Drone Surveying Training

What are the Best Practices for Processing Drone LiDAR Data

mdInfinity Best Practices Video Series

mdInfinity is a complete software suite for your aerial data processing of LiDAR and photogrammetry. In this series of videos, Robert Chrismon, PLS, reviews some of the best practices for processing data that has been collected in the field using drone LiDAR from the Microdrones mdLiDAR1000HR. These videos can assist you with a step-by-step guide to using some of the data processing modules in mdInfinity. However, please contact support for specific questions.

mdInfinity is available in Desktop and Cloud versions, depending on your needs and technical requirements. Each version is similar in functionalities and capabilities, with its own advantages. Custom services are available when you want to leave the processing up to the mdInfinity team, who is always available for help. In our first mdInfinity best Practices video, we review Trajectory Processing, which is the first step in the mdInfinity workflow.

Trajectory Processing

Trajectory Processing is the first step in the mdInfinity workflow where users can validate the trajectory parameters (with or without a base station) before processing and obtain different types of EO files depending on the software they will use afterwards. Trajectory Processing is only available on mdInfinity Cloud.

The Trajectory Processing module is a user friendly workflow for users to process raw GNSS data (with or without base station), hybridize IMU and GNSS data to produce a smooth trajectory file and export it as a EO/sbet files in the desired coordinate system. Users can validate the trajectory parameters before processing and obtain different types of EO files depending on the software they will use afterwards.


After using Trajectory Processing, Georeferencing will anchor geographic coordinates to every point of your point cloud. The Georeferencing module is available in both mdInfinity cloud and desktop to transform the raw data from the LiDAR (range and bearing angles), the GNSS and IMU (orientation and positioning) to 3D point cloud and associated data. With the specifications of the raw data, mdInfinity produces a georeferenced point cloud without any geodetic distortion. You can also select your coordinator reference system that matches the area flown with the drone surveying equipment.

Review Data

After creating the georeferenced .LAS files, it’s a good practice to review the exported data and perform a preliminary assessment of the point cloud. Reviewing the data will help you see any gross errors or determine whether further refinement and processing may be needed.

Precision Enhancement

Precision Enhancement is available on both mdInfinity cloud and Desktop to remove outliers and reduce the noise level of your point clouds. This module is designed to reduce the amplitude of the random errors affecting a LiDAR point cloud and not to eliminate discrepancies due to remaining systematic errors (e.g. poles, cables.) Precision enhancement solves the problem of outlier rejection and denoising in a unified environment. The mdInfinity denoising module reduces the noise level while preserving edges and irregular features in the point cloud. The classification methods will give better results when used on a denoised point cloud, because the precision enhancement improves the visibility of the geometric features of a point cloud.

Reviewing Precision Enhancement Data

At this step, it’s a good idea to compare and review the georeferenced point cloud with the precision enhanced point cloud to compare strips and make sure the enhanced noise levels are satisfactory.

Ground Classification

mdInfinity ground classification is a powerful way to segment bare-Earth points from objects and vegetation in 3D point clouds collected by LiDAR or photogrammetry. This data processing module is available in both mdInfinity desktop and cloud.

The mdInfinity Ground classification tool is based on a combination of several algorithms, including like frontier point analysis, segmentation, TIN densification. As such, it takes benefits of each of these methods and is able to classify ground points in very challenging environments. In particular, steep slopes cliffs are correctly classified which enhance the resolution of ground Digital Terrain Models in proving more true positive ground points than other approaches. It can be customized for different data acquisition systems while proving optimum results in terms of classification rate.

Review Ground Classification

The final step is to review the classified ground and compare the control with other points in the data set for accuracy. It is highly recommended to preprocess the dataset in Strip Adjustment and Precision Enhancement to improve the visibility of the geometric features, before performing Ground Classification. Upon visual inspection of the point cloud and checkpoints, you can determine if the data collected meets the acceptable parameters and move into the deliverable phase for completing the project in other third party software.

For more information on using drone surveying equipment or processing your LiDAR and photogrammetry, please speak with one of our helpful representatives.