心内膜散乱点云边界点检测算法研究

Efficient boundary points extraction method for scattered point cloud of endocardium

针对心内膜散乱点云预处理中的边界点检测,利用截线云理论将散乱点云进行等间隔区域分层,将点云投影至点云切片,得到切片的散乱点集,同时建立链表结构分区存储点云数据;由平面上点的二维坐标定位,提出区域"十"字算法进行切片数据边界点提取,获取切片数据的最外层点,将检测到的边界点存回原始三维数据源,完成预处理过程。实验结果证明,该算法对边界点具有较强的识别能力,能够在快速、有效地简化点云数据的同时保持原始特征的信息,可以提高后续三维建模的精度和速度。

In order to extract boundary points in the preprocessing for scattered point cloud of endocardium,this paper cut the point cloud based on slicing method to obtain the slice data and stored the slice data with list structure.In 2D space,used approximate nearest neighbor（ANN） to represent the spatial topology relationship of the scattered point cloud and to calculate the k-approximate-nearest neighbours for each data point,then it presented the crisscross-shape algorithm to locate the boundary points,and then mapped the boundary points to the corresponding boundary points in 3D space for the next processing of endocardial three-dimensional mapping system.The experimental results show that this approach has a strong ability for distinguishing boundary points,and can reduce data directly and effectively while keeping the original features,which can increase the accuracy and speed of the subsequent three-dimensional model construction.