DoYeob Lee, DongKyoo Shin, DongIl Shin, Journal of Internet Computing and Services, Vol. 17, No. 2, pp. 29-38, Apr. 2016
10.7472/jksii.2016.17.2.29, Full Text:
Keywords: Kinect, Hand Gesture, Hand Gesture Recognition, Depth Image

Abstract

Humans develop and maintain relationship through communication. Communication is largely divided into verbal communication and non-verbal communication. Verbal communication involves the use of a language or characters, while non-verbal communication utilizes body language. We use gestures with language together in conversations of everyday life. Gestures belong to non-verbal communication, and can be offered using a variety of shapes and movements to deliver an opinion. For this reason, gestures are spotlighted as a means of implementing an NUI/NUX in the fields of HCI and HRI. In this paper, using Kinect and the geometric features of the hand, we propose a method for recognizing the number of fingers and detecting the hand area. A Kinect depth image can be used to detect the hand region, with the finger number identified by comparing the distance of outline and the central point of a hand. Average recognition rate for recognizing the number of fingers is 98.5%, from the proposed method, The proposed method would help enhancing the functionality of the human computer interaction by increasing the expression range of gestures.

Statistics
Cite this article