摘要
对图像中的目标进行自动识别,既节省了人力,又提高了信息的时效,是模式识别发展的一个重要方向.文中用矩不变量表示目标的形状特征,并对其在离散时不具有尺寸不变性进行改进,构造了新的矩不变量.根据提取的矩特征设计了支持向量机分类器,用于对不同类型的目标进行分类识别,利用VC++和Libsvm工具开发了自动识别软件.
Automatic identification of the target in an image,which saves manpower and also enhances the efficiency of information,is an important development direction for the pattern recognition.In this paper,moment invariants are used to represent shape features of the target,and that it don＇t have size invariance in the discrete time is improved by constructing a new moment invariant.According to moment feature vector,this paper designs the support vector machine classifier to identify the classification of different types of targets,and develops automatic identification software by VC＋＋and Libsvm tools.
出处
《应用科技》
CAS
2011年第3期15-19,共5页
Applied Science and Technology
关键词
目标识别
形状特征
矩不变量
支持向量机
target recognition
shape feature
moment invariants
support vector machines
