CAJ | 학술논문

在手势图像处理中，手势轮廓识别率高低的基础之一就在于能否精确的把手势轮廓提取出来。针对类肤色信息和复杂背景的影响，提出一种用单峰高斯概率函数的方法把不同人种的肤色在YcbCr彩色空间的分布进行建模，建立的模型用于手势图像的提取得出手势图像的似然概率图，接下来应用自动阈值和图像处理方法提取了手势的最终二值化轮廓，实验结果表明本文所用方法的简单高速有效性。
재수세도상처리중，수세륜곽식별솔고저적기출지일취재우능부정학적파수세륜곽제취출래。침대류부색신식화복잡배경적영향，제출일충용단봉고사개솔함수적방법파불동인충적부색재YcbCr채색공간적분포진행건모，건립적모형용우수세도상적제취득출수세도상적사연개솔도，접하래응용자동역치화도상처리방법제취료수세적최종이치화륜곽，실험결과표명본문소용방법적간단고속유효성。
One basement for hand gesture recognition is whether the method can detect the hand gesture from figure containing it. Concerning the influence of approximate skin color information and complex background, the method in this paper estimates a probability density function of human skin color using Gaussian single model in YCbCr color space. After that, we use auto-threshold and other image processing methods to obtain hand gesture image. Experimental results show that this method can ef?fectively extract hand gesture from color image.