Discover our high-performing ML models for geospatial analysis that can be quickly deployed with our tools: GeoAP — a ready-made platform for your AI-powered geo-analytical solutions with our open-source models or SoilMate — an easy-to-use service for quick geospatial analysis with built-in ML models.
Automated land usage detectionGeneral
Land Use/Land Cover map produced by Microsoft annually. The map is a composite of land use/land cover predictions for 9 classes each year from 2017
Blooming index detectionAgriculture
Model calculates the intensity of blooming for trees with white flowers, for example, almonds. It is useful for pollination and might help agronomists monitor the yearly almonds cycle.
Need a custom
We develop custom ML models tailored to address your business’ needs and requirements.
Fill in the form below, and we will reach you as soon as possible.
Growth anomalies detection. Performed through NDVI index calculation within the area of interest (AoI) and statistical methods application.
Moisture anomaly detectionAgriculture
Moisture index detection. Performed for a given AoI at the desired timeframe, selecting the least cloudy date.
Basic objects detectionGeneral
Vehicles and basic objects detection. Detects the most visible classes with a resolution of 0.3-0.5 m per pixel.
NDVI index calculationAgriculture
NDVI index calculation. Performed for a given AoI at the desired timeframe, selecting the least cloudy date.
Forests above ground biomass (AGB) and carbon stock (AGC) estimation. Model detects amount of carbon in forested areas. *price per 1 sq. km.
Crop type detectionAgriculture
Crop type detection model. Classifies each pixel of satellite data to detect 7 crop classes. It leverages a sequence of SAR and optical imagery to understand crop development dynamics. *price per 1 sq. km.
Plot boundary detectionAgriculture
Automated agricultural field boundaries detection. Performed during the growing season and easily detects the changing boundaries annually. *price per 1 sq. km.
Deforestation, afforestation detectionGeneral
Deforestation detection model. Pixelwise segmentation of optical data. Uses a sequence of 2 images to compare the difference between forest cover and classify it as clearcut or not.