Explore

Week 10 ArcGIS Pro

1. Introduction

This lab introduced end-to-end processing of UAS imagery in Pix4D Mapper, transforming raw photographs gathered during 3D object and mapping missions into analysis-ready spatial products. Because Pix4D implements a Structure-from-Motion (SfM) and Multi-View Stereo workflow, it can appear simple to operate, but accurate outputs depend on proper setup and quality control. In this lab, I focused on verifying image metadata, choosing appropriate processing templates, configuring the Skydio 2 camera model for rolling-shutter correction, and generating deliverables including a densified point cloud and triangle mesh for two missions. Reviewing the Quality Report and documenting project organization were key components in preventing processing and accuracy errors.

⁠

2. Study Area and Source Data

The imagery processed in this lab originated from two prior Skydio 2 missions:

Mission 1: An “accident/structure” scene captured using oblique and lateral angles.

Mission 2: A vertical “light pole” scan focused on a tall, narrow object.

Both flights took place in small, open outdoor environments typical of UAS training areas, with minimal obstructions and consistent daylight conditions. Before processing, I inspected image properties to confirm exposure settings, ISO values, and whether geotags were embedded. This step ensured the datasets were appropriate for SfM reconstruction and guided expectations for model accuracy.

3. Data Management and Project Setup

All processing followed the standard UAS data folder structure:

C:\temp\username_UASmapdata

└── username_MMDDYY_drone_flight

├── 1_Collection

├── 2_Processing

└── 3_Analysis

Images were placed in 1_Collection, and separate subfolders were created in 2_Processing for Mission 1 and Mission 2.

When creating each Pix4D project, I confirmed that the Skydio 2 camera uses an electronic rolling shutter, which can introduce geometric distortion if left uncorrected. I opened the Camera Model editor and set the model type to Linear Rolling Shutter to mitigate distortion.

I selected the 3D Model processing template for both missions because the imagery was captured using oblique, object-focused acquisition rather than a strictly nadir mapping pattern.

⁠

4. Processing Workflow

Pix4D processing followed the standard sequence:

Initial Processing

Identified tie points, calibrated internal camera parameters, aligned images, and produced a Quality Report.

I reviewed the report before proceeding to confirm single-block calibration and acceptable optimization values.

Point Cloud and Mesh

Generated a densified point cloud and triangle mesh.

I assessed point density variation and geometry accuracy, especially for vertical structures.

DSM/Orthomosaic/Index

This stage was not required for this lab and remained disabled.

I also created fly-through animations in the rayCloud to visualize completeness and surface continuity.

5. Results and Deliverables

Both missions reconstructed successfully. The accident-scene dataset produced coherent surfaces and preserved object geometry. The light-pole model captured vertical detail cleanly, demonstrating the importance of oblique coverage.

⁠

6. Discussion and Lessons Learned

This lab reinforced that camera model configuration is critical—especially for rolling-shutter sensors like the Skydio 2. It also demonstrated the importance of reviewing the Quality Report before proceeding, as misalignment or parameter drift early in processing would undermine the final model. For tall vertical features such as poles or towers, consistent oblique coverage mattered more than total image count. Finally, maintaining a clean and standardized data structure simplified reprocessing, troubleshooting, and documentation.

⁠

7. Conclusion

Processing the two Skydio 2 datasets in Pix4D successfully produced densified point clouds, triangle meshes, and animation visualizations. This lab established a foundation for later work incorporating ground control, orthomosaic generation, and coordinate system refinement. Key takeaways included using the Linear Rolling Shutter camera model, inspecting Quality Reports, and adhering to disciplined data organization practices.

Want to print your doc?
This is not the way.

Try clicking the ··· in the right corner or using a keyboard shortcut (

CtrlP

) instead.