| 摘要: |
| 为了探索有效的图形分类与识别的新方法,引进一类正交完备的分段k次多项式系统(简称U-系统).U-系统是一类属于L2[0,1]的正交完备分段k次多项式系统.该系统下的U级数展开式具有良好的平方逼近及一致逼近性质.基于U-系统理论,提出了U描述子的概念,给出了U描述子的性质并在理论上予以证明.为了更好地对图形分类与识别,对U描述子进行了归一化,同时在理论上证明了归一化U描述子具有旋转、平移、尺度大小等不变的性质.实验表明,归一化的U描述子能够高效、准确地对图形进行分类与识别,与Fourier描述子相比,具有更好的识 别率. |
| 关键词: Fourier描述子 U描述子 归一化 图形分类 识别 |
| DOI： |
| 分类号: |
| 基金项目:Supported by the National Natural Science Foundation of China under Grant No.60133020 (国家自然科学基金); the National Grand Fundamental Research 973 Program of China under Grant No.2004CB318000 (国家重点基础研究发展规划(973)); the Guangdong Provincial Science-Technology Project of Guangdong Province of China under Grant No.2004A10302005 (广东省科技攻关项目); the Guangzhou City Science-Technology Project of China under Grant No.2004A10302005 (广州市科技攻关项目) |
|
| A Classification and Recognition Method for Planar Figures Based on Complete Orthogonal U-System |
|
CAI Zhan-Chuan,SUN Wei,QI Dong-Xu
|
| Abstract: |
| In order to probe into classification and recognition method for planar figures in effect, a class of complete orthogonal piecewise k-degree polynomials, so-called U-system, is introduced. U-system is a class of complete orthogonal piecewise k-degree polynomials in L2[0,1]. The expansion in U-series has advantageous properties for approximations in both quadratic norm and uniform. Based on U-system, new U descriptors are defined, and the property of U descriptors and its proof are presented. To classify and recognize figures, a new normalized U descriptor is defined, which is invariant in the rotation, translation and scale transform. The experimental results show the accuracy and efficiency of new normalized U descriptors in classification and recognition method for planar figures. |
| Key words: Fourier descriptor U descriptor normalization figure classification recognition |
