Based on biological principles, too great knee flexion will have the following effects on college students' running posture:
Firstly, too great knee flexion will aggravate the joint burden. Excessive knee flexion will lead to excessive knee joint bending, resulting in excessive joint load, which is easy to cause joint fatigue and injury [15]. Secondly, too great knee flexion will increase the risk of knee injury. Excessive knee flexion can easily lead to knee twisting or spraining during running, thus increasing the risk of knee injury [16]. In addition, too great knee flexion will affect the running effect. Excessive knee flexion will affect running posture and movement, make running effect worse, and even affect running speed and endurance. Finally, too great knee flexion will aggravate physical fatigue. Excessive knee flexion will increase physical fatigue, because knee joints need to consume more energy to support the weight and movement of the body [17].
Therefore, when running, college students should pay attention to the control of knee flexion to avoid excessive knee bending and twisting [18]. It is possible to reduce the load on knee joints through correct running posture and movements, and meanwhile strengthen the exercise of related muscles and improve the endurance and adaptability of the body to better enjoy the fun of running [19, 20].
Unstable center of gravity
The center of gravity is the balance point of human body in sports, which has an important influence on the stability and efficiency of running posture. When the center of gravity is unstable, it will have the following effects on the running posture of college students:
First, the unstable center of gravity will increase the risk of falling while running. When the center of gravity is unstable, the body is easy to lose its balance, leading to an increase in the risk of falling [21]. In this case, runners need to spend extra energy to keep balance, which affects running speed and endurance. Meanwhile, falling down during running is also easy to cause physical injury, which will damage the health benefits of running [22]. Secondly, the unstable center of gravity will increase the joint load. In order to keep balance, people may rely too much on the knees and ankles when running, which will increase the joint load and lead to joint fatigue and injury. In the long run, this situation may lead to diseases such as arthritis and affect the quality of life of runners [23]. In addition, the unstable center of gravity will also affect the running posture and movements, making the running effect worse. When the center of gravity is unstable, runners need to keep their balance by shaking their bodies, which will make the running posture and movements unstable, thus affecting the running effect. In fact, correct running posture and movements can help runners make better use of their body energy, improve running efficiency and avoid unnecessary energy waste.
In order to avoid the negative impact of unstable center of gravity on running, college students should pay attention to maintaining the stability of center of gravity. People can improve the balance ability of the body by strengthening the exercise of core muscles to maintain the position of the center of gravity [24]. In addition, it is necessary to pay attention to the correctness of running posture and movements during running to avoid excessive lateral and up-and-down shaking. This can better protect joints and improve running efficiency and health benefits. In addition to strengthening the exercise of core muscles and paying attention to the correctness of running posture, there are other ways to help college students maintain the stability of their center of gravity [25]. For example, it is very important to choose running shoes that suit you. Running shoes should have sufficient support and buffering capacity to reduce the pressure on the feet and knees, thus improving the comfort and stability of running. In addition, adjusting the breathing rhythm can also help maintain the stability of the center of gravity. Through deep breathing and slow breathing rhythm, the oxygen supply and consumption of the body can be balanced, thus reducing the shaking and shaking of the body. Finally, it is also important to choose the appropriate route and terrain when running, and avoid choosing sections with excessively steep slopes or uneven roads to reduce body shaking and bumps, and thus improve the stability and efficiency of running [26].
In a word, maintaining the stability of the center of gravity is very important for college students to run. By strengthening the exercise of core muscles, paying attention to the correctness of running posture, choosing running shoes suitable for you, adjusting breathing rhythm and choosing appropriate routes and terrain, the stability and efficiency of running can be effectively improved, and the risk of physical injury can be reduced to better enjoy the fun of running.
Design of running posture optimization model based on biomechanics principle
In this paper, the running posture optimization model is an application based on biomechanics principle and machine learning technology, which can help runners improve their running effect and health level. In actual running, bad running posture may lead to sports injury, fatigue, inefficiency and other problems, and by optimizing running posture, these problems can be reduced and the running effect can be improved [27].
The design of running posture optimization model needs to consider many aspects. First, it is necessary to collect a large number of running data of runners of different sizes and levels to establish a reasonable data set [28]. Then, data preprocessing is needed, including data cleaning, standardization, feature extraction and other operations to generate data that meets the input requirements of the model. Then, according to the biomechanical principle and the actual situation, people need to select the characteristics related to running posture optimization, such as knee flexion, center of gravity position and so on [29]. Then, appropriate machine learning algorithms, such as decision tree, neural network and SVM, can be selected to train the model. In model training, cross-validation and other techniques can be used to evaluate the performance and accuracy of the model. Finally, the trained machine learning model is applied to the actual running, and the corresponding running posture optimization suggestions are provided by monitoring the running data of runners [30]. As shown in Figs. 1–4, the code of this model design is displayed.
In Figs. 1–4, this model is a running posture optimization model based on biomechanics principle and machine learning technology. The model collects the running data of runners, including landing point, knee flexion, center of gravity and other parameters, and selects features related to running posture optimization, such as knee flexion, center of gravity and so on, according to biomechanical principles. Then, the machine learning algorithm is used to train the model, and the trained model is applied to the actual running, and the corresponding running posture optimization suggestions are provided. The design of running posture optimization model is based on biomechanics principle. By analyzing the movements of key parts in running, such as knees, feet and hips, the reasonable running posture is determined to improve the running effect and health level. Meanwhile, the running posture optimization model is trained by machine learning algorithm, which can automatically learn the rules and modes of running posture optimization and improve the accuracy and accuracy of running posture optimization. The application of running posture optimization model is very extensive. In the training of athletes, the running posture optimization model can help coaches to better guide athletes' running skills and improve their competitive level. Among ordinary runners, the running posture optimization model can help runners to improve their bad running posture, reduce sports injuries, fatigue, inefficiency and other problems, and improve the running effect and health level. It should be noted that the running posture optimization model is only an auxiliary tool. In actual running, runners need to make appropriate adjustments and optimizations according to their own conditions to achieve the best running effect and health level. Meanwhile, the design of running posture optimization model needs to consider many factors, such as data quality, feature selection, model algorithm and model application to improve the performance and accuracy of the model.
Model parameters and experimental index design
Sports-1M data set is a large-scale video motion recognition data set, which contains more than 1 million video clips and covers 487 different sports categories. The following is the content of data preparation when using Sports-1M data set to train and test the running posture optimization model:
1. Data download: Sports-1M data set can be downloaded from the official website at the following address: http://www.sports1m.com/
2. Data preprocessing: As the Sports-1M data set is video data, the following preprocessing is needed: (1) Video sampling: randomly sample a video segment with a fixed length from each video. (2) Video decoding: converting video into a series of frame images. (3) Image preprocessing: preprocess each image, including cutting, scaling, normalization and other operations.
3. Feature extraction: For each video clip, it is necessary to extract relevant features, such as knee flexion and center of gravity position, as the input of the model. Specific feature extraction methods can be selected and designed according to the actual situation, such as using convolutional neural network to extract features.
4. Data partition: The data set is divided into training set and test set, and cross-validation and other technologies are usually used to evaluate the performance and accuracy of the model.
5. Data enhancement: In order to improve the generalization ability of the model, data enhancement operations can be performed on the data set, such as random rotation, translation and scaling.
The data set that meets the input requirements of the model can be obtained through the above data preparation work, and can be used for the training, test, and evaluation of the operation situation optimization model. In model training, cross-validation and other techniques can be used to evaluate the performance and accuracy of the model, and at the same time, the performance and accuracy of the model can be optimized by trying different model parameters, feature selection and algorithm selection. In addition, based on the performance evaluation of the model by the above data set, this paper evaluates the effect of the model in practical application. This paper tests the running posture of 30 college students in total, and provides a plan to improve the running posture of college students through the model, thus effectively improving the running effect of college students. All image datasets in the original data are publicly available. All methods used in this experiment were carried out in accordance with relevant guidelines and regulations. All experimental plans in this experiment were approved by the Sports Ethics Committee of Shaanxi Normal University. This experiment has received informed consent from all participants. The specific experimental process is as follows:
First, the running posture data of 30 college students is collected, including step frequency, stride length and foot landing angle, and the data is cleaned and standardized to ensure the accuracy and consistency of the data. Then, the trained model is applied to the actual scene, and the running posture optimization scheme is provided for college students. After 15 days of exercise, the adjustment effect of the model designed in this paper on running posture is tested. The running posture of 30 college students is tested, and personalized optimization suggestions are provided according to the output of the model. Finally, the actual effect of running posture optimization scheme is evaluated, and the running effect before and after improvement is compared. Based on this, this paper mainly collects the running postures of college students and makes intelligent analysis on them, and does not involve any other information.
Design of evaluation index for running posture optimization model:
1. Accuracy:
(1) Definition: Accuracy refers to the proportion of correctly classified samples to the total number of samples.
(2) Calculation method: Accuracy = number of correctly classified samples/total number of samples.
2. Recall rate:
(1) Definition: Recall rate refers to the proportion of correctly classified positive cases to all positive cases.
(2) Calculation method: recall rate = number of correctly classified positive cases/ number of all positive cases.
3. F1 value
(1) Definition: F1 value is the harmonic average of accuracy and recall, which is used to comprehensively evaluate the performance of the model.
(2) Calculation method: F1 value = 2* (accuracy * recall rate)/(accuracy + recall rate).
It should be noted that the above evaluation indicators are only a basic framework, and the specific indicators need to be selected and designed according to the actual situation. Meanwhile, in the evaluation of running posture optimization model, other factors, such as error analysis and efficiency, need to be considered to comprehensively evaluate the performance and accuracy of the model.
Evaluation of running posture optimization model based on biomechanics
Training evaluation of running posture model
The concept of training and evaluation of running posture optimization model is based on data-driven. That is, by collecting and processing runners' running data, extracting relevant features, using machine learning algorithm to train the model, and evaluating it through test set, running posture can be optimized, and running effect and health level can be improved. Specifically, the training and evaluation process of running posture optimization model includes data preparation, model training and evaluation. In the data preparation stage, it is necessary to collect runners' running data, perform cleaning, standardization, feature extraction and other operations to generate data sets that meet the requirements of model input. In the model training stage, it is necessary to select appropriate machine learning algorithms, such as decision tree, neural network, SVM, etc., use the training set to train the model, and adjust the model parameters according to the actual situation to improve the model performance and accuracy. In the evaluation stage, it is necessary to use the test set to predict the model, get the prediction results, and select evaluation indicators and evaluation methods according to the actual situation, such as accuracy, recall, F1 value, cross-validation, etc., to evaluate the performance and accuracy of the model. The concept of training and evaluation of running posture optimization model emphasizes the importance of data. By collecting and processing a large number of running data, extracting relevant features and using machine learning algorithm to train and optimize the model, the running effect and health level can be effectively improved. Meanwhile, the training and evaluation of running posture optimization model need to consider many factors, such as data quality, feature selection, model algorithm and model application to improve the performance and accuracy of the model. Based on this, this paper first evaluates the performance of the decision tree algorithm model. In Fig. 5, the results of model evaluation are shown.
In Fig. 5, when the decision tree algorithm is used to train and evaluate the model, the accuracy, recall and F1 value of the training set are higher than those of the test set, indicating that the model has some over-fitting. It is necessary to further optimize the model and reduce the complexity of the model to improve the generalization ability of the model. Based on this, this paper also comprehensively evaluates the performance of the SVM algorithm model. Figure 6 shows the evaluation results of the model.
In Fig. 6, when the model is trained and evaluated by using the SVM algorithm, the accuracy, recall and F1 value of the training set and the test set are close, which shows that the model has good generalization ability. Further optimization of model parameters can be considered to improve the performance and accuracy of the model. Finally, this paper evaluates the performance of the neural network algorithm model. In Fig. 7, the model evaluation results are displayed.
In Fig. 7, when the neural network algorithm is used to train and evaluate the model, the accuracy, recall and F1 value of the training set are higher than those of the test set, indicating that the model has some over-fitting. It is necessary to further optimize the model structure and parameters, reduce the complexity of the model and improve the generalization ability of the model.
To sum up, the above three groups of model evaluation data illustrate the influence of different machine learning algorithms on the running posture optimization model. The complexity, stability, generalization ability and other factors of the algorithm need to be considered when selecting the machine learning algorithm, and the algorithm should be selected and designed according to the actual situation. Meanwhile, in model training and evaluation, it is necessary to comprehensively evaluate the model according to the evaluation index and evaluation method to improve the performance and accuracy of the model.
Evaluation of running posture optimization results
Based on the above model evaluation model, this paper uses the above three models to adjust the running posture of college students to optimize their running posture and improve their running effect. This paper mainly adjusts the body movements of college students during running, and evaluates the running process by testing the EMG signals during running to explore the effect of the model on the optimization of college students' running posture. In Tables 1–3, the results of adjusting the running posture of college students under three models are shown.
Table 1
Optimization effect of running posture under decision tree model
Test sample number
|
Original running posture score
|
Adjusted running posture score
|
1
|
70
|
80
|
2
|
65
|
78
|
3
|
75
|
85
|
4
|
80
|
88
|
5
|
72
|
81
|
Table 2
Optimization effect of running posture based on SVM model
Test sample number
|
Original running posture score
|
Adjusted running posture score
|
1
|
70
|
85
|
2
|
65
|
80
|
3
|
75
|
90
|
4
|
80
|
92
|
5
|
72
|
87
|
Table 3
Optimization effect of running posture under neural network algorithm model
Test sample number
|
Original running posture score
|
Adjusted running posture score
|
1
|
70
|
88
|
2
|
65
|
82
|
3
|
75
|
91
|
4
|
80
|
87
|
5
|
72
|
85
|
In Tables 1–3, after using decision tree model, SVM algorithm model and neural network algorithm model to adjust the running posture of athletes, they can effectively improve the running posture score of athletes, thus improving the running effect and health level. Among them, the adjustment effect of neural network algorithm model is better, which can improve the running score by about 15 points at the highest, which can improve the running posture of athletes and score more. This shows that in the selection and design of running posture adjustment model, the choice of machine learning algorithm has an important influence on the performance and accuracy of the model.