【作者】 刘旭丽； 邹建成； 张波； 铁小匀；

【Author】 Liu Xuli Zou Jiancheng Zhang Bo Tie Xiaoyun (Institute of Image Processing and Pattern Recognition,North China Univ.of Tech.,Beijing 100041)

【机构】 北方工业大学图像处理与模式识别研究所；

【摘要】 本文从基于对称集的形状描述子出发,将形状主方向的思想用于此种形状描述子。首先该描述子是二值对角对称图表,此图表具有图形上成对点的全局对称特性;其次形状主方向消除了边界起始点的影响。在所定义的矩阵内积基础上,计算两图形的匹配值。实验表明由于形状主方向的引入,进一步缩短了匹配值的计算时间,并通过几种类别的图形测试了这种描述子的描述能力。更多还原

【Abstract】 A novel shape descriptor and shape orientation are introduced in this paper.Based on this descriptor,we take the orientation to improve the descriptive ability.First,the descriptor is visualized as a diagonally symmetric diagram with binary valued regions.This diagram is a fingerprint of global symmetry between pairs of points along the shape.Then,we remove the effect of the start point of the border through a shape’s main orientation.From the defined inner product of matrix,we calculate matching value of two shapes.The descriptive ability is tested by many classes of shapes,which shows that the matching time can be saved through shape orientation.更多还原