Land Cover Classification – Cyprus (10m Satellite Imagery)

Location: Cyprus (Island nation in the Eastern Mediterranean)

Project Overview

Nazru’s AI platform was deployed to classify land cover across Cyprus using 10‑metre resolution satellite imagery. The project mapped the island into nine distinct classes: bare_soil, cultivated_crops, forest, industrial, transportation, urban_built, water, wetland, and wild_grass. The output supports environmental monitoring, agricultural planning, and sustainable development assessment for the region.

The Challenge

Classifying land cover over Cyprus from 10m satellite data faces specific challenges:

Mediterranean climate – hot, dry summers and mild, wet winters cause seasonal vegetation changes, affecting classification accuracy.
Spectral similarity – wild_grass and cultivated_crops can be difficult to separate, especially in transitional seasons.
Urban and industrial areas – cities like Nicosia, Limassol, and Larnaca have complex built‑up patterns mixed with vegetation.
Water bodies – reservoirs and seasonal rivers may vary in extent, requiring careful temporal selection.
Agricultural diversity – includes olive groves, vineyards, citrus orchards, and cereal crops.

The Nazru Solution

Nazru’s platform integrated multi‑spectral satellite bands with a deep learning semantic segmentation model. The pipeline:

Pre‑processing – atmospheric correction, cloud masking, and resampling to consistent 10m resolution.
Classification – pixel‑wise assignment to one of nine classes using a model trained on Mediterranean landscapes.
Post‑processing – raster‑to‑polygon conversion with noise removal and minimum mapping unit (MMU) filtering.
Accuracy assessment – validated against high‑resolution reference data or field surveys.

Class definitions (same as previous project):

Class	Description
`bare_soil`	Unvegetated soil, sand, or rock surfaces
`cultivated_crops`	Active agricultural fields, orchards, vineyards
`forest`	Dense tree cover (natural pine forests, e.g., Troodos Mountains)
`industrial`	Factories, warehouses, industrial zones
`transportation`	Roads, highways, airports, and transport corridors
`urban_built`	Residential, commercial, and mixed‑use built‑up areas
`water`	Reservoirs, dams, rivers, and coastal water bodies
`wetland`	Marshlands, salt flats (e.g., Larnaca Salt Lake), and seasonally inundated areas
`wild_grass`	Natural grasslands, steppe, and Mediterranean scrubland (maquis)

Key Results & Benefits

Comprehensive land cover map of Cyprus at 10m resolution.
Clear separation of agricultural land, natural vegetation, urban, industrial, transport, water, wetlands, and bare soil.
Supports:
- Agricultural monitoring (crop types, irrigation planning).
- Environmental conservation (forest cover, wetland protection).
- Urban planning and infrastructure development.
Scalable – can be applied to multiple time‑steps for seasonal or annual change analysis.
GIS‑ready – polygon outputs with class attributes and area measurements.

Key Technologies Used

Semantic segmentation (U‑Net / DeepLabV3+) for pixel‑wise classification.
Multi‑spectral indices – NDVI, NDWI, and SAVI for vegetation, water, and bare soil differentiation.
Polygon vectorisation – clean, topologically correct polygons.
Atmospheric correction – using standard pre‑processing pipelines (e.g., Sen2Cor).