ARASAT
Satellite view of irrigated agricultural fields

Products · Agriculture

Crop Type Classification

Field-level crop identification from the seasonal signature, while the crop is still standing.

Processing service

Processing service delivered per area of interest

Overview

What is planted in each field this season, and how much area does each crop cover?

Every crop has a phenological signature: a characteristic curve of greenness, water content and canopy structure through the season. The product reads that curve from a stack of optical and radar observations and matches it against the crop calendar of the zone, so a classification is made in-season rather than after harvest. Because the method is calibrated per zone rather than hard-coded, it transfers across regions, climates and hemispheres.

The field, per season
Unit
In-season, before harvest
Timing
Optical and radar time series
Inputs
Map, statistics, report
Delivery

Example outputs

  • Parcel polygons outlined over mountainous terrain

    Field-level units over mountainous terrain with small, irregular fields.

How it works

Pipeline and maturity of each stage

  1. 01

    Time-series stack

    Operational

    Cloud-masked optical observations and radar backscatter are assembled for the whole season over the area of interest and aligned to field boundaries.

  2. 02

    Phenological features

    Operational

    Spectral indices and radar profiles are turned into per-field seasonal curves, which are robust to individual cloudy or noisy dates.

  3. 03

    Classification

    Operational

    Time-series models trained on locally labelled fields, anchored to the zone's crop calendar, assign a crop class and a confidence to each field.

  4. 04

    Aggregation and delivery

    Operational

    Field results are aggregated to district and regional cultivated-area statistics and delivered as a crop map, tables and a report.

Data used

  • Sentinel-2
  • Sentinel-1 SAR
  • Landsat 8/9
  • Field boundaries
  • Local crop calendar and field labels

Methods and models

  • Time-series deep learning
  • Phenology-anchored classification
  • SAR and optical fusion
  • Per-zone calibration

Outputs

  • Crop map (GeoTIFF)
  • Per-field class and confidence (vector)
  • Cultivated-area statistics by crop and unit (CSV)
  • PDF report

Who it is for

  • Agriculture ministries and statistics offices
  • Crop insurance and subsidy verification
  • Water allocation and irrigation planning
  • Commodity and supply-chain analysis

More in Agriculture

All agriculture products

From space. For Earth.

Start with ARASAT

Tell us what you need to monitor, understand or protect. Our team works with you from the first conversation to a running program.

Contact usOpen the platform