A real-time tracking method for tennis ball using Kinect

• Main Authors: TAI, YU-TING, 戴妤庭
• Other Authors: LIN, WEI-CHENG
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/ab9n3x

碩士 === 國立高雄應用科技大學 === 資訊工程系 === 104 === Image processing is a technology utilizing machining, analysis and treatment for images to achieve the desired demands. For example, security companies may apply the tracking and face detection of image processing technology to the access control systems, thereby enhancing the security service. Accordingly, many aspects of daily life are closely related to applications of image processing. However, with the development of the city, spaces capable of doing exercise in life are less and less. As a result, in recent years, it’s more and more popular to do exercise with great movements in small indoor spaces, such as indoor batting cages, indoor golf. Furthermore, in terms of tennis, the field required for tennis is too large to play in the crowded city, such that detail information of the moving track which can be got on the actual court cannot be obtained in the small spaces. Therefore, the present research provides the technology of image processing and Kinect to track the movement of tennis in real-time. Specifically, Euler Angles and speed at the moment of tennis flying and instant posture of swinging can be obtained by image processing algorithm. After transferring the data to the simulation software developed by Unreal Engine4 and simulating the flying track of tennis on the actual court, the effect of striking on the actual court can be obtained by training in the small spaces.