Fine-Grained Classification and Anomaly Detection System of Motion Injury Images Based on Improved YOLO Algorithm
Hong Liu · 2024
Sports injuries refer to the undesirable occurrences that arise during human movement, when muscles contract or stretch due to various factors like external forces and sweat. These injuries not only impose a burden on the body but also affect overall health. Accurate diagnosis and timely treatment are crucial to mitigate the consequences of such injuries. To address this issue, this paper employs the improved YOLO algorithm for the identification and classification of static defects related to sports injuries. The algorithm aims to distinguish between different types of sports injury sites effectively. The experimental results demonstrate that the improved YOLO algorithm can distinguish various types of sports injury sites with accuracy rates of 85%, 92%, 78%, 91%, and 88%, respectively. The varying accuracy levels between different categories highlight the strengths and weaknesses of fine-grained classification systems. The study finds that the improved YOLO algorithm is capable of identifying and classifying sports injury sites with a considerable level of accuracy.