GroundTruth · capture guide
How to capture from YOUR tool
GroundTruth needs two things per area: a photo of your markup and the scan session (your radargrams). There are three lanes — pick the one that matches your equipment. Every lane ends at the same place: gtfi.ai/scan.
🐍 CCTV / pipe-inspection cameras
CUES / GraniteNet · Envirosight ROVVER X · IBAK / IKAS · Rausch / RapidView · push & pole cams
CUES / GraniteNet
- Run the crawl normally — GraniteNet stores the video and snapshots per observation.
- Export: Report → Export media — save a defect still (or a short clip) per observation to this device.
- At gtfi.ai/scan: STEP 1 upload the still → STEP 2 attach the clip (or more stills) → Build.
No export handy? Photograph the monitor at each defect — the AI reads it fine.
Envirosight ROVVER X / WinCan
- Export a snapshot or short section clip from the run in WinCan (or screen-record the playback on the tablet).
- Upload the still as STEP 1 and attach the clip as the session.
IBAK / IKAS
- In IKAS, export a defect still or section clip → save to this device.
- Or photograph the control-unit screen at each observation.
Rausch / RapidView
- Export a still from the run → upload. Screen-recording the monitor during playback works too.
Push / pole camera
- Photograph the monitor at each finding, or screen-record the feed with your phone (~30s covers the run).
- Upload the photo as STEP 1 and the recording as the session.
📍 Utility locators & GPR carts
Radiodetection RD8200/RD8100 · Vivax-Metrotech vLoc · Subsite / Ditch Witch · GSSI UtilityScan / MALA carts
Radiodetection RD8200 / RD8100
- Sweep and paint the marks as usual.
- Photo the marked ground (STEP 1) and photo or screen-record the locator display readings during the sweep (STEP 2).
- At gtfi.ai/scan: upload the marks photo → attach the display shots / recording → Build.
The AI cross-checks your paint against the readings — a reading with no paint (or paint with no reading) gets flagged before the dig.
Vivax-Metrotech vLoc
- Photo the marks + the vLoc display, or export from MyLocator / screen-record the sweep and upload the clip.
Subsite / Ditch Witch
- Photo the marked path and the receiver display — or record the sweep on your phone and upload.
GSSI UtilityScan / MALA cart
- Export the scan image from the tablet (Share → Save image) → upload. Or snap the screen with your phone.
🟢 Lane 1 — Tablet-display tools (easiest)
PROCEQ GP8000 / GP8100 / GP8800 (Screening Eagle GPR app on iPad) · any tool driven by a tablet
Proceq / Screening Eagle (iPad)
- Before scanning the area: swipe into Control Center and tap Screen Record (add it once via Settings → Control Center).
- Scan the area normally — every pass and AR view is recorded automatically.
- Mark the concrete. Stop the recording.
- Open gtfi.ai/scan on the same iPad → STEP 1 photograph the marked slab → STEP 2 attach the screen-recording from Photos → Build.
Alternative: export snapshots from Screening Eagle Workspace (Share → Save Image) and attach those as pass photos.
🟡 Lane 2 — Save-your-files tools (no camera needed)
GSSI StructureScan Mini XT · GSSI SIR 4000 (+ Palm) · GSSI Flex NX · Hilti PS 300 · Leica DSX
GSSI (Mini XT / SIR 4000 / Flex NX)
- In the field: save each scan on the unit (it stores every pass as a .DZT file) and photograph each marked area at gtfi.ai/scan (STEP 1 only) → use ➕ Next area between locations.
- End of day: connect the unit's USB to any laptop and copy the project folder.
- Today: attach exported scan images per area (tool: Output/Save Screen → image). Coming next: Data Drop — drag the whole .DZT folder in one go and GroundTruth pairs the passes to your areas automatically.
SIR 4000 also has HDMI-out — a capture dongle can record the live screen if your crew prefers Lane 3.
Hilti PS 300
- Scan and save on the unit; photograph each marked area at gtfi.ai/scan.
- Back at the office: USB the unit into PROFIS Detection, export the scan images (2D views).
- Attach the exported images to each area's record (STEP 2 → attach from file).
Leica DSX
- Scan normally; the DSX tablet app stores results — export the scan/utility-map images from the app.
- Photograph each marked area at gtfi.ai/scan and attach the exported images.
🔴 Lane 3 — Scan-live crews (FieldCam)
Any tool, any age — for crews that read the screen live and don't save files
The FieldCam brim camera
- Clip the micro camera (Insta360 GO 3S class) to your hardhat brim — adhesive Pivot Stand for helmets, Easy Clip for caps. Phone stays in your pocket.
- Press record at the start of each area, press again after you finish marking. The camera sees your screen every time you read it and the slab every time you mark — 2–5 min per area.
- Clips sync to your phone; at gtfi.ai/scan: photo the markup → 🎬 attach the area's clip → Build (or use ➕ Next area as you go).
Battery math: ~20 areas × 3 min ≈ 1 hour of recording/day. One cam + its Action Pod covers a full shift; the Pod recharges the cam between areas.
CCTV / pipe inspection
- Your inspection video already IS the session — upload it directly; GroundTruth extracts the frames.
⚡ Start a scan record →
🛡️ Copilot, not autopilot — GroundTruth's AI is a second set of eyes; the certified technician reviews and signs off. Questions? Your onboarding code came with concierge setup — reach out any time.