Accurate recognition of high-difficulty movements and secure management of training data remain critical challenges in the domain of intelligent aerobics training. This study introduces a novel framework that integrates convolutional neural networks with blockchain technology to address these issues from both algorithmic and data governance perspectives. At the algorithmic level, we developed a deep action recognition network that integrates multi-channel multi-scale convolution with a dual attention mechanism. By jointly modeling spatiotemporal image features and channel state information, this architecture significantly enhances the system’s ability to identify complex and challenging aerobics movements. At the system level, blockchain technology is employed to establish a decentralized framework for training data storage and access control, ensuring data integrity verification, traceability management, and cross-device consistency. Experimental evaluation demonstrates that the proposed system achieves superior performance on both benchmark datasets and real-world training environments. Specifically, it attains a recognition accuracy of 96.8% on a custom aerobics dataset, outperforming traditional methods, while blockchain integration increases traceability efficiency by 38.2%. These results underscore the potential of the proposed fusion model to enhance both the precision of movement analysis and the robustness of data security in intelligent sports training systems.
Zhang et al. (Tue,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: