A method and device for predicting exercise energy consumption

Abstract

The present invention provides a method and device for predicting exercise energy consumption, comprising: obtaining acceleration data and a target exercise intensity level of a target subject during exercise; performing feature extraction on the acceleration data and obtaining characteristic information of the acceleration data using principal component analysis; constructing input parameters for the target subject based on the characteristic information and the target subject's physical condition information; and inputting the input parameters into a target exercise energy consumption prediction model based on the target exercise intensity level to obtain the target subject's exercise energy consumption. The present invention selects different target exercise energy consumption prediction models based on different target exercise intensity levels. The target exercise energy consumption prediction models not only utilize the characteristic information of the acceleration data but also utilize the target subject's physical condition information to predict the target exercise energy consumption, and use the predicted results as the target subject's exercise energy consumption, thereby improving the accuracy of exercise energy consumption prediction.

Description

Technical Field

[0001] The present invention relates to the technical field of sports energy consumption measurement, and in particular to a sports energy consumption prediction method and device. Background Art

[0002] The widespread use of wearable devices has made it possible to exercise on treadmills, rowing machines, and bicycles, and has facilitated the measurement of energy consumption during human activity through motion sensors such as mobile phones or sports watches. The use of inertial units to collect motion data from various parts of the human body and accurately measure energy consumption has broadened its application, making it increasingly convenient for exercisers and rehabilitation personnel to use portable energy consumption monitoring devices for health monitoring in their daily lives. Therefore, research on energy consumption prediction for specific scenarios is of great significance.

[0003] Currently, two algorithms, pedometer and acceleration, are used to monitor exercise energy consumption, or calories consumed. The pedometer energy consumption algorithm studies the correlation between human movement speed and human energy consumption, and derives an equation for calculating walking energy consumption based on a pedometer. However, the type and intensity of exercise significantly impact the pedometer's distance and energy calculations, and the pedometer cannot achieve good measurement accuracy under complex movements. The acceleration energy consumption algorithm, based on collected acceleration data, studies the correlation between the geometric mean of the three-axis accelerometer and human energy consumption, and proposes an equation for calculating human energy consumption based on the ensemble average of acceleration. However, the constructed linear regression equation fails to utilize the richer characteristics of acceleration, and therefore cannot accurately determine human exercise energy consumption.

[0004] It can be seen from this that it is difficult to obtain accurate measurement results of exercise energy consumption using existing exercise energy consumption measurement methods. Summary of the Invention

[0005] The present invention provides a method, device, electronic device and storage medium for predicting exercise energy consumption, which are used to solve the defect in the prior art that it is difficult to accurately measure exercise energy consumption.

[0006] In a first aspect, the present invention provides a method for predicting exercise energy consumption, comprising: obtaining acceleration data and a target exercise intensity level of a target object during exercise; performing feature extraction on the acceleration data, and obtaining feature information of the acceleration data using a principal component analysis method; constructing input parameters of the target object based on the feature information and the physical state information of the target object; inputting the input parameters into a target exercise energy consumption prediction model based on the target exercise intensity level to obtain the exercise energy consumption of the target object; wherein the target exercise energy consumption prediction model is obtained based on sample input parameters at the target exercise intensity level and sample exercise energy consumption training corresponding to the sample input parameters.

[0007] According to a method for predicting exercise energy consumption provided by the present invention, a target exercise intensity level of a target object during exercise is obtained, including: obtaining the exercise tolerance of the target object during exercise; and determining the target exercise intensity level according to the value range of the exercise tolerance.

[0008] According to a method for predicting exercise energy consumption provided by the present invention, the input parameters of the target object are constructed based on the characteristic information and the physical condition information of the target object, including: standardizing the characteristic information and the physical condition information so that the average value of the input parameter is a preset value and the variance is less than a preset variance threshold.

[0009] According to a method for predicting exercise energy consumption provided by the present invention, when the exercise intensity level is divided into mild, moderate and severe, the input parameters are input into a target exercise energy consumption prediction model according to the target exercise intensity level to obtain the exercise energy consumption of the target object, including: when the target exercise intensity level is mild, the input parameters are input into the mild exercise energy consumption prediction model to obtain the first exercise energy consumption of the target object; when the target exercise intensity level is moderate, the input parameters are input into the moderate exercise energy consumption prediction model to obtain the second exercise energy consumption of the target object; when the target exercise intensity level is severe, the input parameters are input into the severe exercise energy consumption prediction model to obtain the third exercise energy consumption of the target object.

[0010] According to a method for predicting exercise energy consumption provided by the present invention, the target exercise energy consumption prediction model is a neural network model; the transfer function of the input layer and the hidden layer of the neural network model is a hyperbolic tangent curve function.

[0011] According to a method for predicting exercise energy consumption provided by the present invention, the characteristic information includes at least one of the following characteristics: upper quartile, median, lower quartile, maximum value, mean, minimum value, kurtosis, skewness, variance and standard deviation; the physical state information includes at least one of the following physical state parameters: the age of the target object, the height of the target object and the weight of the target object.

[0012] According to a motion energy consumption prediction method provided by the present invention, acceleration data of a target object during motion is obtained, including: obtaining the acceleration data using an acceleration sensor set at a target human body position of the target object.

[0013] In a second aspect, the present invention further provides an exercise energy consumption prediction device, comprising: a first module, a second module, a third module and a fourth module.

[0014] The first module is used to obtain the acceleration data and target motion intensity level of the target object during the motion process;

[0015] The second module is used to extract features from the acceleration data and obtain feature information of the acceleration data using principal component analysis;

[0016] A third module is configured to construct input parameters of the target object based on the feature information and the physical condition information of the target object;

[0017] A fourth module is configured to input the input parameters into a target exercise energy consumption prediction model according to the target exercise intensity level to obtain the exercise energy consumption of the target object;

[0018] The target motion energy consumption prediction model is obtained through training based on sample input parameters at the target motion intensity level and sample motion energy consumption corresponding to the sample input parameters.

[0019] In a third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of any of the above-described methods for predicting exercise energy consumption are implemented.

[0020] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above-described methods for predicting exercise energy consumption.

[0021] The exercise energy consumption prediction method and device provided by the present invention select different target exercise energy consumption prediction models according to different target exercise intensity levels. The target exercise energy consumption prediction model not only uses the characteristic information of acceleration data but also uses the physical state information of the target object to predict the target exercise energy consumption, and uses the prediction result as the exercise energy consumption of the target object, thereby improving the prediction accuracy of exercise energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0023] Figure 1 This is one of the flow charts of the exercise energy consumption prediction method provided by the present invention;

[0024] Figure 2This is the second flow chart of the exercise energy consumption prediction method provided by the present invention;

[0025] Figure 3 This is the third flow chart of the exercise energy consumption prediction method provided by the present invention;

[0026] Figure 4 It is a structural diagram of the exercise energy consumption prediction device provided by the present invention;

[0027] Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0028] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0029] It should be noted that, in the description of the embodiments of the present invention, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "include a ..." do not exclude the presence of other identical elements in the process, method, article or device comprising the elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances.

[0030] The terms "first," "second," and the like in this application are used to distinguish similar objects, and are not used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, so that the embodiments of this application can be implemented in orders other than those illustrated or described herein. Furthermore, the objects distinguished by "first," "second," and the like generally refer to a class of objects and do not limit the number of objects. For example, the first object may be one or more.

[0031] The following combination Figure 1-Figure 5 The present invention describes an exercise energy consumption prediction method and device provided by an embodiment of the present invention.

[0032] Figure 1 This is one of the flow charts of the method for predicting exercise energy consumption provided by the present invention, such as Figure 1 As shown, including but not limited to the following steps:

[0033] Step 101: Acquire acceleration data and target motion intensity level of a target object during motion.

[0034] The present invention can obtain the acceleration data by using an acceleration sensor arranged at a target human body position of the target object. The target object can be an adult.

[0035] It is understandable that the target human body position may be a leg, arm, back, etc. of the target object.

[0036] The acceleration sensor may be an inertial unit sensor MPU6050 GY-521; the acceleration sensor may be placed on the user's calf to acquire acceleration.

[0037] In one scenario, after the acceleration sensor is set up, the target subject can exercise on a treadmill with an exercise intensity level measurement function. During the exercise, the acceleration data and the target exercise intensity level of the target subject are measured.

[0038] The exercise intensity level may include multiple levels. Optionally, the exercise intensity level includes: light, moderate and heavy.

[0039] Step 102: extracting features from the acceleration data and obtaining feature information of the acceleration data using principal component analysis.

[0040] Optionally, before performing feature extraction on the acceleration data, the acquired acceleration data may be subjected to noise reduction processing. The noise reduction method may be a wavelet noise reduction method. Feature extraction on the noise-reduced acceleration data can more accurately obtain feature information of the acceleration data.

[0041] It should be noted that the feature information includes at least one of the following features: upper quartile, median, lower quartile, maximum value, mean, minimum value, kurtosis, skewness, variance and standard deviation.

[0042] Step 103: Construct input parameters of the target object based on the feature information and the physical condition information of the target object.

[0043] It should be noted that the target object's physical condition information includes at least one of the following physical condition parameters: the target object's age, the target object's height, and the target object's weight. Furthermore, the physical condition information is not limited to age, height, and weight.

[0044] The present invention can perform standardization processing on the characteristic information and the physical state information so that the average value of the input parameter is a preset value and the variance is less than a preset variance threshold.

[0045] Optionally, the preset value of the average value is 0; the preset method threshold is 2.

[0046] The input parameters include characteristic information of acceleration data and physical condition information of the target object. Compared with the existing technology that only uses acceleration data to measure exercise energy consumption, it utilizes richer features and can improve the measurement accuracy of exercise energy consumption based on data.

[0047] Step 104: According to the target exercise intensity level, the input parameters are input into a target exercise energy consumption prediction model to obtain the exercise energy consumption of the target object.

[0048] The present invention constructs different exercise energy consumption prediction models based on different target exercise intensity levels. After obtaining input parameters, the target exercise intensity level corresponding to the input parameters can be used to determine the target exercise energy consumption prediction model.

[0049] The input parameters are then fed into a pre-trained target exercise energy consumption prediction model to obtain the target object's exercise energy consumption, which can be represented by a specific calorie value.

[0050] It can be understood that the target motion energy consumption prediction model is obtained by training based on sample input parameters at the target motion intensity level and sample motion energy consumption corresponding to the sample input parameters.

[0051] It is understandable that the target motion energy consumption prediction model in the present invention can be a regression model such as a support vector machine model, a neural network model, etc.

[0052] The exercise energy consumption prediction method provided by the present invention selects different target exercise energy consumption prediction models according to different target exercise intensity levels. The target exercise energy consumption prediction model not only uses the characteristic information of acceleration data but also uses the physical state information of the target object to predict the target exercise energy consumption, and uses the prediction result as the exercise energy consumption of the target object, thereby improving the prediction accuracy of exercise energy consumption.

[0053] Based on the contents in the above embodiments, as an optional embodiment, the present invention provides an exercise energy consumption prediction method for obtaining a target exercise intensity level of a target object during exercise, including: obtaining the exercise tolerance of the target object during exercise; and determining the target exercise intensity level according to the value range of the exercise tolerance.

[0054] In the present invention, a dedicated oxygen mask can be used to measure the oxygen consumption of the target subject and calculate the specific value of exercise tolerance. It is understandable that the measurement of exercise tolerance can be achieved by adopting existing technology, and the method of obtaining exercise tolerance is not further described here.

[0055] Optionally, when the exercise tolerance is less than 3 METs, the target exercise intensity level is determined to be mild; when the exercise tolerance is greater than or equal to 3 METs and less than or equal to 6 METs, the target exercise intensity level is determined to be moderate; when the exercise tolerance is greater than 6 METs, the target exercise intensity level is determined to be severe.

[0056] Based on the content in the above embodiment, as an optional embodiment, the exercise energy consumption prediction method provided by the present invention is:

[0057] Figure 2 This is the second flow chart of the method for predicting exercise energy consumption provided by the present invention. Figure 2 As shown, the present invention inputs the input parameters into the target exercise energy consumption prediction model according to the target exercise intensity level to obtain the exercise energy consumption of the target object.

[0058] When the target exercise intensity level is light, the input parameters are input into a light exercise energy consumption prediction model to obtain the first exercise energy consumption of the target object; when the target exercise intensity level is moderate, the input parameters are input into a moderate exercise energy consumption prediction model to obtain the second exercise energy consumption of the target object; when the target exercise intensity level is heavy, the input parameters are input into a heavy exercise energy consumption prediction model to obtain the third exercise energy consumption of the target object.

[0059] In order to more clearly illustrate the technical solution of the present invention, the implementation process of the present invention is described below in conjunction with a complete experiment.

[0060] Figure 3 This is the third flow chart of the method for predicting exercise energy consumption provided by the present invention. Figure 3 The experimental process of the present invention is described.

[0061] The experiment in the present invention detects body motion signals through the inertial unit sensor MPU6050 GY-521 and obtains acceleration data, which is three-axis acceleration data. After filtering and voltage division processing, the acceleration data enters the microprocessor STM32F103ZET6, and the analog signal is converted into a digital signal through the built-in AD converter. After further processing, it is sent to the static memory IS62WV51216 for storage. The data is sent to the PC through the serial port for data processing and algorithm analysis, and then the different characteristics of the human body's motion state are obtained, providing accurate data support for the subsequent establishment of the exercise energy consumption prediction model. To facilitate data collection, the inertial unit sensor in the present invention adopts two power supply modes: button battery and DC power supply. Button battery can be used when collecting data for offline use.

[0062] To balance stability and reliability, a sampling frequency of 50Hz and 2000 sampling points per axis were considered. Since the acceleration generated during human motion is not very large, and a high range inevitably reduces accuracy, the accelerometer range was considered to be ±2g. Since the accelerometer's ADC is 16-bit, the sensitivity is: 65536 / 4 = 16384LSB / g. Since the STM32F103ZET6 uses an internal 8M RC crystal by default, which may miss the motion characteristics generated at specific locations during human motion, a PLL clock source using the X / Y / Z axis gyroscope as a reference was selected. Here, the X-axis gyroscope was selected as the reference. The system reference voltage is 3.3V. Based on this configuration, the acceleration generated during human motion is collected to provide data support for developing a motion energy consumption prediction model for specific scenarios.

[0063] The experiment selected 30 testers aged 19-45 as research subjects and collected acceleration data of human movement in a treadmill scenario.

[0064] After acceleration data processing is complete, supervised learning is currently being considered for categorizing exercise intensity levels based on pre-set exercise intensity grading criteria. Exercise intensity levels are categorized based on the exercise tolerance value output by the oxygen mask. This invention defines mild intensity as <3 METs, moderate intensity as 3-6 METs, and severe intensity as >6 METs.

[0065] When the target motion energy consumption prediction model is a neural network model, the following describes the process of building an artificial neural network model based on intensity classification.

[0066] (1) Data preprocessing and feature extraction

[0067] Alternatively, the present invention uses existing measurement equipment to obtain human calorie consumption every 10 seconds and uses this as exercise energy consumption. Therefore, feature extraction is performed on acceleration data at every 50 sampling frequencies within each 10-second period. Principal component analysis is used to understand the interrelationships between features, and the feature information ultimately extracted is the upper quartile, median, lower quartile, maximum, mean, minimum, kurtosis, skewness, variance, and standard deviation.

[0068] (2) Input of input parameters.

[0069] The test subject's age, height, weight, and other information are selected and added to the extracted feature information to construct the input parameters. The predicted output result label is the exercise energy consumption within the 10 seconds.

[0070] (3) Classification of exercise intensity.

[0071] Supervised learning is performed based on the pre-set exercise intensity level classification criteria. Classification is performed based on the obtained exercise tolerance METs value.

[0072] Mild is defined as <3 METs, while moderate intensity is 3-6 METs, and severe is >6 METs.

[0073] (4) Construction of artificial neural network model (ANN) at each exercise intensity level.

[0074] The importance of each feature in the input parameters is expressed by a weight. A single neuron does not have the ability to solve difficult problems because it cannot handle all the required nonlinear factors or interactions, so the neurons are arranged into several computational layers, which are connected by transfer functions.

[0075] (5) Target motion energy consumption prediction model training.

[0076] Before training begins, each network layer's weights are assigned a small random value. To improve estimation errors, the ANN model's learning process uses a feedforward / backward propagation algorithm (BP neural network algorithm). The feedforward process involves calculating the transformation between input and output, calculating the data for each window based on the current weights and transfer function. The feedforward estimate is compared with the observed value as it passes through the network, and an error is calculated.

[0077] The gradient (rate of change) of this error is used to iteratively adjust the weights. This weight update process constitutes backpropagation (BP), an optimization process that follows most standard optimization methods. By repeatedly feeding the same data into the network model, the weights adapt, the error signal decreases, and an optimal set of weights is achieved. This training continues until an acceptable error is achieved. Once the optimized weights are achieved, the model has calculated the energy consumption expressed by the acceleration signal.

[0078] The ANN model uses a three-layer network structure with 10 nodes in the input layer, 20 nodes in the hidden layer, and 1 node in the output layer. The hyperbolic tangent function was chosen as the transfer function for the input and hidden layers because the data in this experiment was normalized, with a mean of 0 and a variance of 1, to reduce significant deviations in the input parameters before training. Data training used gradient descent with a learning rate of 0.0001.

[0079] (6) Target motion energy consumption prediction model output.

[0080] The final output of the ANN model is the energy consumption of exercise over a period of time, that is, the calorie value.

[0081] Compared to other methods for calculating exercise energy expenditure, this method establishes separate energy expenditure prediction models for each intensity level based on exercise intensity classification, integrating energy consumption with intensity information and resulting in more accurate calculated energy expenditure results. Furthermore, an ANN algorithm is designed for each intensity classification, using a feedforward / backward propagation algorithm. Feedforward estimates are compared with observed values ​​through the network to calculate an error. This error is used to iteratively adjust weights. The weight update process constitutes backpropagation (BP), achieving weight adaptation and error reduction, resulting in a set of optimal weights and a more accurate exercise energy expenditure calculation result.

[0082] Figure 4 Schematic diagram of the structure of the exercise energy consumption prediction device provided by the present invention. Figure 4 As shown, the device includes: a first module 401, a second module 402, a third module 403 and a fourth module 404.

[0083] The first module 401 is used to obtain acceleration data and target motion intensity level of the target object during motion;

[0084] The second module 402 is used to extract features from the acceleration data and obtain feature information of the acceleration data using principal component analysis;

[0085] The third module 403 is configured to construct input parameters of the target object based on the feature information and the physical condition information of the target object;

[0086] The fourth module 404 is configured to input the input parameters into a target exercise energy consumption prediction model according to the target exercise intensity level to obtain the exercise energy consumption of the target object;

[0087] The exercise energy consumption prediction device provided by the present invention selects different target exercise energy consumption prediction models according to different target exercise intensity levels. The target exercise energy consumption prediction model not only uses the characteristic information of acceleration data but also uses the physical state information of the target object to predict the target exercise energy consumption, and uses the prediction result as the exercise energy consumption of the target object, thereby improving the prediction accuracy of exercise energy consumption.

[0088] It should be noted that the exercise energy consumption prediction device provided in the embodiment of the present invention can execute the exercise energy consumption prediction method described in any of the above embodiments during specific operation, which will not be described in detail in this embodiment.

[0089] Figure 5 Schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 5As shown, the electronic device may include: a processor 510, a communication interface 520, a memory 530, and a communication bus 540, wherein the processor 510, the communication interface 520, and the memory 530 communicate with each other via the communication bus 540. The processor 510 may call the logic instructions in the memory 530 to execute a method for predicting exercise energy consumption, which includes: obtaining acceleration data and a target exercise intensity level of a target object during exercise; performing feature extraction on the acceleration data and obtaining feature information of the acceleration data using a principal component analysis method; constructing input parameters of the target object based on the feature information and the physical state information of the target object; inputting the input parameters into a target exercise energy consumption prediction model based on the target exercise intensity level to obtain the exercise energy consumption of the target object; wherein the target exercise energy consumption prediction model is obtained based on sample input parameters at the target exercise intensity level and sample exercise energy consumption corresponding to the sample input parameters.

[0090] In addition, the logic instructions in the above-mentioned memory 530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0091] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the exercise energy consumption prediction method provided by the above-mentioned embodiments, the method including: obtaining the acceleration data and target exercise intensity level of the target object during exercise; performing feature extraction on the acceleration data, and obtaining feature information of the acceleration data using principal component analysis; constructing input parameters of the target object based on the feature information and the physical state information of the target object; inputting the input parameters into a target exercise energy consumption prediction model based on the target exercise intensity level to obtain the exercise energy consumption of the target object; wherein, the target exercise energy consumption prediction model is obtained based on sample input parameters at the target exercise intensity level and sample exercise energy consumption corresponding to the sample input parameters.

[0092] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the exercise energy consumption prediction method provided in the above-mentioned embodiments, the method comprising: obtaining acceleration data and target exercise intensity level of the target object during exercise; performing feature extraction on the acceleration data, and obtaining feature information of the acceleration data using principal component analysis; constructing input parameters of the target object based on the feature information and the physical state information of the target object; inputting the input parameters into a target exercise energy consumption prediction model based on the target exercise intensity level to obtain the exercise energy consumption of the target object; wherein, the target exercise energy consumption prediction model is obtained based on sample input parameters at the target exercise intensity level and sample exercise energy consumption training corresponding to the sample input parameters.

[0093] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0094] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0095] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for predicting exercise energy consumption, characterized in that: include: Acquiring acceleration data and a target exercise intensity level of a target subject during exercise, including: acquiring exercise tolerance of the target subject during exercise; and determining the target exercise intensity level according to a value range of the exercise tolerance; Performing feature extraction on the acceleration data, and obtaining feature information of the acceleration data using a principal component analysis method; constructing input parameters of the target object based on the feature information and the physical condition information of the target object; In the case where the exercise intensity levels are divided into mild, moderate and severe, the input parameters are input into a target exercise energy consumption prediction model according to the target exercise intensity level to obtain the exercise energy consumption of the target object, including: when the target exercise intensity level is mild, the input parameters are input into a mild exercise energy consumption prediction model to obtain a first exercise energy consumption of the target object; when the target exercise intensity level is moderate, the input parameters are input into a moderate exercise energy consumption prediction model to obtain a second exercise energy consumption of the target object; when the target exercise intensity level is severe, the input parameters are input into a severe exercise energy consumption prediction model to obtain a third exercise energy consumption of the target object; Among them, the target motion energy consumption prediction model is obtained based on the sample input parameters at the target motion intensity level and the sample motion energy consumption corresponding to the sample input parameters. Each motion intensity level corresponds to a motion energy consumption prediction model, and the target motion energy consumption prediction model is a neural network model; the transfer function of the input layer and the hidden layer of the neural network model is a hyperbolic tangent curve function; the learning process of the neural network model adopts a feedforward / backward propagation algorithm.

2. The method for predicting exercise energy consumption according to claim 1, wherein: The step of constructing input parameters of the target object based on the feature information and the physical condition information of the target object includes: The characteristic information and the physical condition information are standardized so that the average value of the input parameter is a preset value and the variance is less than a preset variance threshold.

3. The method for predicting exercise energy consumption according to claim 1, wherein: The characteristic information includes at least one of the following characteristics: Upper quartile, median, lower quartile, maximum, mean, minimum, kurtosis, skewness, variance, and standard deviation; The body state information includes at least one of the following body state parameters: the age of the target object, the height of the target object, and the weight of the target object.

4. The method for predicting exercise energy consumption according to claim 1, wherein: Obtain the acceleration data of the target object during motion, including: The acceleration data is acquired by using an acceleration sensor arranged at a target human body position of the target object.

5. A sports energy consumption prediction device, characterized in that: include: The first module is configured to obtain acceleration data and a target exercise intensity level of a target subject during exercise, including: obtaining the exercise tolerance of the target subject during exercise; and determining the target exercise intensity level based on a value range of the exercise tolerance; The second module is used to extract features from the acceleration data and obtain feature information of the acceleration data using principal component analysis; A third module is configured to construct input parameters of the target object based on the feature information and the physical condition information of the target object; a fourth module, configured to, when the exercise intensity level is divided into light, moderate, and heavy, input the input parameters into a target exercise energy consumption prediction model according to the target exercise intensity level to obtain the exercise energy consumption of the target object, including: when the target exercise intensity level is light, inputting the input parameters into the light exercise energy consumption prediction model to obtain a first exercise energy consumption of the target object; when the target exercise intensity level is moderate, inputting the input parameters into the moderate exercise energy consumption prediction model to obtain a second exercise energy consumption of the target object; and when the target exercise intensity level is heavy, inputting the input parameters into the heavy exercise energy consumption prediction model to obtain a third exercise energy consumption of the target object; Among them, the target motion energy consumption prediction model is obtained based on the sample input parameters at the target motion intensity level and the sample motion energy consumption corresponding to the sample input parameters. Each motion intensity level corresponds to a motion energy consumption prediction model, and the target motion energy consumption prediction model is a neural network model; the transfer function of the input layer and the hidden layer of the neural network model is a hyperbolic tangent curve function; the learning process of the neural network model adopts a feedforward / backward propagation algorithm.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the exercise energy consumption prediction method according to any one of claims 1 to 4 are implemented.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the exercise energy consumption prediction method according to any one of claims 1 to 4 are implemented.

Images