An AI-based badminton smash detection using residual-shuffle network optimized based on upgraded pufferfish optimizer

The growth in the field of sports analytics has observed a paradigm shift with the advent of artificial intelligence (AI) and deep learning techniques. The accurate detection of smashes in badminton is significant for strategic decision-making and performance enhancement. This research presents an innovative AI-based framework, called DeepSmash, which uses a Residual-Shuffle Network (ResNet) optimized by an upgraded Pufferfish Optimizer (UPO) to automatically detect badminton smashes from broadcasted video footage. By a combination of the strengths of ResNet's hierarchical feature representation and UPO's efficient parameter tuning, the model achieves high-precision recognition of smashes, drops, clears, net actions, and lifts. A comprehensive comparative analysis with state-of-the-art models demonstrates the superiority of our proposed approach, underscoring its potential to revolutionize sports analytics and athlete performance enhancement. This innovation not only sets a new benchmark for AI applications in sports technology but also covers the way for the development of more efficient sports analytics tools.