局部相似目标匹配的微分几何模型

根据微分几何原理,基于平面曲线作刚体变换时其曲率的不变性,利用曲率来表达目标轮廓的内在特征。筛选出待匹配目标轮廓上内在特征相似的点,形成点对集合。在点对集合中寻找相似线段来定位可能的平面变换,通过得分函数,求出点对集合中相似线段平面变换的最佳值,得出最佳匹配。仿真实验表明,该模型适合局部相似情况下的目标匹配,特别对于复杂形状目标,运算复杂度较低,具有较好的识别效果。

Differential Geometry Model to Partially Similar Object Matching

According to differential geometry,considering the invariance of curvature when rigid transform was applied to plane curve,a new model utilizing curvature to illustrate the inherent characteristic of object contours was proposed;point-pair set was constructed by means of filtrating points with similar inherent characteristic in object contours;possible transform was located by similar straight line segments pair;finally,optimum transform was illustrated and optimum matching was determined by score function. Simulation experiments indicate an encouraging matching efficiency and low run time complexity of the algorithm for partially similar object matching,especially for complex shape.