Design of an Olympic performance prediction system based on optical sensing technology and facial expression recognition

Abstract In the development research of sports, predicting performance is a crucial task. Through scientific prediction methods, athletes can be better guided to train and develop scientific training plans. At present, the prediction of sports performance mainly relies on expert experience, and its accuracy is relatively low. Based on this, in order to address the challenge of large-scale data, we propose an Olympic performance prediction system based on genetic algorithm and related facial expression recognition technology. The system utilizes the self-organization, adaptability, and intelligence of genetic algorithm, with the main goal of improving search efficiency. After multiple comparative experiments, the algorithm proposed in this article shows higher accuracy in feature selection of multidimensional data, almost surpassing the level of all other algorithms. In order to improve the overall running speed of the system, parallel processing technology was adopted. The results showed that our designed system demonstrated excellent performance advantages in both user connection count and HTTP connection count testing, providing support for in-depth research on Olympic performance and expanding competitive sports. At the same time, it also improved the accuracy of Olympic performance prediction and provided data support.