Workflow Narrative

UAV Image Intake QA

Problem It Solves

Is this image set good enough to proceed into expensive downstream processing, and what should be fixed first if not?

Who It Is For

UAV mission operators, photogrammetry technicians, and geospatial production teams.

Primary User

Drone services teams, geomatics operations managers, and data-engineering teams responsible for production intake quality.

What It Does

Scans UAV imagery and builds a metadata/coverage inventory.
Computes intake QA metrics (GPS/EXIF completeness, overlap estimate, image-count sufficiency).
Extracts per-image blur scores and flags blurry images with warnings.
Parses RTK fix status and gimbal/flight orientation priors from DJI XMP metadata where available.
Returns workflow status (`pass`, `review`, `fail`) with warnings suitable for operator triage.

How It Works

Recursively discovers supported image files (`jpg`, `jpeg`, `tif`, `tiff`, `png`).
Parses EXIF fields for timestamp and GPS coordinates; parses DJI XMP sidecar fields for RTK fix status and gimbal yaw/pitch/roll and flight yaw.
Estimates overlap from GPS nearest-neighbor spacing against a nominal footprint heuristic.
Computes a per-image blur score from a Laplacian variance kernel (configurable as `off`, `fast`, or `full` mode) with wide-SIMD acceleration.
Metadata extraction and blur scoring run in parallel across all images using Rayon thread pools.
Applies profile thresholds to classify readiness and emit actionable warnings.
Indicative heuristic: overlap_proxy ~= 1 - (nn_spacing_m / nominal_footprint_m), clipped to [0, 1].

Why It Wins

Converts ad hoc manual preflight checks into a single reproducible QA workflow with machine-readable outputs, including blur quality, RTK status, and orientation priors alongside the standard GPS and overlap diagnostics.

Typical Buying Trigger

Frequent failed/rework-heavy photogrammetry or registration runs caused by avoidable intake quality issues.

Typical Presets

fast: permissive intake thresholds with fast-mode blur scoring for rapid field feedback.
balanced: default operations profile with fast-mode blur scoring.
strict: tighter quality gate with full-mode blur scoring before production processing.

Inputs

Parameter	Optional	Description
images_dir	no	Input directory containing UAV images to screen.
profile: fast \| balanced \| strict	yes	[pro] QA profile controlling overlap, metadata readiness, and blur thresholds; drives pass/review/fail classification. Defaults to `balanced`.
recursive	yes	If true, scans subdirectories under images_dir. Defaults to `true`.
output_prefix	yes	Prefix used to name all output artifacts.
blur_mode: off \| fast \| full	yes	Blur scoring mode. `off` skips blur scoring; `fast` downsamples before scoring; `full` scores at original resolution. Defaults to `fast`.

Outputs

Parameter	Type	Description
image_inventory	CSV	Per-image inventory with metadata flags, GPS coordinates, capture fields, blur score, and gimbal/RTK orientation columns (`*_image_inventory.csv`).
qa_report	JSON	Structured QA checks and warning details for intake triage, including blur and orientation hint sections (`*_intake_qa_report.json`).
summary	JSON	Workflow summary contract with status and aggregate metrics including blur coverage and RTK coverage fractions (`*_intake_summary.json`).
html_report	HTML	Human-readable customer-facing report generated from the summary contract for stakeholder review and QA traceability.
image_centers	GeoJSON	Image center points from EXIF GPS with per-feature blur score and orientation properties (`*_image_centers.geojson`).
flight_path_lines	GeoJSON	Flight path line geometry built from ordered image centers with GPS (`*_flight_path_lines.geojson`).

Python Example

import whitebox_workflows as wbw

wbe = wbw.WbEnvironment(include_pro=True, tier="pro")

result = wbe.guided_uav_image_intake_workflow(
    images_dir="data/uav_mission/images",
    profile="balanced",
    recursive=True,
    output_prefix="output/uav_intake",
    blur_mode="fast",
)

print(result)

License Notice

Use of this function requires a license for Whitebox Workflows Professional (WbW-Pro). Please visit www.whiteboxgeo.com to purchase a license.

Project Links

WbW Homepage User Manual Learn More