Koda'ly Musical Hand Signs Recognition without Visual Background Modeling

CHUR > College Computer Science and Informatics > Department of Computer Science and Information Engineering > Seminar Papers > Item 987654321/31740

Please use this identifier to cite or link to this item: http://chur.chu.edu.tw/handle/987654321/31740

Title:	Koda'ly Musical Hand Signs Recognition without Visual Background Modeling
Authors:	連振昌 Lien, Cheng-Chang
Contributors:	資訊工程學系 Computer Science & Information Engineering
Keywords:	以方塊為基礎;雙前景;膚色偵測;不變矩;追蹤;紋理;支持向量機。 Kodály musical hand signs;Dual foregrounds fusion;Moment invariants;Texture-based fist tracking.
Date:	2012
Issue Date:	2014-06-27 01:36:20 (UTC+8)
Abstract:	以視覺為基礎的手勢辨識技術常會面臨到下列幾個問題。首先是光線的變化常在偵測前景物時容易造成誤判。再者是手部的追蹤於複雜的背景下不易穩定地定位出手部的正確位置。因此我們提出了一項無需背景模型之手勢辨識系統克服上述問題。在本論文中我們提出了兩項創新技術。第一、我們發展以方塊為基礎之連續畫面差異法快速偵測移動的手部。第二、經由長短期雙重之前景融合(fusion of short-term and long-term foregrounds)分析，我們可以獲得正確且完整之手部區域。過去在複雜的背景下不易穩定的定位 In this study, we develop a novel vision-based Kodály musical hand signs recognition system to recognize the gestures of the musical notes. Vision-based gesture recognitions often face the following problems. First, the illumination change can influence t
Appears in Collections:	[Department of Computer Science and Information Engineering] Seminar Papers

Files in This Item:

File	Description	Size	Format
Koda'lyMusicalHandSignsRecognitionwithoutVisualBackgroundModeling.pdf		29Kb	Adobe PDF	234	View/Open