The intelligent tracking and detection of athletes’ actions and the improvement of action standardization are of great practical significance to reducing the injury caused by sports in the sports industry. For the problems of nonstandard movement and single movement mode, this exploration takes the video of sports events as the object and combines it with the video general feature extraction of convolutional neural network (CNN) in the field of deep learning and the filtering detection algorithm of motion trajectory. Then, a target detection and tracking system model is proposed to track and detect targets in sports in real-time. Moreover, through experiments, the performance of the proposed system model is analyzed. After testing the detection quantity, response rate, data loss rate, and target detection accuracy of the model, the results show that the model can track and monitor 50 targets with a loss rate of 3%, a response speed of 4 s and a target detection accuracy of 80%. It can play an excellent role in sports events and postgame video analysis, and provide a good basis and certain design ideas for the goal tracking of the sports industry.

This study uses the video image information in sports video image analysis to realize scientific sports training. In recent years, game video image analysis has referenced athletes’ sports training. The sports video analysis is a widely used and effective method. First, the you only look once (YOLO) method is explored in lightweight object detection. Second, a sports motion analysis system based on the YOLO-OSA (you only look once-one-shot aggregation) target detection network is built based on the dense convolutional network (DenseNet) target detection network established by the one-shot aggregation (OSA) connection. Finally, object detection evaluation principles are used to analyze network performance and object detection in sports video. The results show that the more obvious the target feature, the larger the size, and the more motion information contained in the sports category feature, the more obvious the effect of the detected target. The higher the resolution of the sports video image, the higher the model detection accuracy of the YOLO-OSA target detection network, and the richer the visual video information. In sports video analysis, video images of the appropriate resolution are fed into the system. The YOLO-OSA network achieved 21.70% precision and 54.90% recall. In general, the YOLO-OSA network has certain pertinence for sports video image analysis, and it improves the detection speed of video analysis. The research and analysis of video in sports under the lightweight target detection network have certain reference significance.