screen of a Bluetooth headset). It should be noted that the Bluetooth headset device communicates with the mobile phone in real time, and the user can also view and process the prompt information through the mobile phone APP, thereby helping the user to adjust his or her eating and exercise habits in a timely manner.

In summary of the above technical solutions: the present invention includes a database, an information collection unit, a processing and analysis unit, and an interactive planning unit. The information collection unit collects food image information and the user's body function parameters, and first constructs a food comparison model to classify food in different scenarios, generate corresponding food types, and then obtain corresponding nutrients and volume sizes, and generate food calories; also through the analysis of body function parameters, a first body surface coefficient and a first motion coefficient are generated, and a four-dimensional monitoring vector is constructed and analyzed to judge the degree of exercise consumption, and then the consumed calories are generated; finally, the calories of food and the calories consumed are analyzed to achieve health planning, which helps users adjust their eating and exercise habits in time.

The size of the interval and threshold is set for the convenience of comparison. The size of the threshold depends on the amount of sample data and the number of bases set by technicians in this field for each group of sample data; as long as it does not affect the proportional relationship between the parameter and the quantized value;

The above formulas are all dimensionless and numerical calculations. The formula is a formula obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims (6)

1. Calorie analysis matching system based on thing networking equipment, its characterized in that: the system comprises a database, an information acquisition unit, a processing analysis unit and an interactive planning unit;

the information acquisition unit comprises an image acquisition module and a data acquisition module; the image acquisition module acquires images of food, and the data acquisition module acquires physical function parameters of a user and sends the physical function parameters to the database for storage;

The processing analysis unit is connected with the information acquisition unit, is connected with a database of a system background and performs corresponding data transmission; the food identifying and processing module is used for acquiring and analyzing various parameter information of target food stored in a database, establishing a food comparison model and calculating calories of the food; the exercise data processing module is used for establishing an exercise consumption model and calculating consumed calories by acquiring and analyzing physical function parameters of a user stored in the database; the calories of the food and the calories consumed are also transmitted to a database for storage;

The interactive planning unit is connected with the processing analysis unit, is connected with a database of a system background and performs corresponding data transmission, and realizes health planning by analyzing calories of food and calories consumed;

And a database for storing a food type judgment table of the target food.

2. The internet of things device-based calorie analysis matching system of claim 1, wherein: the specific analysis process of the food recognition processing module is as follows:

Step A1, firstly establishing a food comparison model, and retrieving a food type judgment table in a database, so as to compare the similarity of the target food and the food type judgment table of the target food, wherein the preset food type judgment table data comprise food type data in different scenes, and then obtaining the food type of the target food according to the matching similarity in the picture comparison process;

A2, obtaining food types of target foods, obtaining corresponding nutritional components and occupied volumes according to the food types, constructing an influence analysis model, analyzing the nutritional components and the volumes, and respectively obtaining influence coefficients; wherein the nutritional ingredients comprise energy (kJ), sodium (mg), protein (g), oil (g), carbohydrate (g), calcium (mg), iron (mg) and organic acid (mg) contained in each hundred grams of each ingredient;

step A2-1, firstly constructing an influence analysis model:

Marking input information as a parameter set U, wherein the set U comprises N element values, marking any element value as Xi, respectively endowing the N element values with corresponding weight coefficients phi, and establishing an influence coefficient Ue of the formula output parameter set U;

A2-2, substituting the nutritional ingredients and the volumes into an influence analysis model in sequence to obtain nutritional ingredient influence coefficients and volume influence coefficients respectively, and marking the nutritional ingredient influence coefficients and the volume influence coefficients as U1 and U2 respectively;

and A3, calculating the calories K0 of the target food according to the type of the target food and the corresponding volumes and nutritional components of the type of the target food.

3. The calorie analysis matching system based on the internet of things equipment according to claim 2, wherein: the specific establishment process of the food comparison model comprises the following steps:

Step A1-1, obtaining a contrast scene of target food, and obtaining color characteristics of the target food and preset judging food by carrying out regional framing on the target food, wherein the color characteristics comprise a red element R, a green element L and a blue element B, a space rectangular coordinate system is established by the same base point, the transverse axis of the space rectangular coordinate system is a color space of R/G (red element/green element), and the longitudinal axis is a color space of B/G (blue element/green element);

step A1-2, obtaining the distribution quantity m1 and the distribution quantity m2 of the target food and the preset judgment food in the color space, and comparing the distribution quantity m1 with the distribution quantity m 2:

When m1=m2, the target food is completely matched with the preset judgment food, and the matching similarity is 100%;

When m1 is not equal to m2, indicating that the target food is not completely matched with the preset judging food, recording difference data between distribution quantity, performing model training again, acquiring a set Q of the difference data, correspondingly acquiring a minimum value Qmin of the set Q, and calculating the matching similarity at the moment;

Step A1-3, setting a threshold value of the matching similarity, and comparing the matching similarity with a preset threshold value:

when the matching similarity is smaller than a set threshold value, continuously changing network parameters, and carrying out model training again;

and when the matching similarity is larger than a set threshold value, storing the adopted model to generate a food comparison model.

4. The internet of things device-based calorie analysis matching system of claim 1, wherein: the specific analysis process of the motion data processing module is as follows:

step B1, firstly acquiring physical function parameters of a user at the current time point, wherein the physical function parameters comprise a first function parameter and a second function parameter, and then sequentially carrying out processing analysis on the first function parameter and the second function parameter;

Step B1-1, the first functional parameters comprise a user movement speed direction and a movement speed value, and a movement gesture analysis model is constructed to analyze the first functional parameters:

The motion speed direction is decomposed firstly: respectively marking the included angles between the motion speed direction and the X-axis direction, the Y-axis direction and the Z-axis direction of the three-dimensional coordinate axis as an inclination angle theta 1, an inclination angle theta 2 and an inclination angle theta 3 (theta 1, theta 2 and theta 3 are all between 0 DEG and 90 DEG) by taking the three-dimensional coordinate axis as a reference; mapping the motion speed value according to the motion speed direction to obtain a sub-speed value, correspondingly marking the sub-speed value as a speed value v1, a speed value v2 and a speed value v3, and combining the speed value with the inclination angle to generate a first motion coefficient D1 of the current time point;

step B1-2, wherein the second functional parameters comprise heart rate, temperature and blood oxygen concentration of the user, are correspondingly marked as c1, c2 and c3, and are endowed with weight coefficients corresponding to the heart rate, the temperature and the blood oxygen concentration, so as to generate a first body table coefficient D2;

Step B2, dividing a day into a plurality of Time periods, acquiring a first motion coefficient D1, a first body table coefficient D2 and a motion class S to construct a four-dimensional monitoring vector < first motion coefficient D1, the first body table coefficient D2, the motion class S and a Time Time >, counting the average value and the fluctuation value of the first motion coefficient D1 and the second motion coefficient D2 in each Time period, for any Time period Ta, marking the motion class of the Time period Ta as Sa, the average value of the first motion coefficient D1 and the first body table coefficient D2 as T _HD1 and T _HD2 respectively, the fluctuation value as T _RD1 and T _RD2 respectively, and constructing a judgment vector according to different combinations;

step B3, a motion consumption model is built again: marking input information as a vector W, wherein the vector W contains N0 components, marking any component as Yj, respectively endowing the N0 components with corresponding consumption coefficients sigma j, and establishing a judgment value Wd of the formula output vector W;

inputting any judgment vector into the motion consumption model to obtain a corresponding judgment value, setting a threshold value of the judgment value, comparing and analyzing the judgment value with a preset threshold value, and judging the motion consumption degree;

And step B4, acquiring the weight of the user, and setting a formula to calculate the consumed calories K1.

5. The internet of things device-based calorie analysis matching system of claim 7, wherein: the specific generation process of the motion category comprises the following steps:

Acquiring a motion speed value and a minute speed value within a preset time, setting a threshold vo1 and a threshold vo2, comparing the minute speed value with the preset threshold, and analyzing to generate a corresponding motion category S:

When v3 is larger than v1 and v3 is larger than v2, marking the motion speed value at the moment as s2, editing the motion speed value into a secondary character, and combining the s2 and the secondary character to generate a secondary motion category;

Otherwise, when v3 is larger than a preset value vo1 and v1 is larger than a preset value vo2, marking the motion speed value at the moment as s1, editing the motion speed value as a first-level character, and combining the s1 and the first-level character to generate a first-level motion category;

Otherwise, the motion speed value at the moment is marked as s3, the motion speed value is edited into three-level characters at the same time, and the three-level motion category is generated by combining the s3 and the three-level characters.

6. The internet of things device-based calorie analysis matching system of claim 1, wherein: the specific analysis process of the health plan is as follows:

Step C1, using calories K0 of food as an abscissa and calories K1 consumed as an ordinate, establishing a calorie consumption change relation graph, recording inflection point moments, and marking the inflection point moments as change points; meanwhile, an ideal threshold curve is preset in a coordinate system;

Step C2, respectively obtaining the length of a line segment above a preset threshold curve and the area surrounded by the upper line segment and the preset threshold curve of the calorie K0-consumed calorie K1 curve of the food, and the length of a line segment below the preset threshold curve and the area surrounded by the lower line segment and the preset threshold curve of the calorie K0-consumed calorie K1 curve of the food;

and C3, drawing a circle outwards by taking the changing point as the center of a circle and taking r as the radius, and if the enclosed area mj1 or area mj2 does not completely fall in the circle, generating a reminding signal and carrying out health planning.