evidoria

← Back to browse

Good practice Imported

Perivallon — AI and Satellite Imagery Detect Illegal Waste Dumps With Over 90% Precision Across Lombardy, Italy

Italy · Milan · See the Italy profile · See the Milan profile

Evidence: Observational / pre–post Top 24% 73/100 · Ask Evidence Copilot about this practice

Politecnico di Milano's Perivallon project uses neural networks on satellite imagery to flag illegal waste dumps with over 90% accuracy, piloted across 100+ Lombardy municipalities with ARPA Lombardia and Italy's Carabinieri Ecological Unit.

>90%
Illegal dump detection accuracy
100+
Municipalities covered in field testing
6,500+
AerialWaste dataset downloads
Perivallon — AI and Satellite Imagery Detect Illegal Waste Dumps With Over 90% Precision Across Lombardy, Italy

Details

Maturity
Pilot
Promoter
Politecnico di Milano / ARPA Lombardia / Carabinieri Ecological Unit (Perivallon project, Horizon Europe)
Period
2024-2025
Region (NUTS)
ITC4
Keywords
environmental crime, waste management, law enforcement, satellite AI

Context

Illegal waste dumping and hazardous-material storage are hard to detect in time for enforcement, a problem widely discussed in Italy around the 'Terra dei Fuochi' waste-burning crisis.

Activities

Perivallon, a Horizon Europe-funded project led technically by Politecnico di Milano, applies neural networks to optical and radar satellite imagery to detect illegal waste dumps, piloted with ARPA Lombardia and the Carabinieri's Ecological Unit across more than 100 Lombardy municipalities.

Results

Field testing found the system identifies illegal dump sites with over 90% accuracy, building on the existing AerialWaste satellite-imagery dataset (downloaded 6,500+ times by researchers). The consortium spans 24 partners across 12 countries.

Conclusions

As of August 2025 Perivallon remains at pilot/demonstration stage in Italy; its 90%+ accuracy reflects controlled field trials, and scale-up to routine enforcement use has not been independently confirmed.

Implementation
Indicative cost
High (€500k–€5M)
Time to results
Medium (1–3 years)
Staffing & skills
Politecnico di Milano, ARPA Lombardia, Carabinieri Ecological Unit, 24-partner Horizon Europe consortium

Conditions for success

  • optical and radar satellite imagery
  • existing labelled waste-imagery dataset (AerialWaste)

Common failure modes

  • still pilot/demonstration stage, not embedded in routine enforcement
  • accuracy figure not independently audited

Commonly funded by

Horizon Europe National / regional programmes

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful