Specified Gestures Identification using Gaussian Mixture Model

• Main Authors: Yung-Chun Sue, 蘇詠鈞
• Other Authors: 周復華
• Format: Others
• Language: zh-TW
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/15219540833691164661

碩士 === 清雲科技大學 === 電子工程所 === 100 === Sign language recognition technique is composed by the hand images detection and the hand gestures recognition. Hand images detection is locating the sign language select, sign language capture, the palm and fingers part from the sensed image, and rotating them to the appropriate hand posture, both are the important pre-processing for sign language identification and recognition. This paper first introduced sequentially throughout the study practices, as well as the process of image pre-processing instructions. The major work in the hand gestures recognition is to identify the variance of the fingers. In this paper the creation of sign language image of slash encoding, Department of the advantages of slash encoding the difference between your fingers the number of changes, and the Gaussian mixture model (GMM) to establish the model of sign language and identification. Such as poor recognition rate is adjusted probability distribution of weight values to improve the recognition rate. The entire the paper Shushing is the Gaussian mixture model (GMM), slash code, adjust the probability distribution of the weight value. Finally, after adjusting the probability distribution of weight values, we learned from the conclusion that the overall recognition results rose to 98.33%from 92.66% of the original, so changing the probability distribution of the weight value can effectively improve the recognition rate.