Journal of Networking Technology Xinfei 15 4 2024 https://doi.org/10.6025/jnt/2024/15/4/126-133 https://www.dline.info/jnt/fulltext/v15n4/jntv15n4_2.pdf Body part motion recognition is an important research direction in computer vision and artificial intelligence, widely used in motion analysis, rehabilitation medicine, and game interaction. Traditional methods for recognizing body part movements are usually based on image processing, machine learning, or deep learning techniques. This article proposes a body part motion recognition method based on fuzzy theory and a double-layer hidden Markov model (HMM). This method aims to improve the accuracy and robustness of motion recognition, providing an effective solution for real-time motion capture and pose recognition applications.