【摘要】 针对字母手势的检测和跟踪问题,文章提出一种基于最大似然准则Hausdorff距离的手势识别算法。该算法首先对字母手势图像进行二值化处理,并由字母手势图像的边缘信息中提取字母手势的关键点(指根和指尖);然后采用基于最大似然准则的Hausdorff距离对手势进行识别,搜索策略采用类似于Rucklidge提出的多分辨率搜索方法,在不影响成功率和目标定位精度的情况下,可以显著地缩短搜索时间。实验结果表明此方法可以较好地识别字母手势,同时对部分变形(旋转和缩放)手势也有良好的效果。更多还原

【Abstract】 Aiming at the detection and tracking issues of hand gesture this paper proposes a new method of hand gesture recognition based on maximum likelihood criterion Hausdorff distance. The method first detects the binary image from the hand gesture image, and extracts the key points as feature vectors from hand gesture’s edge image, and then uses the maximum likelihood Hausdorff distance to recognize the hand gesture. The searching strategy adopts Rucklidge’s methods, so that a much higher searching speed can be achieved while maintaining the search success rate and target location accuracy. Experiment results show that the proposed algorithm can recognize hand gesture efficiently, and it also can match partially deformable hand gesture accurately.更多还原