DJI-2 + Edits

The transcription provides instructions on how to process data from the Zenmuse L2 LiDAR system using the DGI Terra software. It covers topics such as importing LiDAR data, setting parameters for point cloud processing, ground point classification, generating DEMs, advanced settings for accuracy checks, and calibration of the Zenmuse L2. It also mentions the option to view and save reports, and suggests troubleshooting steps if calibration fails. The aircraft will autonomously return to the last waypoint and resume the mission. If the L2 shows abnormalities during use, run the DGI Pilot 2 app and enter the Health Management System. Data Processing Welcome to the DGI tutorial video. In this video, we will explain how to process the data from Zenmuse L2. Data Import Launch DGI Terra and choose New Mission. Then select LiDAR Point Cloud and name the mission. Processing LiDAR Point Clouds uses a substantial amount of computer resources, so ensure your computer configuration can smoothly complete the LiDAR Point Cloud processing. Click the folder icon and import the LiDAR Point Cloud data collected by L2. If you need to use the DRTK2 mobile station or a third-party base station for PPK processing, you should determine the time period of the task operation based on the Point Cloud data files obtained from the L2. Search for the base station data from the same time period within the files stored by the DRTK2 mobile station or the third-party RTK base station. If you use a DRTK2 mobile station, you can directly copy that day's base station data files to the DGI Terra. DGI Terra will automatically recognize them. After the import is complete, if you did not set the base station center coordinates during the operation or need to replace the original self-converging coordinates, you can modify the base station center coordinates in the base station center point settings. Here's how to do it. Basic Parameters Point cloud density can be selected by percentage or by distance. The amount of point clouds involved in processing can be adjusted by setting. If quick processing is needed or computer capability is insufficient, the percentage can be reduced or the distance can be increased. Choose Point Cloud Processing under Scenarios. Point Cloud Processing Under the Point Cloud Processing menu, Point Cloud Effective Distance refers to the distance between the effective point cloud data used for processing and the L2 LiDAR. If the points collected by the LiDAR exceed this effective distance, those points will be filtered out and will not participate in the processing. This setting is practical when reconstructing a close target but inevitably collects the distance background area. Optimize Point Cloud Accuracy When enabled, DGI Terra will adjust and optimize the point cloud data scanned at different times enhancing the overall accuracy. It is recommended to turn on this feature. Smooth Point Cloud When enabled, this will reduce the thickness of the point cloud, remove isolated noise, make local structures appear more apparent, and improve the effect of ground point extraction. It is recommended to enable this feature. Ground Point Type When enabled, it can classify ground and non-ground points. Ground points can be used for creating the topography of the survey area. Choose a ground type based on the actual situation. Flat ground is suitable for areas with dense buildings or plains. Gentle slope applies to common mountainous and hilly areas. A steep slope is ideal for areas with dramatic elevation changes such as high mountains and river valleys. The algorithm of DGI Terra's ground point classification selects seed point units for ground points evenly within the survey area based on the set building dimensions. These seed points are then used to construct an initial triangular mesh. To identify ground points among other laser points, DGI Terra calculates their angles and distances relative to the vertices and plane of the respective triangle. Points with angles and distances below preset values are classified as ground points, updating the triangular mesh. This process repeats until all ground points are identified. Building max diagonal should be set based on the diagonal length of the largest rooftop plane in the survey area. This rooftop may be recognized as ground if the value is set too small. DGI Terra has preset iteration angle and iteration distances according to different ground types, such as flat ground, gentle slope, and steep slope, making it suitable for most common scenarios. These two parameters can be appropriately increased if the terrain has more variations. Generate DEM. When enabled, a DEM will be generated in TIF format based on the ground point cloud data. By scale, you can select 1 to 500, 1 to 1,000, and 1 to 2,000. Alternatively, the DEM resolution can be set by GSD. Advanced Settings. In the Advanced Settings menu, the Accuracy Check feature has been added. Click the Import button to bring in checkpoints or control points. Note that the coordinate system should be geodetic and consistent with the point cloud coordinate system. After importing, select either Checkpoints or Control Points and click Process Now. Control points can be used to optimize the elevation accuracy of the point cloud, while checkpoints can be used to assess the quality of the point cloud's accuracy. Set the output settings according to your requirements, including the coordinate system and point cloud format. If you choose to output point cloud files in LAS, PLY, or PCD formats, the Merged Output option can be selected to merge multiple result files into one. Click Start Processing and verify each setting in the pop-up reconstruction parameter checklist. After reconstruction, processed results can be moved, scaled, and rotated. At the bottom of the interface, you can choose five different display modes to view the results, including RGB, Reflectivity, Height, Return, and Type. Click the profile icon to access the cross-section feature. Through the diagram, geographical features such as vegetation and ground can be quickly identified. Choose the cross-section line on the model by clicking the left button to start and the right button to finish. You can set the width in the menu. The default is 5 meters. The shortcut Control-Alt-F can be used on the project page to open the folder of the current mission. You can view and save reports in HTML format. The reports include original data-related information, software parameters, output formats, and point cloud processing time. Users can refer to the document, How to Read a DGI Terra Quality Report, on the DGI website for detailed information on the report. LiDAR Self-Calibration Welcome to the DGI tutorial video. This video will explain the self-calibration method of the XenMUSE L2. It is necessary to perform LiDAR self-calibration. Data Collection Use area route, choose a survey area with buildings, select LiDAR mapping. Single return, 240 kHz, repetitive scanning, and turn on RGB coloring. Use the ortho collection to plan a route for more than 5 minutes with altitude set to relative to takeoff point, Alt, and 100 meters. Enable elevation optimization and IMU calibration. Set 10 meters per second for speed in advanced settings, and side overlap for LiDAR should be 50%. After planning, execute the mission and collect the calibration data. Data Generation with DGI Terra Use the latest version of DGI Terra, version 3.9.0 or above. Create a new mission of LiDAR point cloud. Import the collected calibration data and use calibration for scenario selection. After processing, click to export the calibration files. The generated calibration file is a TAR file under the project folder terra-lidar-cali. Calibration Copy the calibration file to the root directory of the microSD card. Mount the L2 onto the M350RTK. Insert the microSD card into the Zenmuse L2 requiring calibration and power of the aircraft. Wait approximately 5 minutes to complete the calibration. Verification After calibration, remove the microSD card from the L2, connect it to a computer, and open the text log file. If it displays All Succeed, it indicates the calibration was successful. You can also record a new set of point cloud data and check whether the time parameters in the CLI file have been updated. If calibration fails multiple times, it is recommended to reset to factory settings and recalibrate. After reconstruction, it is recommended to check whether the point cloud has a layering issue and whether the coloring has overlaps. If there are such issues, it will be necessary to recollect the calibration data.