AI-Powered Snow Logistics

Intelligent Snow Volume & Weight Estimation

A camera-based AI system that automatically counts snow transport trucks, identifies them via license plate recognition, retrieves vehicle dimensions from a registry database, and estimates both the volume and weight of transported snow — enabling accurate billing and operational analytics.

How the System Works

End-to-end automated pipeline from camera feed to billing data

1
Video Capture

Cameras positioned at the snow dumping polygon capture arriving trucks in real-time.

2
Truck Detection

YOLO v12 deep learning model identifies and localizes trucks in every frame with high confidence.

3
Plate Recognition

ALPR extracts license plate text from the detected truck region with OCR confidence scoring.

4
Database Lookup

The plate is matched against the vehicle registry to retrieve truck dimensions and cargo bed volume.

5
Snow Estimation

Volume × SWE density = estimated snow weight. Results are logged for billing & analytics.

System Architecture

Modular design for flexible deployment

Input Layer
IP Camera Feed RTSP / Video File
SWE Data Feed FMI / SYKE API
AI Processing Core
Truck Detector YOLOv12 Medium
Plate Reader fast-alpr OCR
Volume Estimator DB dims + CV fill
Weight Calculator Volume × SWE
Output Layer
Vehicle Registry SQLite / Traficom
Billing Records Per-truck events
Analytics Dashboard Reports & CSV

Key Technologies

Built on proven, state-of-the-art components

YOLOv12 Object Detection

Real-time truck detection using the latest YOLO architecture. The medium model balances accuracy and speed for production deployments. Trained on COCO dataset with truck (class 7) specialization.

mAP: 52.5% Speed: ~18ms
Automatic License Plate Recognition

Fast-ALPR provides both plate localization and OCR in a single pipeline. Confidence scoring enables multi-frame confirmation, reducing false reads and ensuring reliable truck identification.

Multi-frame verify Finnish plates
SWE-Based Weight Estimation

Snow Water Equivalent (SWE) data from meteorological sources, combined with known truck cargo volume, enables weight estimation without physical scales. SWE varies by snow type and transport conditions.

100–500 kg/m³ Date-adjusted

Snow Weight Estimation Methodology

From truck dimensions to billing-ready weight data

Volume Determination

The cargo bed dimensions (Length × Width × Height) are retrieved from the vehicle registry database after license plate identification. For known trucks, the maximum cargo volume is a fixed, reliable value.

Vcargo = L × W × H
Example: 6.50 × 2.55 × 1.20 = 19.89 m³
Fill Level Estimation

Computer vision can estimate the fill percentage of the truck bed by analyzing the snow surface relative to the truck walls. For the current prototype, a default 100% fill is assumed.

Vsnow = Vcargo × fill%
Future: CV-based fill level detection
Weight Calculation via SWE

Snow Water Equivalent provides the density of the transported snow. Combined with volume, it yields the estimated weight.

Wsnow = Vsnow × SWE (kg/m³)
Example: 19.89 × 350 = 6,962 kg ≈ 7.0 tonnes
SWE Reference Values
Snow Type SWE (kg/m³) Context
Fresh powder50 – 100Newly fallen, dry
Settled snow100 – 200Days-old, compacting
Wind-packed200 – 350Wind-driven compaction
Transported300 – 450Loaded by machinery
Wet / Spring350 – 500Melting, saturated
Glacial / Ice500 – 917Refrozen, dense
For accurate results, SWE should be sourced from the Finnish Meteorological Institute (FMI) or Finnish Environment Institute (SYKE) and adjusted for local conditions and date.

Development Roadmap

Phased approach to a production-ready system

Phase 1 – Complete
Truck Detection & License Plate Recognition

YOLOv12-based truck detection, ALPR integration, multi-frame confirmation logic, web-based processing interface with real-time video feed.

Phase 2 – Complete
Vehicle Registry & Database Integration

SQLite vehicle registry with truck dimensions, automatic plate-to-vehicle matching, snow event logging, enriched results with vehicle specs and volume data.

Phase 3 – In Progress
SWE Integration & Weight Estimation

Integration with FMI/SYKE API for real-time SWE data. Date-aware snow density calculation. CV-based fill level estimation for more accurate volume readings.

Phase 4 – Planned
Traficom Vehicle Data Integration

Connect to Traficom's open vehicle registry API for automatic retrieval of truck specifications by license plate. Requires GDPR-compliant data access agreement.

Phase 5 – Planned
Production Deployment & Billing

Live RTSP camera feed processing, real-time dashboard, automated billing report generation, multi-site support, and API access for third-party integration.

Requirements for Production Deployment

What is needed to take the system from prototype to production

Camera Infrastructure
  • IP cameras at snow dumping polygon entry points
  • Adequate resolution for plate reading (1080p+)
  • Proper mounting angles for truck identification
  • City permits for camera placement on public land
Client responsibility
Data Access & GDPR
  • Traficom data request for vehicle dimensions
  • GDPR compliance framework for plate data
  • Data processing agreement (DPA)
  • Retention policy for detection logs
Client responsibility
SWE Data Source
  • FMI or SYKE API access for snow density data
  • Location-specific SWE measurements
  • Historical data for calibration
  • Real-time feed for daily estimates
Open data available