Join us for a brief demonstration of the new smoothing tool we have added to LP360 (Advanced)/LP360 for sUAS. In this demonstration we will show several examples of the smoothing using data from UAV LIDAR systems. Pros and cons of smoothing vs. thinning and using classified or unclassified points will also be discussed.
Working with automotive class LIDARs – the term we use for the smaller, lighter, less expensive LIDARs being developed for autonomous vehicles – has taught us a lot about how to get the most out of these cost-effective, but noisy, LIDARs. We use the Quanergy M8 Ultra on our True View 410 3D imaging sensors. It produces very accurate data for its class, less than 5 cm network accuracy, even beneath canopy, but like all lidars in this class has significantly more shot noise than traditional survey-grade LIDARs. To an experienced airborne or mobile lidar data user, the point cloud from such UAV LIDAR systems looks “fuzzy”. To address this we have added a new Point Cloud Task (PCT) to effectively smooth noisy point cloud data.
The algorithm reduces the shot-to-shot noise on hard surfaces – reduces the fuzziness – while maintaining the overall accuracy and preserving fine details on above ground features and structures. This significantly improves the overall quality of point cloud data sets from UAV LIDARs such as Quanergy and Velodyne.
The new tool will be available in the next release of LP360 (Advanced) and LP360 for sUAS and is already implemented for True View EVO users.