CN-121987203-A - Real-time running fatigue monitoring method, system, electronic equipment and storage medium

CN121987203ACN 121987203 ACN121987203 ACN 121987203ACN-121987203-A

Abstract

The invention relates to the technical field of sports health monitoring, and particularly discloses a real-time running fatigue monitoring method, a real-time running fatigue monitoring system, electronic equipment and a storage medium, wherein the method comprises the steps of receiving real-time heart rate data in a running process; the method comprises the steps of extracting a plurality of heart rate variability characteristics from real-time heart rate data, inputting the plurality of heart rate variability characteristics into a fatigue monitoring model, wherein the fatigue monitoring model is obtained by training based on indoor running data in advance, performing cross-scene optimization by using outdoor running data through migration learning and incremental training strategies, processing the plurality of heart rate variability characteristics through the fatigue monitoring model to generate a real-time fatigue quantification result, and outputting fatigue state information according to the real-time fatigue quantification result. The method solves the technical problems of poor real-time performance of subjective evaluation, application limitation of the existing sensing technology and insufficient dynamic adaptability of the model, and realizes the improvement of running fatigue state real-time monitoring capability and cross-scene adaptability.

Inventors

PAN BINGYU
CUI WEI
WANG YEXUAN
SHEN YANFEI

Assignees

北京体育大学

Dates

Publication Date: 20260508
Application Date: 20260206

Claims (10)

1. A method for real-time running fatigue monitoring, comprising: Receiving real-time heart rate data during running; extracting a plurality of heart rate variability features from the real-time heart rate data; The plurality of heart rate variability characteristics are input into a fatigue monitoring model, wherein the fatigue monitoring model is obtained by training based on indoor running data in advance, and is obtained by performing cross-scene optimization by using outdoor running data by adopting a migration learning and incremental training strategy; Processing the plurality of heart rate variability characteristics through the fatigue monitoring model to generate a real-time fatigue quantification result; And outputting fatigue state information according to the real-time fatigue quantification result.
2. The method of real-time running fatigue monitoring according to claim 1, wherein the step of receiving real-time heart rate data during running comprises: And acquiring an original heart rate signal from the wearable heart rate monitoring equipment through a wireless communication protocol in the running process, performing real-time filtering processing and heart rate peak detection on the original heart rate signal, and generating the real-time heart rate data.
3. The method of real-time running fatigue monitoring according to claim 1, wherein the step of extracting a plurality of heart rate variability features from the real-time heart rate data comprises: Acquiring a continuous heartbeat interval sequence from the real-time heart rate data; intercepting an analysis segment from the heartbeat interval sequence based on a preset time window, and calculating a time domain heart rate variability index and a frequency domain heart rate variability index of the analysis segment; the time domain heart rate variability index and the frequency domain heart rate variability index are combined to the plurality of heart rate variability features characterizing a current physiological state.
4. The method for real-time running fatigue monitoring according to claim 3, further comprising: training an initial neural network model based on an indoor running heart rate variability characteristic sample set to obtain a basic fatigue monitoring model; and loading parameters of the basic fatigue monitoring model by adopting a migration learning strategy, performing cross-scene fine adjustment on the model parameters by utilizing an outdoor running heart rate variability characteristic sample set, and continuously integrating newly-added outdoor running heart rate variability characteristic data by using an incremental training strategy to update the fine-adjusted model parameters so as to obtain the cross-scene optimized fatigue monitoring model.
5. The method of real-time running fatigue monitoring according to claim 4, wherein the step of processing the plurality of heart rate variability features by the fatigue monitoring model to generate real-time fatigue quantification results comprises: And arranging the heart rate variability characteristics into input characteristic vectors according to a preset sequence, inputting the input characteristic vectors into the fatigue monitoring model, calculating the input characteristic vectors through a multi-layer feedforward neural network forming the fatigue monitoring model, obtaining probability distribution of different fatigue grades generated by an output layer of the multi-layer feedforward neural network, and calculating the real-time fatigue quantification result according to the probability distribution.
6. The method for real-time running fatigue monitoring according to claim 5, wherein the step of outputting fatigue status information according to the real-time fatigue quantification result comprises: Comparing the real-time fatigue quantification result with a preset fatigue grade threshold value, and determining a fatigue state grade according to the comparison result; Acquiring a text template and an icon identifier corresponding to the fatigue state grade by inquiring a preset mapping relation, filling the text template by combining the real-time heart rate data and running duration data and associating the icon identifier, and generating the fatigue state information comprising state description and intervention advice; And sending the fatigue state information to a display screen of the user terminal and synchronously outputting the fatigue state information and the loudspeaker.
7. The real-time running fatigue monitoring method according to any of claims 1 to 6, further comprising: Inquiring a preset adjustment strategy table according to the fatigue state grade, acquiring a target speed distribution range parameter and a recommended rest duration parameter corresponding to the fatigue state grade, and combining the target speed distribution range parameter, the recommended rest duration parameter and the fatigue state information into comprehensive prompt information for output.
8. A real-time running fatigue monitoring system, comprising: the acquisition module is used for receiving real-time heart rate data in the running process; An extraction module for extracting a plurality of heart rate variability features from the real-time heart rate data; The input module is used for inputting the heart rate variability characteristics into a fatigue monitoring model, wherein the fatigue monitoring model is obtained by training based on indoor running data in advance, and is obtained by performing cross-scene optimization by using outdoor running data by adopting a migration learning and incremental training strategy; the processing module is used for processing the heart rate variability characteristics through the fatigue monitoring model to generate a real-time fatigue quantification result; And the output module is used for outputting fatigue state information according to the real-time fatigue quantification result.
9. An electronic device comprising a processor coupled to a memory, the memory having stored therein at least one computer program that is loaded and executed by the processor to cause the electronic device to implement the real-time running fatigue monitoring method of any of claims 1 to 7.
10. A computer readable storage medium, characterized in that at least one computer program is stored in the computer readable storage medium, which at least one computer program, when being executed by a processor, implements the real-time running fatigue monitoring method according to any of claims 1 to 7.

Description

Real-time running fatigue monitoring method, system, electronic equipment and storage medium Technical Field The invention relates to the technical field of sports health monitoring, in particular to a real-time running fatigue monitoring method, a real-time running fatigue monitoring system, electronic equipment and a storage medium. Background With the rapid development of body building and competitive sports of the whole people, running is taken as a basic sports project, and scientific training and sports safety management of the running project are increasingly paid attention to. Sports fatigue is a key factor affecting running performance and increasing sports injury risk, and real-time accurate monitoring of fatigue state in running process has important significance for optimizing training load and preventing sports injury. However, the prior art still has obvious defects in real-time running fatigue monitoring, and is difficult to meet the engineering application requirements. The traditional exercise fatigue evaluation mainly depends on subjective evaluation methods such as subjective effort score and the like, the methods take subjective perception of an athlete as a core, fatigue information is obtained through a post-hoc or intermittent query mode, the method is easily influenced by individual cognitive deviation, real-time automatic detection and quantitative output in the exercise process cannot be realized, and the automatic application requirements of real-time early warning and dynamic training regulation and control are difficult to meet. In recent years, the progress of wearable sensing technology provides a new technical means for exercise monitoring, the existing research realizes the quantitative management of training load by collecting physiological exercise data such as heart rate, acceleration and the like, but related applications are concentrated on staged training summary or post statistical analysis, and the real-time fatigue state online detection capability aiming at running exercise characteristics is lacking, so that instant feedback and risk early warning cannot be provided in the exercise process. Further, the existing fatigue recognition technology based on wearable sensing mainly focuses on multi-source information fusion and static mode recognition, and sensor data are classified or subjected to regression analysis through a pre-training model to obtain a fatigue recognition result. The technical scheme does not fully consider the dynamic process characteristics and the fatigue accumulation change trend of running exercise, lacks the targeted modeling of a fatigue state evolution mechanism, and lacks a systematic technical scheme at the engineering realization level of a real-time detection mechanism and an online early warning feedback logic, so that the real-time adaptability and the dynamic updating capability of the model in an actual running scene are insufficient, and the performance fluctuation caused by indoor and outdoor environment differences and individual differences is difficult to effectively realize. Therefore, a technical scheme capable of realizing real-time monitoring of running fatigue state, having cross-scene adaptive capacity and supporting dynamic updating optimization is needed, so as to solve the technical problems of poor real-time performance of the existing subjective evaluation method, application limitation of the existing wearing sensing technology and insufficient dynamic adaptability of the existing fatigue recognition model, and provide reliable real-time fatigue monitoring technical support for motion safety management and scientific training. Disclosure of Invention In order to solve the technical problems, the invention provides a real-time running fatigue monitoring method, a real-time running fatigue monitoring system, electronic equipment and a storage medium. In a first aspect, the present invention provides a real-time running fatigue monitoring method, which has the following technical scheme: Receiving real-time heart rate data during running; extracting a plurality of heart rate variability features from the real-time heart rate data; The plurality of heart rate variability characteristics are input into a fatigue monitoring model, wherein the fatigue monitoring model is obtained by training based on indoor running data in advance, and is obtained by performing cross-scene optimization by using outdoor running data by adopting a migration learning and incremental training strategy; Processing the plurality of heart rate variability characteristics through the fatigue monitoring model to generate a real-time fatigue quantification result; And outputting fatigue state information according to the real-time fatigue quantification result. The real-time running fatigue monitoring method has the following beneficial effects: according to the method, the heart rate data are collected in real time, the heart rate variability chara