基于回波强度和采样点距离的点云滤波研究

本文根据三维激光扫描的特点,设计提出一种从点云数据中分离出目标对象的特征提取方法。首先根据激光回波信号特征和采样点距离特征进行阈值确定,然后通过将回波信号强度和距离信号位于各自阈值范围之外的对象滤除,从而得到滤波后的目标物体点云数据集。应用上述方法对室外建筑物数据进行试验,得到较好的效果,表明该方法适用于部分目标对象的初步分离。

Study on points cloud filtering based on reflectance intensity and range

A method of feature extraction that separates target objects from points cloud was presented in this paper according to the characteristics of three-dimensional laser scanning. The threshold values were firstly determined in terms of the laser reflectance and the range of sample points. The objects beyond each threshold value mentioned above were then filtered to get the best estimates of the target objects. The outdoor experiments on buildings by using the above method show that it is useful to separate the target objects primarily.