| Summary: | 碩士 === 國立中興大學 === 資訊科學與工程學系 === 104 === Recently, there are a lot of sports related researches are published, especially baseball, golf and tennis. Exercisers can check if their postures are standard, according to those analyzing result. They will further understand how to improve their skills and what to improve.
In contrast, studies about badminton are rarely seen. This may result in much deeper learning curve on badminton. Although there are still some costumed sensors that combined with Apps providing users these information, but they are limited in pitches analysis and only few basic pitched are available.
Therefore, this paper discusses how to use a six-axis sensor to further analyze badminton action, even for athletes without a coach, they still can use the sensor to train themselves and improve their performance.
To achieve this goal, we combined the three-axis accelerometer and gyroscope to obtain posture analysis parameters. Through the measured data to obtain the feature, then use these features to classify the pitch type. The results showed that the proposed classification method can achieve 97% average accuracy rate, which is much higher than the result of Support Vector Machine (SVM) which is 87%.
|
|---|
