http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
몰입형 인터페이스 구현을 위한 DTV와 HNN 기반 손 제스처 인식 알고리즘에 대한 연구
현정환(Jeong-Hwan Hyun),강대성(Dae-Seong Kang) 한국정보기술학회 2017 한국정보기술학회논문지 Vol.15 No.3
In this paper, we suggested an algorithm that learns hand gesture by learning multiple information obtained by Kinect camera and DTV(Distance Transform Vector) through hippocampal neural network algorithm. First, we propose a detection technique of palm area through a preprocessing using Kinect camera and DTV. After that we implemented a hippocampal neural network learning algorithm based on multiple information. And experiments were performed by hand center point, 8-direction gesture, mouse click and screen gesture setting for hand gesture recognition. The proposed hand gesture recognition algorithm is compared with the conventional learning algorithms BP and HMM. And it is proved through experiments that the excellent results can be obtained in terms of recognition rate.