为了准确地在三维网格模型上定位特征角点,提出了一种基于变形分析的三维Susan角点检测算法。算法首先利用邻接区域信息定义顶点的变形函数,由变形函数值得到候选角点集合;对于候选角点,设定比较区域,在区域内用相似比较函数确定角点;最后,在一定邻接区域内使用非极大值抑制去除虚假角点。使用特定模拟数据和真实三维人脸模型进行角点检测,实验结果验证了该算法的优越性和有效性。
In order to locate the corner vertices on 3D mesh precisely,the paper proposed a novel method by extending the Susan operator to the 3D scanning data based on deformation analyzing.The algorithm first selected the candidate corners by deformation value which determined by multi-stage adjacent information.Then,to each candidate corners,calculated the value of comparison function based on predefined regions.Finally applied non-maximum suppression within the specified range for removing false point.Experiment results in special simulant data and real 3D models show the algorithm's validity and superiority.