Built for tolling & transportation

Neural Sentinel

ALPR and vehicle attribute extraction with a confidence-gated review queue. Your operators only see the plates the model isn't sure about — everything else auto-approves.

Neural Sentinel batch upload dashboard showing 8 vehicles with plate numbers, states, make/model, color, and confidence scores.

Four attributes, one pass

Upload an image or a batch of video frames. Neural Sentinel returns plate number, state, make/model/year, and color — each with a confidence score.

Plate Number
SNF8881
Confidence 93.6%

Trained on tolling imagery — angled plates, reflections, partial occlusion, mixed U.S. formats.

State
TEXAS
Confidence 99.6%

State inferred from plate artwork and layout. 51 U.S. jurisdictions supported, including specialty issues.

Make / Model / Year
Ford Ranger 2013
Visual classification

Trim-level identification when possible. Reviewers can override the model if needed.

Color
Black
Per-vehicle palette

Normalized color field for filtering and search — not affected by plate enhancement filters.

A review queue, not a dumping ground

High-confidence detections auto-approve. Low-confidence detections land in a queue your team can burn through with the keyboard.

Confidence-gated auto-approval

Each vehicle is scored on plate, state, and vehicle attributes. Anything above your threshold is auto-approved and exported. Anything below routes to the queue — usually 15–25% of traffic.

  • Per-reviewer stats: total, auto-accepted, edited, rejected, escalated
  • Average review time tracked per operator
  • Supervisor view for claim history and escalations
  • Approval rate and queue depth on every refresh
Neural Sentinel review queue showing pending review counts, per-reviewer stats, keyboard shortcuts, and two vehicles awaiting review.

Enhance stubborn plates in seconds

When the plate is dark, glared, or low-contrast, your reviewer doesn't need Photoshop — the enhancements panel is built in.

Presets first, sliders when you need them

One-click presets — Auto, Hi-Contrast, Night, Deblur, Glare, Grayscale — cover the common cases. Fine-tune brightness, contrast, sharpness, saturation, exposure or gamma when a plate needs it. Enhancements apply only to the plate crop, so the vehicle shot stays true.

  • Keyboard shortcuts for navigate, approve, reject, escalate
  • Save & Approve commits extracted attributes in one action
  • Manual override for any field with a non-obvious read
  • Zoomable plate and vehicle crops
Neural Sentinel review modal with license plate, enhancement presets and sliders, and vehicle details form.

Every detail, one keystroke away

Every reviewed vehicle keeps the full record: the vehicle crop, the plate crop, extracted plate number and state with confidence, make/model/year, color, and detection metadata — frame index, timestamp, bounding box. Previous / Next moves you through the batch without a single mouse click.

  • Edit any field; the audit trail preserves the original model output
  • Detection metadata for timeline reconstruction and dispute handling
  • Per-batch export: JSON, CSV, PDF
  • REST API and webhooks for tolling, CAD/RMS, and evidence pipelines
Neural Sentinel vehicle details panel showing vehicle crop, plate crop, plate number SNF8881, state Texas, make/model Ford Ranger 2013, color Black, and detection metadata.

Under the hood

The pipeline, the people tiers, and what you get as a buyer.

Detection pipeline

  • YOLO vehicle detection and cropping
  • PaliGemma VLM for plate, state, make/model, color
  • State-inference fallback when the crop is low-quality
  • Batch or streaming (WebSocket) input
  • Single forward pass — no OCR + classifier chain to maintain

People tiers

  • Reviewer — confidence-gated queue, keyboard-first modal
  • Supervisor — claim history, operator identity, time-per-item
  • Org admin — users, thresholds, exports
  • Platform admin — tenanting, roles, audit
  • Every edit audited against the original model output

Training & data

  • Fine-tuned on real tolling imagery, not web-scraped COCO
  • Angled plates, reflections, partial occlusion covered
  • 51 U.S. jurisdictions, including specialty and vanity plates
  • PostgreSQL-backed with versioned Alembic migrations
  • Your review edits can feed the next training cycle
Inputs MP4, AVI, MOV, JPG, PNG · up to 500 MB
Exports JSON, CSV, PDF · REST + webhooks
Deploy SaaS, private cloud, on-prem
Isolation Per-tenant data, RBAC, full audit log

Bring your review workload down

Pilot Neural Sentinel on a day's worth of your tolling imagery and measure how much of it auto-approves. Our team handles the integration.

AI Assistant
Online
Hi! I'm your AI assistant. How can I help you learn more about Gravity Digital Solutions' products?