A running pilot on your 634k-image survey format: georeferenced orthomosaic, a Google-Maps-style viewer, and a computer-vision model that already counted and sized … polymetallic nodules across the sample line.
The 30-frame sample you shared, processed end-to-end. These update the moment new imagery lands.
Your sample is a single ~18 m survey line. The full engagement is ~634,000 frames — the same pipeline, scaled horizontally on AWS batch compute.
Every frame carries a .jgw world file, so we know its exact ground position. That single fact makes the hard parts easy.
Pair each JPG with its world file, read geometry into a database that also drives batch scheduling at scale.
Place every pixel at its true UTM position. No fragile feature-matching — the coordinates are known.
Voronoi ownership assigns each seabed patch to one frame, cancelling the 70% overlap so nothing is counted twice.
Warp & blend frames into one seamless north-up image of the seabed.
Slice into 256px web tiles across zoom levels — the browser loads only what's on screen.
Detect, count & size every nodule; pixel scale converts boxes to real centimetres.
Flag megafauna candidates; low-confidence ones route to a scientist, never auto-labelled.
Density, size, taxa, anomalies — plus GeoJSON/CSV for QGIS, ArcGIS and pandas.
Each .jgw pins a frame to the ground, so global structure never drifts over 634k frames — unlike panorama-style feature-stitching, which is slow and fragile on repetitive nodule texture. 69.7% overlap — from the files, not eyeballed.
We QA'd the georeferencing and found a consistent 40 mm camera/nav timing offset. Phase-correlating each overlap snaps frames together: 40.1 mm → 0.6 mm seam misalignment.
Each seabed patch is owned by one frame and counted once; sizes come out in real cm from the known 0.887 mm/px scale. Correcting the offset removed ~3,000 double-counts.
Pan and zoom the georeferenced seabed. Toggle nodule, fauna and footprint layers; click any nodule to measure it.
Size distributions, density by region, dedup proof, the fauna review queue, and one-click scientific exports.
Direct answers to all five of your questions, the 5–8 person team, timeline, pilot price and AWS cost model.