VisionShips Q2 2027
SpillGuard AI
A small ceiling-mounted camera that watches shop and warehouse floors.
- Spots spills and debris in seconds
- Detects when someone falls
- Alerts the nearest staff member
From $399 per edge box
PilotReservation not completed. Checkout was cancelled and you have not been charged. You can try again whenever you are ready.
LoadingDock AI
Count, verify and inspect every pallet at the dock automatically
In development · Ships Q3 2027Pilot applications open
From$990per 2-door kit (waived on a 12-month plan)
Software: $249 per dock door per month (2-door minimum)
Pilots are scheduled from Q1 2027, ahead of general availability (Q3 2027). The pilot fee is refundable until installation and credited in full to your first invoice.
Real-time 3D render · drag to rotate · final design may vary
What it is
A rugged box the size of a paperback that links to your loading-dock cameras.
What it does
- Counts pallets in and out
- Reads labels and QR codes
- Spots damaged or torn wrap
- Checks loads against the paperwork
Who it's for
Warehouses, 3PLs, factory shipping docks and air-cargo handlers.
Look inside
What's inside LoadingDock AI
Keep scrolling: it opens up in real time to show the parts that do the work.
Overview
LoadingDock AI is a 15 cm fanless edge box that connects to 2–4 dock cameras and verifies every incoming and outgoing load. It counts and tracks pallets, reads labels and QR codes, detects damaged goods and torn wrap, and checks loads against the manifest — creating photo-evidenced records in real time.
- Pallet detection, counting and tracking through the dock door
- Damage detection — torn wrap, crushed cartons, leaning stacks
- Label, barcode and QR reading with manifest verification
- 2–4 existing or new cameras per box, 256 GB local video
The problem
Mis-shipments, short deliveries and damage disputes cost warehouses, 3PLs and airline cargo handlers money every day. Dock teams count pallets by hand under time pressure, damage is discovered after the carrier has left, and there is rarely visual proof of condition at handover. Claims become arguments without evidence.
Use cases
Inbound verification
Count pallets as they come off the truck and match them to the ASN or purchase order automatically.
Outbound load checks
Confirm the right pallets go on the right trailer before doors close.
Damage evidence
Detect torn stretch-wrap and crushed cartons and capture time-stamped photos at handover.
Label and QR reading
Read SSCC labels, barcodes and QR codes without stopping the forklift.
Airline and airport cargo
Verify ULD and pallet build-ups and capture condition records at cargo acceptance.
Dock utilisation
Measure door dwell time and loading speed to improve scheduling.
How it works
- 1
Watch the dock
Cameras at each dock door stream to the LoadingDock box in real time.
- 2
Count and read
On-device models detect and track pallets and read labels and QR codes as they pass.
- 3
Inspect
Each pallet is checked for wrap damage, crushed cartons and unstable stacking.
- 4
Verify
Counts and IDs are matched to the manifest from your WMS; discrepancies raise an alert before the truck leaves.
- 5
Report
A photo-evidenced load report is generated and sent to your WMS, team and carrier.
In the app
LoadingDock Console
The LoadingDock dashboard shows every door's live status, counts and discrepancies, with photo-evidenced load reports for each shipment.
- Live door status board
- Manifest vs. actual reconciliation
- Damage evidence gallery
- Carrier handover reports (PDF)
- Dock dwell-time analytics
- Searchable shipment history
Door 4 · Inbound · PO-88213
Pallets
22 / 24
Labels read
22 / 22
Damaged
1
Door time
18 min
- Short shipment · Door 409:41
22 of 24 pallets received vs manifest
- Damage · Pallet 1709:36
Torn stretch wrap, photo saved
- Door 2 idle 40 min09:20
Trailer docked with no movement
- Load verified · Door 109:05
Outbound SO-5520, 16 / 16 pallets
AI capabilities
Runs on-device · Hailo-8
Pallet detection and counting
Detects pallets on forklifts and pallet jacks and counts them across a virtual dock line.
Model: YOLO11-m + ByteTrack line-crossing
Damage detection
Detects torn wrap, crushed corners, leaning stacks and leaks.
Model: RT-DETR fine-tuned on dock damage dataset
Label and QR reading
Reads SSCC labels, 1D barcodes, QR codes and printed text.
Model: PaddleOCR (PP-OCRv4) + ZXing decoder
Load verification
Matches detected pallets and IDs to manifest lines and flags shortages or extras.
Model: Rules engine with WMS integration
Anomaly detection
Flags unusual dock events such as unattended open doors or long dwell.
Model: Isolation Forest over dock-event features
Hardware & target specifications
Targets for production units. Subject to change during development.
Hardware
| AI processor | Hailo-8 (26 TOPS) or NVIDIA Jetson Orin Nano 8 GB |
|---|---|
| Camera | 2–4 existing or new IP cameras per box (RTSP / ONVIF) |
| Sensors | Optional dock-door contact sensor |
| Connectivity | Gigabit Ethernet ×2 · Wi-Fi 6 · Optional 4G LTE · GPIO for dock lights and door sensors |
| Power | 10–20 W fanless, 12–24 V DC |
| Enclosure | Fanless aluminium chassis, wall / DIN mount, industrial grade |
| Dimensions | 15 × 15 × 5 cm |
| Weight | ~900 g |
| Ingress rating | IP40 (IP65 enclosure option for cold or dusty docks) |
AI and video
| Processor | Hailo-8 (26 TOPS) / Orin Nano 8 GB |
|---|---|
| Cameras | 2–4 per box |
| Input resolution | Up to 4K |
| Detection | Pallets, damage, labels, QR |
| Throughput | Forklift speed, no stop required (target) |
Storage and data
| Local storage | 256 GB SSD (local video) |
|---|---|
| Evidence | Per-pallet photos and clips |
| Retention | Configurable, default 30 days |
| Processing | On-device |
Connectivity
| Ethernet | 2× Gigabit |
|---|---|
| Wireless | Wi-Fi 6, optional 4G |
| I/O | 4× GPIO |
| APIs | REST, webhooks, SFTP export |
Power and physical
| Power | 10–20 W, fanless |
|---|---|
| Input | 12–24 V DC |
| Size | 15 × 15 × 5 cm |
| Operating temp. | −20 °C to 55 °C |
Built with
The real parts inside
Proven, off-the-shelf components our prototypes are built on. Final production parts may differ.
Raspberry Pi 5
Main computer for our edge boxes
Photo: SimonWaldherr,CC BY 4.0, via Wikimedia Commons · AI overlay added
Raspberry Pi AI HAT+ (Hailo AI accelerator)
Runs the vision models on the device
Photo: SparkFun Electronics,CC BY 2.0, via Wikimedia Commons · AI overlay added
NVIDIA Jetson Orin module
Higher-power option for large sites and robots
Photo: 4300streetcar,CC BY 4.0, via Wikimedia Commons · AI overlay added
Standard CCTV cameras (bullet and dome)
The cameras you already have — we connect to them
Photo: Myotus,CC BY-SA 4.0, via Wikimedia Commons · AI overlay added
Reference photos of commercially available components. IEMI Labs is not affiliated with or endorsed by their manufacturers. Photo credits
In the box
Planned contents
- LoadingDock AI edge box
- Power supply
- Wall / DIN mounting kit
- Dock camera mounting guide
- Quick-start guide
Software & app
The LoadingDock dashboard shows every door's live status, counts and discrepancies, with photo-evidenced load reports for each shipment.
- Live door status board
- Manifest vs. actual reconciliation
- Damage evidence gallery
- Carrier handover reports (PDF)
- Dock dwell-time analytics
- Searchable shipment history
Integrations
- WMS via REST / EDI files (SAP EWM, Manhattan, Blue Yonder, Oracle via API)
- Existing ONVIF / RTSP cameras
- Handheld scanners via export
- Email and Teams / Slack alerts
- SFTP and S3-compatible evidence export
Pricing
Hardware: $990 per 2-door kit (waived on a 12-month plan)
Software: $249 per dock door per month (2-door minimum)
The $990 two-door kit is waived when you choose a 12-month plan. Start with a 30-day single-door paid pilot for $500, credited in full to your first invoice.
Start here
30-day paid pilot · $500
Put one dock door on camera for 30 days and see the evidence you would have had for every claim.
- Edge box and camera kit for one dock door
- Timestamped clip and still for every pallet
- Label and BOL reading with pallet counts
- Per-shipment proof packs for claims and chargebacks
- Credited in full to your first invoice if you continue
Book the pilot
Refundable until installation
Evidence
$249/door/mo
Hands-free proof of every pallet in and out.
- Clip and still per pallet
- Pallet counting
- Label and BOL reading (OCR)
- Per-shipment proof pack PDF
Verify
$349/door/mo
Adds manifest matching against your WMS.
- Everything in Evidence
- WMS manifest matching
- Exception alerts
- Damage flagging (beta)
Enterprise
Custom
For DCs, 3PL networks and cargo terminals.
- Multi-site dashboard
- Custom WMS / TMS integration
- Extended retention
- Priority support and SLA
Indicative launch pricing in USD, excluding taxes and shipping. Final pricing is confirmed before any balance is charged.
Estimated impact
- Manual counting time
- up to 80% less (est.)
- Mis-shipment rate
- up to 50% lower (est.)
- Damage claim disputes
- up to 40% fewer with photo evidence (est.)
Illustrative estimates based on internal modelling and typical deployments; not measured results. Actual outcomes depend on site conditions and usage.
Designed for
- Warehouses and distribution centres
- 3PL providers
- Manufacturers' shipping docks
- Airline and airport cargo handlers
- Cold-chain and food distributors
Roadmap
Target milestones. We will email reservation holders and pilot partners as they change.
Q4 2026
2027 Pilot Program applications open; reference hardware running on customer RTSP cameras
Q1 2027
Pilot deployments at selected sites; model fine-tuning on site data
Q1 2027
EVT / DVT hardware builds, thermal and 24/7 reliability testing
Q3 2027
General availability and first production shipments of LoadingDock AI
Q4 2027
ULD build-up verification for airline cargo and dimensioning add-on planned
2027 Pilot Program
Pilot LoadingDock AI on your site
Pilot partners get early hardware, direct engineering support and preferential launch pricing. Tell us about your site and cameras.
How the pilot program worksThanks, we have it.
A member of the IEMI Labs team will reply within two business days.
LoadingDock AI FAQ
General FAQWhat does the 2027 Pilot Program cost?
Applying is free and carries no obligation. When you are ready, you can book a 30-day single-door paid pilot for $500. We install an edge box and camera kit on one dock door and capture evidence for every pallet for 30 days. The fee is fully refundable until your pilot hardware is installed, and it is credited in full to your first invoice if you continue.
Do I need new cameras?
Not usually. LoadingDock works with existing IP cameras that see the dock door clearly. For best label reading we may recommend one additional camera per door during pilot onboarding.
Is my data processed on the device?
Yes. Video is processed and stored on the box's local SSD. Only counts, discrepancies and selected evidence photos are sent to the dashboard. People in frame can be automatically blurred in evidence images.
Which WMS systems can it connect to?
LoadingDock exposes a REST API and webhooks and can import manifests by file. Connectors for major WMS platforms are being built with pilot partners.
How long does installation take?
Typically half a day per site: mount the box, connect cameras, draw the dock line in the setup wizard, and connect manifests.
Is LoadingDock AI available now?
LoadingDock AI is in development. Specifications shown are target specifications and may change as we complete engineering validation. We are targeting a Q3 2027 ship window and will email reservation holders with every major milestone.
What warranty will it come with?
We are planning a 2-year limited hardware warranty covering manufacturing defects, with advance replacement for pilot and business customers. Final warranty terms will be published before shipping begins.
Related products
VisionShips Q2 2027
ParkSense AI
A palm-sized edge box that plugs into your existing parking cameras.
- Counts free bays in real time
- Flags overstays and wrong parking
- Guides drivers to open spaces
From $499 per edge box (waived on a 12-month plan)
PilotVisionShips Q2 2027
ShelfScan AI
A tiny cube camera that clips onto a shop shelf or ceiling.
- Spots empty and low shelves
- Finds products in the wrong place
- Catches wrong price tags
From $299 per camera
PilotLoadingDock AI
From $990 · Ships Q3 2027