提出了一种基于自适应模板半径的形状上下文描述子描述图像轮廓特征的手势动作模式识别方法,并探讨了这种方法在基于视觉的汉语字母手势识别中的应用.首先,在计算手势轮廓各点的形状上下文时,提出了一种距离直方图法求取自适应的模板半径的方法,并以χ2统计量计算各轮廓点之间的匹配距离,再以改进Hausdorff距离计算各手势轮廓间的匹配距离,最后用最近邻法进行分类识别.对来自50个实验对象的手势图像分类识别的实验结果表明,该方法对30个汉语字母手势图像的正确识别率达到了74.3%,明显高于基于不变矩及傅里叶描述子特征提取方法的识别结果.

1. Wu Ying,Huang Thomas S.Human hand modeling,analysisand animation in the context of HCI[C].Proceedings of International Conference on Image Processing,Kobe,1999:6-10.
2. Zhou Hanning,Lin DJ,Huang TS.Static Hand Gesture Recognition based on Local Orientation Histogram Feature Distribution Model[J].Proceed-ings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Worksh-ops:1063-6919.
3. Starner T,Weaver J,Pentland A.Real-Time American Sign Language Recognition Using Desk and Wearable Computer Based Video[J].IEEE Transactions on Patt-ern Analysis and Machine Intelligence,1998,20(12):1371-1375.
4. 任海兵,徐光,林学.基于特征线条的手势识别[J].软件学报,2002,13(5):987-993.
5. 朱继玉,王西颖,等.基于结构分析的手势识别[J].计算机学报,2006,29(12):2130-2137.
6. Hyeon-Kyu Lee,Kim JH.An HMM-based threshold model approach for gesture recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1999,21(10):861-972.
7. Serge Belongie,Jitendra Malik,Jan Puzicha.Shape Matching and Object Recognition Using Shape Con-texts[J].IEEE transactions on Pattern Analysis and Machine Inteligence,2002,24(24):509-522.
8. 田晓亮,刘忠.基于形状相似度的水下目标识别算法[J].声学技术,2007,26(3):493-497.
9. 陈实,马天骏,等.基于形状上下文描述子的步态识别[J].模式识别与人工智能,2007,20(6):794-799.
10. Dubuission MP,Jain AK.A Modified Hausdorff Distance for Object Matching[J].Proc of the 12th International Conference on Pattern Recognition,1994,I:566-568.