Non-contact vital sign monitoring method and system based on thermal infrared imaging

Abstract

The invention discloses a non-contact vital sign monitoring method and system based on thermal infrared imaging, relates to the technical field of infrared imaging, and is used for solving the problem of how to effectively utilize temperature data of thermal infrared imaging to analyze training effects of different sports items on all parts of a body. According to the method, the temperature difference, the variance and the mean value are quantitatively analyzed through an entropy method, and the temperature change of each part before and after exercise is systematically evaluated. The method not only can reflect the training intensity and effect of the body parts, but also can recommend more targeted sports items for the user through the sorting result; according to the scheme, through correlation analysis of temperature data and training indexes, an evaluation model for exercise item recommendation is constructed, and the training scheme can be optimized according to the influence degree of different exercises on all parts. The problem that traditional exercise recommendation lacks pertinence and scientific basis is solved.

Description

Technical Field

[0001] The present invention relates to the field of infrared imaging technology, and more specifically, to a non-contact vital sign monitoring method and system based on thermal infrared imaging. Background Art

[0002] In recent years, non-contact vital sign monitoring technology has received extensive attention in the fields of sports health, medical rehabilitation, and sports training. Traditional contact measurement means, such as heart rate belts, thermometers, etc., although they can accurately measure vital sign parameters, have many deficiencies, such as poor comfort, being easily affected by sweat or sliding during exercise. In addition, contact devices require frequent calibration and maintenance, and in a sports scenario, using contact devices may interfere with normal sports performance.

[0003] As a non-contact measurement means, thermal infrared imaging technology can capture the thermal radiation on the human body surface, generate a temperature distribution image, and then extract vital sign information. Due to its non-contact, high-resolution, and high-sensitivity characteristics, this technology has gradually become an emerging direction for vital sign monitoring and evaluation. However, because thermal infrared imaging devices are sensitive to environmental temperature, they are often affected by environmental factors such as direct sunlight and heat source interference in practical applications, resulting in inaccurate measurement results. In addition, how to effectively utilize the temperature data of thermal infrared imaging, analyze the training effects of different sports on various parts of the body, and optimize the sports plan through a scientific recommendation algorithm is still a technical difficulty.

[0004] In view of the above problems, the present invention proposes a solution. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a non-contact vital sign monitoring method and system based on thermal infrared imaging to solve the problems raised in the above background art.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A non-contact vital sign monitoring method based on thermal infrared imaging includes the following steps:

[0008] Calibrate the temperature of the thermal infrared imaging device;

[0009] Utilize the temperature distribution information in the thermal infrared image to extract the temperature data of the user's body surface through image processing technology;

[0010] Construct a body temperature calibration model with the obtained temperature data to calibrate the temperature data, and calculate the influence degree of sports items on the training indicators of various parts of the body based on the entropy method;

[0011] Recommend sports events that have a great impact on training indicators.

[0012] In a preferred embodiment, the obtained temperature data is used to construct a body temperature calibration model to calibrate the temperature data, which specifically includes the following steps:

[0013] The initial temperature measured by the uncalibrated infrared thermometer is T m , the surface temperature of the part to be measured is T 测 , the ambient temperature is T temp , and the final calibration result is T 实 ;

[0014] First, there is a difference kb between the T m measured by the infrared thermometer and the final calibration result of T 实 , and kb is the calibration value; that is: T 实 = T m + kb;

[0015] The difference b is obtained by summing and averaging the temperature values T i measured by the infrared thermometer multiple times, and then subtracting the surface temperature T 测 of the part to be measured from the average value; that is:

[0016] The coefficient k before b is calculated using the covariance formula to calculate the overall error T m , T temp of T 封 and the encapsulation temperature T m - T temp , T m - T 封 , let T 封 = 0, that is

[0017] Finally, the formula is integrated to obtain the body temperature calibration model, that is

[0018]

[0019] In a preferred embodiment, the method for extracting the temperature data of the user's body surface is as follows:

[0020] The user's body is divided into multiple parts. According to the human body proportion, the standing person image is simplified into a rectangle. Let the height of the person be h and the shoulder width be j, and a rectangular coordinate system is established with the lower left corner of the rectangle as the origin;

[0021] The coordinates of each body part are obtained, where a = h / 9, b = j / 3

[0022] Let the minimum temperature difference of the i-th body part be T imin, the maximum temperature difference is T imax , the variance of the temperature difference is T istd , the average value of the temperature difference is T iave ; the temperature difference is the difference in temperatures measured at the same position before and after exercise;

[0023] The method for analyzing temperature data is as follows:

[0024] Take (T imax -T imin ), T istd , T iave Perform mean normalization and multiply by 100. The new values are denoted as (T imax -T imin ) new , T istdnew , T iavenew And sort them from largest to smallest respectively. According to (T imax -T imin ) new , T istdnew , T iavenew Determine the training effect of each part based on the numerical size.

[0025] In a preferred embodiment, the method for evaluating the influence of a sports event on the training indicators of each part of the body and the method for calculating the recommended index of a sports event for each part of the body are as follows:

[0026] The data set is the training indicators of each part of the body under each sports event ((T imax -T imin ) new / T istdnew / T iavenew ). Each training indicator needs to be input into the model in batches, and this scheme needs to be input 3 times respectively;

[0027] Index selection: Suppose there are r trainings, n sports events, and m body parts. Then x θij is the training indicator value of the jth body part of the ith sports event under the θth training;

[0028] To remove the influence of dimensionality, standardize each indicator, x θ ' ij =x θij / x max ;

[0029] Determine the weight of the training indicator of the jth body part:

[0030] Calculate the entropy value of the training indicator of the jth body part:

[0031] where \(l_n\) is the natural logarithm, \(k\geq0\), and \(k = l_n(r_n)\);

[0032] Calculate the coefficient of difference \(g\) of the training index of the \(j\)th body part j : \(g\) j \(= 1 - e\) j ;

[0033] Calculate the weight of the training index of the \(j\)th body part:

[0034] Calculate the influence degree of each sports item on each body part:

[0035] The recommendation index \(\omega\) of the \(i\)th sports item for the \(j\)th body part ij The calculation formula of \(\omega\) is as follows ij \(= a\) 1 \(\alpha\) ij \(+ a\) 2 \(\beta\) ij \(+ a\) 3 \(\gamma\) ij ; \(a\) 1 \(+ a\) 2 \(+ a\) 3 \(= 1\);

[0036] where \(\alpha\) ij ,\(\beta\) ij and \(\gamma\) ij are the influence degrees of the \(i\)th sports item on the training intensity, training effect, and training status of the \(j\)th body part, and \(a_1\), \(a_2\), \(a_3\) are the weights of the training intensity, training effect, and training status.

[0037] In a preferred embodiment, the non-contact vital sign monitoring system based on thermal infrared imaging includes a body temperature calibration module, a thermal image processing module, a sports item recommendation module, and a data storage module, and the signals of each module are connected;

[0038] The body temperature calibration module is used to obtain the environmental temperature, the environmental temperature data measured by an uncalibrated thermal infrared imaging device, the body temperature data of the user, and the body temperature data measured by the thermal infrared imaging device multiple times, and perform temperature calibration;

[0039] The thermal image processing module is used to obtain the user's body surface data and perform data analysis;

[0040] The sports item recommendation module is used to evaluate the influence degree of sports items on the training indexes of each body part and calculate the recommendation index;

[0041] The data storage module is used to store all data during the processing of the platform.

[0042] Technical effects and advantages of the non-contact vital sign monitoring method and system based on thermal infrared imaging of the present invention:

[0043] Through a partition model based on human body proportions, the present invention divides the human body into 24 parts, combines the pixel distribution and geometric features of the infrared thermal image, accurately extracts the temperature data of each part, provides high-quality input for subsequent data analysis, and quantitatively analyzes the temperature difference, variance and mean through the entropy method to systematically evaluate the temperature changes of each part before and after exercise. This method can not only reflect the training intensity and effect of body parts, but also recommend more targeted exercise programs for users based on the sorting results; the solution constructs an evaluation model for exercise program recommendation through the correlation analysis between temperature data and training indicators, and can optimize the training program according to the influence degree of different exercises on each part. This solves the problem of lack of pertinence and scientific basis in traditional exercise recommendations. Brief Description of the Drawings

[0044] Figure 1 It is a flow chart of the non-contact vital sign monitoring method based on thermal infrared imaging of the present invention. Detailed Embodiments

[0045] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0046] The non-contact vital sign monitoring method based on thermal infrared imaging includes the following steps:

[0047] Install thermal infrared imaging devices at appropriate positions in front of and behind the user;

[0048] To reduce the influence of environmental temperature on the measurement of body surface temperature, calibrate the temperature of the thermal infrared imaging device.

[0049] Utilize the temperature distribution information in the thermal infrared image to extract the temperature data of the user's body surface through image processing technology.

[0050] Perform data analysis on the obtained temperature data to evaluate the training effect.

[0051] Evaluate the influence of exercise programs on the training indicators of each part of the body through the entropy method model, and the greater the influence of the exercise program on the training indicators, the greater the recommendation index. Thus, exercise program recommendation for users is realized.

[0052] Further, the method for installing the thermal infrared imaging device is as follows:

[0053] Select two thermal infrared imagers with high resolution, high frame rate, wide temperature measurement range, good stability and anti-interference ability, and place them at positions d in front of and behind the user respectively. Ensure that the thermal infrared imagers can fully cover the user's activity area, while avoiding direct sunlight, heat source interference and reflection effects.

[0054] The calculation formula for the distance d between a single thermal infrared imager and the user is as follows:

[0055] d = h / (2 * tan(θ / 2))

[0056] θ is the field of view angle of the thermal infrared imager. h is the height of the user. If h is taken as 2m, then the formula is simplified to

[0057] d = 1 / tan(θ / 2)

[0058] Furthermore, the steps for temperature calibration of the thermal infrared imaging device are as follows:

[0059] The initial temperature measured by the uncalibrated infrared thermometer is T m , the surface temperature of the part to be measured is T 测 , the ambient temperature is T temp , and the final calibration result is T 实 . It is confirmed by the literature that the influence of the packaging temperature of the infrared thermometer on the experiment can be ignored.

[0060] First, there is a difference kb between the T m measured by the infrared thermometer and the final calibration result T 实 . kb is the calibration value. That is:

[0061] T 实 = T m + kb

[0062] The difference b is obtained by summing and averaging the temperature values T i of the infrared thermometer measured multiple times, and then subtracting the surface temperature T 测 of the part to be measured from the average value. That is:

[0063]

[0064] The coefficient k before b can be calculated using the covariance formula for the overall errors T m , T temp and the packaging temperature T 封 of the infrared thermometer, and T m - T temp , T m - T 封 . Here, the packaging temperature is not considered, and let T 封 = 0, that is

[0065]

[0066] Finally, integrate the formulas to obtain the body temperature calibration model, that is

[0067]

[0068] The corresponding hardware program design is as follows:

[0069] Prepare the driver program for the infrared thermometer, the driver program for measuring the ambient temperature, and the driver, display, and key programs for the LCD display. First, activate the chip and peripheral driver programs of the experimental device and enter the waiting verification state; second, the interface displays the ambient temperature in real time. When the temperature of the infrared thermometer is measured, record the measurement data in an array; then, use the timer to time, measure once per second. Each time a measurement is taken, display the measured temperature value on the LCD screen. A total of 10 measurements are taken. If more than 10 measurements are taken, the real-time measured temperature value will be displayed in a certain area, but no record will be made; finally, refresh the interface through the key and enter the next round of calibration temperature measurement. The program flow Figure 1 As shown in the figure.

[0070] Furthermore, the method for extracting the temperature data of the user's body surface is as follows:

[0071] Divide the user's body into 24 parts, 10 in the front (left chest, right chest, left abdomen, right abdomen, front left upper arm, front left lower arm, front right upper arm, front right lower arm, front left thigh, front right thigh), and 12 in the back (left back, right back, back left upper arm, back left lower arm, back right upper arm, back right lower arm, left buttock, right buttock, back left thigh, back right thigh, left calf, right calf,)

[0072] According to the human body ratio, simplify the standing person image into a large rectangle. Let the height of the person be h and the shoulder width be j. Then the length of the rectangle is h and the width is j. Establish a rectangular coordinate system with the lower left corner of the rectangle as the origin (establish it once in the front and once in the back). Each body part is also regarded as a small rectangle.

[0073] The coordinates of each body part can be obtained, where a = h / 9 and b = j / 3

[0074] Front body:

[0075] Front left upper arm: (0, 7.5a), (0.5b, 7.5a), (0, 6a), (0.5b, 6a)

[0076] Left chest: (0.5b, 7.5a), (1.5b, 7.5a), (0.5b, 6a), (1.5b, 6a)

[0077] Right chest: (1.5b, 7.5a), (2.5b, 7.5a), (1.5b, 6a), (2.5b, 6a)

[0078] Anterior right upper arm: (2.5b, 7.5a), (3b, 7.5a), (2.5b, 6a), (3b, 6a)

[0079] Anterior left lower arm: (0, 6a), (0.5b, 6a), (0, 4.5a), (0.5b, 4.5a)

[0080] Left abdomen: (0.5b, 6a), (1.5b, 6a), (0.5b, 4.5a), (1.5b, 4.5a)

[0081] Right abdomen: (1.5b, 6a), (2.5b, 6a), (1.5b, 4.5a), (2.5b, 4.5a)

[0082] Anterior right lower arm: (2.5b, 6a), (3b, 6a), (2.5b, 4.5a), (3b, 4.5a)

[0083] Anterior left thigh: (0.5b, 4.5a), (1.5b, 4.5a), (0.5b, 2.5a), (1.5b, 2.5a)

[0084] Anterior right thigh: (1.5b, 4.5a), (2.5b, 4.5a), (1.5b, 2.5a), (2.5b, 2.5a)

[0085] Back:

[0086] Posterior left upper arm: (0, 7.5a), (0.5b, 7.5a), (0, 6a), (0.5b, 6a)

[0087] Left back: (0.5b, 7.5a), (1.5b, 7.5a), (0.5b, 6a), (1.5b, 6a)

[0088] Right back: (1.5b, 7.5a), (2.5b, 7.5a), (1.5b, 6a), (2.5b, 6a)

[0089] Posterior right upper arm: (2.5b, 7.5a), (3b, 7.5a), (2.5b, 6a), (3b, 6a)

[0090] Posterior left lower arm: (0, 6a), (0.5b, 6a), (0, 4.5a), (0.5b, 4.5a)

[0091] Left hip: (0.5b, 6a), (1.5b, 6a), (0.5b, 4.5a), (1.5b, 4.5a)

[0092] Right hip: (1.5b, 6a), (2.5b, 6a), (1.5b, 4.5a), (2.5b, 4.5a)

[0093] Rear right forearm: (2.5b, 6a), (3b, 6a), (2.5b, 4.5a), (3b, 4.5a)

[0094] Rear left thigh: (0.5b, 4.5a), (1.5b, 4.5a), (0.5b, 2.5a), (1.5b, 2.5a)

[0095] Rear right thigh: (1.5b, 4.5a), (2.5b, 4.5a), (1.5b, 2.5a), (2.5b, 2.5a)

[0096] Rear left calf: (0.5b, 2.5a), (1.5b, 2.5a), (0.5b, 0), (1.5b, 0)

[0097] Rear right calf: (1.5b, 2.5a), (2.5b, 2.5a), (1.5b, 0), (2.5b, 0)

[0098] Let the minimum temperature difference of the i-th body part be T imin , the maximum temperature difference be T imax , the variance of the temperature difference be T istd , and the mean value of the temperature difference be T iave . The temperature difference is the difference in the temperatures measured at the same position before and after exercise.

[0099] Furthermore, the method for analyzing the temperature data is as follows:

[0100] Normalize the mean of (T imax - T imin ), T istd , T iave and multiply by 100. The new values are denoted as (T imax - T imin ) new , T istdnew , T iavenew and sort them from largest to smallest respectively. Mean normalization is a common method and will not be elaborated here.

[0101] (T imax - T imin ) new The larger it is, the greater the temperature difference of the corresponding body part i before and after exercise, the faster the temperature rise of this body part, and the higher the training intensity;

[0102] T istdnew The smaller it is, the smaller the variance of the temperature difference of the corresponding body part i before and after exercise, the more uniform the temperature rise distribution of this body part, and the more stable the training effect.

[0103] T iavenew The larger it is, the larger the mean value of the temperature difference of the corresponding body part i before and after exercise, the better the overall temperature rise effect of this body part, and the better the training state.

[0104] Furthermore, the method for evaluating the influence of sports events on the training indicators of each body part and the calculation method of the recommended index of sports events on each body part are as follows:

[0105] For the research on the sports effect evaluation method, the influence degree of sports events on the training indicators of each body part is calculated based on the entropy value method in the objective weighting method:

[0106] Information entropy is a measure of the degree of disorder of a system, and information is a measure of the degree of order of a system. The absolute values of the two are equal but the signs are opposite. The greater the variation degree of the index value of a certain index, the smaller the information entropy, the greater the amount of information provided by this index, and the greater the weight of this index; conversely, the smaller the variation degree of the index value of a certain index, the greater the information entropy, the smaller the amount of information provided by this index, and the smaller the weight of this index. Therefore, the weights of each index can be calculated using the information entropy tool according to the variation degree of each index value.

[0107] The data set is the training indicators of each body part under each sports event ((T imax -T imin ) new / T istdnew / T iavenew ), and each training indicator needs to be input into the model in batches. In this solution, it needs to be input 3 times respectively.

[0108] 1. Index selection: Suppose there are r trainings, n sports events, and m body parts. In this solution, m is selected as 24. Then x θij is the training index value of the jth body part of the sports event i under the θth training.

[0109] 2. To remove the influence of dimensionality, standardize each index.

[0110] x′ θij =x θij / x max

[0111] 3. Determine the weight of the training index of the jth body part:

[0112]

[0113] 4. Calculate the entropy value of the training index of the j-th body part:

[0114]

[0115] In the formula, ln is the natural logarithm, k≥0, k = ln(rn).

[0116] 5. Calculate the coefficient of variation g of the training index of the j-th body part j :

[0117] g j =1 - e j

[0118] 6. Calculate the weight of the training index of the j-th body part:

[0119]

[0120] 7. Calculate the influence degree of each sports item on each body part:

[0121]

[0122] The calculation formula of the recommendation index ω of the i-th sports item for the j-th body part ij is as follows

[0123] ω ij =a 1 α ij + a 2 β ij + a 3 γ ij

[0124] a1 + a2 + a3 = 1

[0125] Among them, α ij , β ij and γ ij are the influence degrees of the i-th sports item on the training intensity, training effect, and training state of the j-th body part, and a1, a2, a3 are the weights of the training intensity, training effect, and training state. Users can set them by themselves or choose the default weights of the system.

[0126] The following introduces the calculation methods of the default weights of the training intensity, training effect, and training state of the system:

[0127] Judging from common sense, users with different body fat percentages will have differences in training intensity, training effect, and training state. Anaerobic exercise and aerobic exercise will also have differences in training intensity, training effect, and training state.

[0128] 1. Users with a BMI value ∈ [18.5, 23.9] are classified as normal users, denoted as nor, users with a BMI < 18.5 are classified as underweight users, denoted as sho, and users with a BMI > 23.9 are classified as overweight users, denoted as lar. The BMI value is calculated by dividing the weight (in kilograms) by the square of the height (in meters). Exercise items are divided into two major categories: aerobic exercise (denoted as ae) and anaerobic exercise (denoted as an).

[0129] 2. Collect the training data of users

[0130] (1) Calculate the average value of all training intensity indicators of normal users, denoted as nor1; the average value of all training status indicators of normal users, denoted as nor2; the average value of all training effect indicators of normal users, denoted as nor3; and so on to calculate sho1, sho2, sho3, lar1, lar2, lar3.

[0131] (2) Calculate the average value of all training intensity indicators under aerobic exercise, denoted as ae1; the average value of all training status indicators under aerobic exercise, denoted as ae2; the average value of all training effect indicators under aerobic exercise, denoted as ae3; and so on to calculate an1, an2, an3.

[0132] (3) For ease of understanding, record it in the following table:

[0133] Training intensity Training effect Training status Normal user <![CDATA[nor1]]> <![CDATA[nor2]]> <![CDATA[nor3]]> Underweight user <![CDATA[sho1]]> <![CDATA[sho2]]> <![CDATA[sho3]]> Overweight user <![CDATA[lar1]]> <![CDATA[lar2]]> <![CDATA[lar3]]> Aerobic exercise <![CDATA[ae1]]> <![CDATA[ae2]]> <![CDATA[ae3]]> Anaerobic exercise <![CDATA[an1]]> <![CDATA[an2]]> <![CDATA[an3]]>

[0134] 3. Calculate the weights of training intensity, training effect, and training status of normal users under aerobic exercise as follows:

[0135] a1 = (nor1 + ae1) / (nor1 + ae1 + nor2 + ae2 + nor3 + ae3)

[0136] a2 = (nor2 + ae2) / (nor1 + ae1 + nor2 + ae2 + nor3 + ae3)

[0137] a3 = (nor3 + ae3) / (nor1 + ae1 + nor2 + ae2 + nor3 + ae3)

[0138] The weights of training intensity, training effect, and training status of normal users under anaerobic exercise are as follows:

[0139] a1 = (nor1 + an1) / (nor1 + an1 + nor2 + an2 + nor3 + an3)

[0140] a2 = (nor2 + an2) / (nor1 + an1 + nor2 + an2 + nor3 + an3)

[0141] a3 = (nor3 + an3) / (nor1 + an1 + nor2 + an2 + nor3 + an3)

[0142] The weights of training intensity, training effect, and training status for underweight users during aerobic exercise are as follows:

[0143] a1 = (sho1 + ae1) / (sho1 + ae1 + sho2 + ae2 + sho3 + ae3)

[0144] a2 = (sho2 + ae2) / (sho1 + ae1 + sho2 + ae2 + sho3 + ae3)

[0145] a3 = (sho3 + ae3) / (sho1 + ae1 + sho2 + ae2 + sho3 + ae3)

[0146] The weights of training intensity, training effect, and training status for underweight users during anaerobic exercise are as follows:

[0147] a1 = (sho1 + an1) / (sho1 + an1 + sho2 + an2 + nor3 + sho3)

[0148] a2 = (sho2 + an2) / (sho1 + an1 + sho2 + an2 + nor3 + sho3)

[0149] a3 = (sho3 + an3) / (sho1 + an1 + sho2 + an2 + nor3 + sho3)

[0150] The weights of training intensity, training effect, and training status for overweight users during aerobic exercise are as follows:

[0151] a1 = (l ar1 + ae1) / (l ar1 + ae1 + l ar2 + ae2 + l ar3 + ae3)

[0152] a2 = (l ar2 + ae2) / (l ar1 + ae1 + l ar2 + ae2 + l ar3 + ae3)

[0153] a3 = (l ar3 + ae3) / (l ar1 + ae1 + l ar2 + ae2 + l ar3 + ae3)

[0154] The weights of training intensity, training effect, and training status for overweight users during anaerobic exercise are as follows:

[0155] a1 = (l ar1 + an1) / (l ar1 + an1 + l ar2 + an2 + l ar3 + an3)

[0156] a2 = (lar2 + an2) / (lar1 + an1 + lar2 + an2 + lar3 + an3)

[0157] a3 = (lar3 + an3) / (lar1 + an1 + lar2 + an2 + lar3 + an3)

[0158] A non-contact vital sign monitoring method and system based on thermal infrared imaging, including a body temperature calibration module, a thermal image processing module, a sports item recommendation module, and a data storage module, with signal connections between the modules;

[0159] The body temperature calibration module is used to obtain the ambient temperature, the ambient temperature data measured by an uncalibrated thermal infrared imaging device, the body temperature data of the user, and the body temperature data measured by the thermal infrared imaging device multiple times, and perform temperature calibration.

[0160] The thermal image processing module is used to obtain the user's body surface data and perform data analysis.

[0161] The sports item recommendation module is used to evaluate the influence degree of sports items on the training indexes of various parts of the body and calculate the recommendation index.

[0162] The data storage module is used to store all data during the platform processing. The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the real situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0163] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product.

[0164] Those of ordinary skill in the art can realize that the modules and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application of the technical solution and the invention constraints. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0165] In addition, in each embodiment of the present application, the functional modules can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0166] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, and all should be covered by the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims described above.

[0167] Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A non-contact vital sign monitoring method based on thermal infrared imaging, characterized in that, It includes the following steps: Perform temperature calibration on the thermal infrared imaging device; Extract the temperature data of the user's body surface by using the temperature distribution information in the thermal infrared image through image processing technology; Construct a body temperature calibration model with the obtained temperature data to calibrate the temperature data, and calculate the influence degree of sports items on the training indexes of various body parts based on the entropy value method; Recommend the sports items with a great influence on the training indexes.

2. The non-contact vital sign monitoring method based on thermal infrared imaging according to claim 1, characterized in that Construct a body temperature calibration model with the obtained temperature data to calibrate the temperature data, which specifically includes the following steps: The initial temperature measured using an uncalibrated infrared thermometer is T m , the surface temperature of the part to be measured is T 测 , the ambient temperature is T temp , and the final calibration result is T 实 ; First is the T measured using an infrared thermometer m and the final calibration result is T 实 There is a difference kb between them, and kb is the calibration value; that is: T 实 = T m + kb; The difference value b is obtained by summing up and averaging the temperature values T of the infrared thermometer measured multiple times, and then subtracting the surface temperature T of the part to be measured from the average value; that is: i Sum and calculate the average value, and then subtract the surface temperature T of the part to be measured 测 from the average value; namely: The coefficient k before b is calculated using the covariance formula for T m , T temp and the overall error T of the packaging temperature T of the infrared thermometer 封 is obtained. Let T m -T temp , T m -T 封 be obtained. Let T 封 = 0, that is Finally, integrate the formula to obtain the body temperature calibration model, that is 3. The non-contact vital sign monitoring method based on thermal infrared imaging according to claim 2, wherein: The method for extracting the temperature data of the user's body surface is as follows: Divide the user's body into multiple parts. According to the human body proportion, simplify the standing person image into a rectangle. Let the height of the person be h and the shoulder width be j, and establish a rectangular coordinate system with the lower left corner of the rectangle as the origin; Find the coordinates of each body part, where a = h / 9 and b = j / 3 Let the minimum temperature difference of the \(i\)-th body part be \(T\). imin The maximum temperature difference is \(T\). imax The variance of the temperature difference is \(T\). istd The mean value of the temperature difference is \(T\). iave The temperature difference is the difference in temperatures measured at the same position before and after exercise. The method for analyzing the temperature data is as follows: The (T imax -T imin ), T istd , T iave are mean-normalized and all multiplied by 100. The new values are denoted as (T imax -T imin ). new , T istdnew , T iavenew And they are sorted from largest to smallest respectively. According to the (T imax -T imin ). new , T istdnew , T iavenew The training effects of each part are determined by the numerical size.

4. The non-contact vital sign monitoring method based on thermal infrared imaging according to claim 3, wherein ； The method for evaluating the influence of sports items on the training indexes of various body parts and the calculation method of the recommendation index of sports items for various body parts are as follows: The dataset is the training metrics of each body part under each sports event ((T imax -T imin ) new / T istdnew / T iavenew ). Each training metric needs to be input into the model in batches, and this scheme needs to be input 3 times respectively; Index selection: Suppose there are r training sessions, n sports events, and m body parts. Then x θij is the training index value of the jth body part of the ith sports event in the θth training session; To remove the influence of dimensionlessization, standardize each index, \(x\) θ ' ij = \(x\) θij / \(x\) max ; Determine the weight of the training index for the j-th body part: Calculate the entropy value of the training index of the j-th body part: In the formula, ln is the natural logarithm, k ≥ 0, and k = ln(rn); Calculate the difference coefficient g of the training index of the jth body part j : g j = 1 - e j ; Calculate the weight of the training index for the j-th body part: Calculate the degree of influence of each sports event on each part of the body: The recommended index ω of the i-th sports event for the j-th body part ij is calculated as follows: ω ij = a 1 α ij + a 2 β ij + a 3 γ ij ; a 1 + a 2 + a 3 = 1; Among them, α ij , β ij and γ ij are the degrees of influence of the i-th sports event on the training intensity, training effect, and training status of the j-th body part, and a1, a2, and a3 are the weights of the training intensity, training effect, and training status.

5. A non-contact vital sign monitoring system based on thermal infrared imaging, which is used to implement the non-contact vital sign monitoring method based on thermal infrared imaging according to any one of claims 1-4, wherein: It includes a body temperature calibration module, a thermal image processing module, a sports item recommendation module, and a data storage module, and the modules are connected by signals; The body temperature calibration module is used to obtain the environmental temperature, the environmental temperature data measured by using an uncalibrated thermal infrared imaging device, the body temperature data of the user, and the body temperature data measured by the thermal infrared imaging device multiple times, and perform temperature calibration; The thermal image processing module is used to obtain the user's body surface data and perform data analysis; The sports item recommendation module is used to evaluate the influence degree of sports items on the training indexes of various body parts and calculate the recommendation index; The data storage module is used to store all data during the processing of the platform.

Images