Study Finger Detection Techniques Using Physical Feature

• Main Authors: Chao-Yueh Hsiao, 蕭朝岳
• Other Authors: Wen-Yuan Chen
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/68189641290444567747

碩士 === 國立勤益科技大學 === 電子工程系 === 102 === With the advancement of technology, human-computer interaction interface has been the subject of many people’s researches, from the early keyboard and mouse to the current cameras and microphones and other tablet products. In recent years, the sold somatosensory games such as Nintendo Wii, Xbox and Kinect can allow users to manipulate the game by using the human body rather than the remote control or the handle to operate the game. Through the color detection and image subtraction technique, this study found the hand image, and the histogram analysis was then used to find the palm and wrist arm junction points. After obtaining the palm image, to the convex hull algorithm was used to analyze the characteristics of the palm to identify the portion of fingertips. Finally, by the midpoint of the fingertips and the wrist, the possible gestures of wrists were calculated. The experimental results show that this method can effectively remove background and keep the hand image, so the palm image is segmented out to find exactly the fingertip position and identify a variety of gestures.