Human-Robot Interaction Method Using Geometric Feature of Hand Gesture 

 

 In this paper, we propose hand gesture recognition of six kind method using geometric feature and apply the smart contents Geometric feature of hand gesture defines histogram using distance between the center point of the extracted feature pointThe outer point is sequential for conditions of establishing the histogram.Therefore this paper uses the filtering snake model in the process of outer feature extraction.The entire configuration of Hand gesture recognition systems is as follows. Hand center point is extracted using skin color and difference filter and hand's ROI track using simple CAM-SHIFT based on skin color. Extracts clear hand region from traced ROI. And simultaneously recognize hand gesture and hand orientation using histogram between contour point and center point extracted by filtering snake method. The robot control application was implemented using this method for smart OS environment. This paper use Massey database(public database) and built the self database of different environment, angle, size. This paper derive 8.8msec reduction in computation speed and 12.5% rising in recognition rate through comparison analysis about proposal algorithm and the existing algorithm

​

Yang, Dongseok, Jong-Kuk Lim, and Younggeun Choi. "Early childhood education by hand gesture recognition using a smartphone based robot." Robot and Human Interactive Communication, 2014 RO-MAN: The 23rd IEEE International Symposium on. IEEE, 2014.