The personal fitness model identification research

• Main Authors: TSENG, SHAO-TING, 曾紹庭
• Other Authors: CHEN, YUAN-HSIN
• Format: Others
• Language: zh-TW
• Published: 2019
• Online Access: http://ndltd.ncl.edu.tw/handle/5wz4pv

碩士 === 國立聯合大學 === 電子工程學系碩士班 === 107 === In recent years, the proportion of Taiwan's sports population has increased year by year. More and more people are willing to enter the gymnasium, and everyone has gradually moved from simple running to Fitness and Power Lifting exercises . In the past, few documents were discussed or To study the pattern recognition of Power Lifting, in this paper I will use the sensor to obtain the signal of the exercise, and then use these signals to build the model of each exercise, and then use the established model to identify the current motion and improve the accuracy of the recognition. In this paper, the smart phone built-in three-axis acceleration sensor is used to collect the motion data, and the collected motion data is transmitted to the host computer through wifi, select the identification feature and use the KNN algorithm to establish a variety of motion models, and at the same time establish a confusion matrix, The model predicted data is compared with the test data, and the classification effect of the model can be clearly observed, then we try to train the Artificial Neural Network to observe if it can be better identified by training Artificial Neural Network. Finally, we find the model that is most suitable for the algorithm to distinguish the current model of the exercise.