Basketball Shooting and Goaling Detection Based on DWC-YOLOv8

Smart sports provide intelligent and digital sports services and experiences through the application of artificial intelligence technology, promoting the development and progress of the sports field. This article introduces a basketball shooting and goaling detection algorithm based on DWC-YOLOv8, aiming to promote the intelligent development of campus basketball games. Firstly, we use the self-made dataset to train the YOLOv8 object detection model based on depth-wise separable convolution with the of 98.9% and FPS of 52. The training results show that the model has high detection accuracy for basketball and basket, and meet the requirements of real-time detection. Then, a judgment method for detecting shots and goals is proposed. Four sets of basketball game shooting clips are selected for experiments and the recognition rate for shots and goals is 95.92% and 95.83%, respectively. The experimental results show that the algorithm effectively realizes the detection of basketball shots and goals. This study is of great significance for promoting the intelligent and digital development of campus basketball games.