Peg-Free Hand Features Extraction

• Main Authors: Cheng-Chi Loo, 駱成麒
• Other Authors: Li-Min Liu
• Format: Others
• Language: zh-TW
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/36710958937234658959

碩士 === 中原大學 === 數學研究所 === 100 === 　　Existing palm-recognition machines require using pegs for image acquisition. The main purpose of this study is to acquire high-quality and high-accuracy palm features in a peg-free environment in order to reduce environmental constraints during image acquisition. The experiments use three peg-free pictures acquired from the same individual, and use computer programs to automatically extract features on palm. Extracted features may be different due to muscle scaling and the direction fingers pointing to. The final features are derived by averaging features from those three samples. The final features are compared with the original features and the error ranges are (1) length feature error: 1.52%, (2) area feature error: 1.60%, and (3) perimeter feature error: 1.15%.