 Google has introduced India-first artificial intelligence models designed to bring more detailed, field-level insights to agriculture. Developed by Google DeepMind’s AnthroKrishi team, the Agricultural Landscape Understanding (ALU) and Agricultural Monitoring and Event Detection (AMED) models can help map farmland, monitor agricultural activity, and support data-driven farming decisions.
Key Highlights:
- AI models: Agricultural Landscape Understanding (ALU) and Agricultural Monitoring and Event Detection (AMED).
- ALU: Helps map agricultural fields and analyse land patterns using spatial and farm-related data.
- AMED: Tracks changes in crops, land use, and other agricultural events at scale.
- Google Earth: Data from the models has been integrated as a visual agricultural data layer.
- Global expansion: The models have expanded beyond India to countries including Kenya, Uganda, Ghana, Rwanda, Nigeria and Zambia.
- Farmland mapping: TerraStack has used the APIs to map more than 140 million hectares of farmland.
- Water management: Karnataka’s Water Resources Department has used the data across 2.6 million hectares of irrigated land.
- Sustainable farming: CarbonFarm is using ALU with Gemini for low-carbon rice cultivation, targeting 2 million hectares by 2030.
- The models could support precision agriculture, crop monitoring, irrigation planning, and digital agriculture services.
|