一种推扫式光子计数激光雷达点云滤波算法及其验证

发布时间：2018-01-15 07:01

本文关键词：一种推扫式光子计数激光雷达点云滤波算法及其验证　出处：《科学技术与工程》2017年09期 　论文类型：期刊论文

【摘要】：提出一种用于推扫式光子计数激光雷达点云滤波方法。定义了一种线性径向基函数,利用这种基函数可以计算密度点密度值;基于此实现了一种粗滤波和精去噪结合的点云滤波算法,经过两个步骤的滤波之后能够对诸如植被等分层地物进行精细滤波。使用NASA的机载激光雷达MABEL的实际飞行数据对这种滤波算法进行了验证,并与中值滤波算法的效果进行了对比。结果表明算法能有效滤除与地面非常接近的噪声点,适合低密度地物目标点云的滤波处理,能够很好保留细节,其中植被滤波精度90.26%,地面点滤波精度96.11%。
[Abstract]:A point cloud filtering method for push-sweep photon counting lidar is proposed, and a linear radial basis function is defined, by which density density can be calculated. Based on this, a point cloud filtering algorithm combining coarse filtering and fine denoising is implemented. After two steps of filtering, the fine filtering of layered ground objects such as vegetation can be carried out, and the actual flight data of airborne lidar MABEL of NASA are used to verify the filtering algorithm. The results show that the algorithm can effectively filter the noise points which are very close to the ground, which is suitable for the processing of the point cloud of low density ground objects, and can preserve the details well. The accuracy of vegetation filtering is 90.26 and that of ground point filtering is 96.11.
【作者单位】：中国科学院上海技术物理研究所空间主动光电技术与系统实验室;
【基金】：国家“973”计划(2009CB724004)资助
【分类号】：TN958.98;TN713
【正文快照】： 作为一种新型激光三维雷达,光子计数激光雷达的发展时间较短,其特点是回波信号强度低,噪声大于激光回波,光电探测系统的响应率处于少光子水平[1]。系统设计中超窄带滤波,但是由于探测系统灵敏度极高,目标背景环境产生的大量噪声仍会对探测电路形成触发,产生大量噪声数据。激光

【参考文献】