Preconference Training

Preconference training will be offered on November 17, 2026. These trainings are available at an additional cost to those who are signed up to attend the conference. There are full-day and half-day options available. Details about each training are offered below. If you are interested in attending a workshop, you will need to select it on the form when you register for the conference. Space is limited, so make sure you register soon! 

8:00-5:00 - Cloud-Native GIS with GeoLibre: Building Modern Geospatial Workflows

Dr. Qiusheng Wu - Associate Professor and Director of Graduate Studies | Department of Geography and Sustainability, University of Tennessee, Knoxville

This hands-on workshop introduces participants to GeoLibre, a free and open-source cloud-native GIS platform for visualizing, analyzing, and sharing geospatial data. Participants will learn how to work with modern cloud-native formats and web services directly from a web browser or desktop application without relying on proprietary software. Through practical exercises, attendees will explore interactive mapping, cloud data access, 3D visualization, AI assistant, and reproducible geospatial workflows using open standards.

Hands-on topics:

• Introduction to GeoLibre and the cloud-native GIS ecosystem
• Accessing cloud-hosted geospatial data
• Working with COG, GeoParquet, PMTiles, and Zarr
• Connecting to STAC catalogs and OGC web services
• Visualizing vector, raster, LiDAR, and 3D data
• Extending GeoLibre with plugins and AI-powered tools
• Sharing maps and reproducible geospatial projects

8:00-12:00 - Designing Mobile Data Collection and Inspection Workflows with ArcGIS

Stephanie Baker | ESRI

We will explore how ArcGIS can enhance asset management workflows through mobile data collection and inspection, with a focus on ArcGIS Field Maps and ArcGIS Survey123. Topics will include authoring smart forms and surveys, assigning work using Field Maps Tasks, investigating AI capabilities such as computer vision in Survey123, and exploring options for workflow orchestration and automation. This half-day workshop will be a mix of lecture and hands-on training, with a possible walk about the UNR campus to test what we’ve built! Participants will need to bring existing access to an ArcGIS Online organizational account with at least a Creator user and Publisher role or higher.

1:00-5:00 - Building an Image Selection Workflow for UAS Data: R, sf, Leaflet, and Shiny

Ken Nussear - Associate Professor | Department of Geography | University of Nevada, Reno

A typical drone mission returns hundreds to thousands of photos, but only a subset of those is needed to map the area of interest. Images captured during launch, climb-out, battery changes, descent, and end-of-line turns are often at altitudes and angles that differ from the planned survey height, and images captured while the aircraft is accelerating or moving at high speed can have motion blur. Flights also routinely extend well beyond the actual area of interest while the drone travels to and from the target area. Feeding all collected images into photogrammetry software (e.g., Agisoft Metashape, Pix4D, or OpenDroneMap) costs hours of processing time and often degrades the resulting orthophotomosaics. The usual remedy, scrolling through a folder of thumbnails and deleting unnecessary photos by hand, does not scale and is not reproducible.

This workshop presents a data-driven alternative built on metadata that the camera and aircraft write to every file. We will use the R package exifr (an ExifTool wrapper) to read EXIF and XMP tags from a full image directory into a data frame, then treat that data frame as a flight log: position, timestamp, absolute and relative altitude, or the speed in the x, y, z velocity vector, and camera or gimbal angle and heading. We begin in R, exploring the distributions of these variables to discover what the mission actually did relative to what the flight plan intended. From there we select images using thresholds on parameters such as altitude and speed, convert the selected records to an sf point layer, and subset them spatially against our area of interest, either drawn interactively on a satellite basemap or supplied as an existing KML or shapefile boundary. Selected files are copied to a clean processing directory with a CSV manifest that records exactly which source images produced the output, making the selection auditable and repeatable.

We then apply the same logic in a Shiny application that removes the coding burden for routine use. Participants leave with commented R scripts, the Shiny application, and a sample image set, ready to apply to their own missions. The workshop is aimed at anyone who processes drone imagery and wants a defensible, documented answer to the question of which photographs went into the model.

Learning Objectives

By the end of the workshop, participants will be able to:

1. Extract EXIF and XMP metadata from a directory of UAS imagery into an R data frame, including vendor-specific altitude, velocity, and gimbal attitude tags.
2. Use exploratory plots of altitude, speed, heading, and timestamp to reconstruct mission behavior and identify frames unsuitable for photogrammetric processing.
3. Apply numeric thresholds and spatial intersection against a drawn or imported area-of-interest polygon to produce a defensible image subset.
4. Export the subset with a manifest that documents the selection for reproducibility.
5. Run and modify a Shiny application that performs the entire workflow interactively.