Application of Video Abnormal Behavior Detection Algorithm in Evaluation of Track and Field Teaching and Training Effect

Introduction

The struggle for talent is growing more intense as society develops, and the value of education is becoming more and more significant. Since education has a significant impact on both the amount and quality of talent, governments everywhere have implemented a number of initiatives to establish high-quality education as a fundamental national strategy. One of the required courses in sports colleges is track and field teaching, which is crucial for advancing the mental and physical well-being of the pupils. But as “skill-based athletics” education continues to evolve and be innovative, so does the need for athletics education. The conventional view of education holds that the distribution of student grades should follow a normal distribution, beginning with the hierarchical and static concept of education. Thus, this “small at both ends and large in the middle” condition has always been maintained by the teacher's instruction. But given the current state of communist education, all pupils must receive an education, and raising the standard of living for all citizens is necessary for the socialist economy to grow. The enthusiasm of college athletics teachers has been severely hampered, and the comprehensive and quick development of female athletics teaching in universities has been impeded by reports in certain media outlets of frequent teaching accidents in college athletics. This has also resulted in mental harm to parents of students and had a strong negative impact. Thus, the primary goal of this research is to use video abnormal behavior detection algorithms to thoroughly identify and assess the risk elements of track and field teaching. Thus, the primary goal of this research is to use video abnormal behavior detection algorithms to thoroughly identify and assess and investigate the primary risk factors associated with track and field teaching, as well as build a risk assessment index system and manage and quantitatively evaluate the risks.

The innovation of this article lies in certain concepts:

• (1)

  In the process of feature extraction, in order to improve the robustness and effectiveness of the extracted features and remove some unavoidable noises in the process of feature extraction and foreground background separation, a complex wavelet domain denoising method is proposed.

• (2)

  In the classification model, we use particle swarm optimization twin support vector machines (PSO TSVM). The support vector machine (SVM) only constructs a classification hyperplane for different classification samples, while the twin support vector machin (TSVM), developed by SVM, constructs a hyperplane for each type of samples, which has a good processing ability for unbalanced data. Compared with SVM, the complexity of TSVM in solving second class optimization problems is ¼ of that of SVM, and it has greater advantages in speed than SVM.

Literature Review

As a physical education teacher, you must be quite familiar with and understand the technical principles, action structures, etc. of the sports events you teach and be able to find the inherent laws and essence in the technical links. If you do not have standardized technical actions or are not clear about the technical principles and action structures, it is unimaginable to teach, and you are completely incompetent for sports teaching. Sports teaching contents includes track and field, outdoor exercise, and directional cross-country(Zhang, 2023). The resetting of track and field curriculum not only changed the training objectives, but also updated the teaching contents, including outdoor sports, orienteering, outdoor life survival, and so on. Visible, the current track and field course teaching evaluation system also needs to be rebuilt (Yildirim et al.,2023).