The next green revolution may be happening in the server room.
What’s new: Microsoft open-sourced a set of AI tools designed to help farmers cut costs and improve yields.
How it works: FarmVibes-AI includes systems that analyze overhead imagery and sensor data to guide farm operations.
- AsyncFusion uses drone imagery, satellite imagery, and data from soil sensors to map soil conditions in real time. Farmers can use the output to plan where and when they should plant their fields.
- DeepMC is a neural network that combines data from soil sensors, climate sensors, and weather predictions to forecast field temperature, precipitation, and soil moisture up to 120 hours ahead. Its output can enable farmers to prepare for extreme temperatures and other events.
- SpaceEye, another neural network, filters clouds from satellite imagery for use by AsyncFusion and DeepMC. Microsoft engineers trained the network via an adversarial method using infrared and visible-light images partly covered with synthetic clouds.
Behind the news: Nonprofits and academic institutions provide other open-source AI systems to increase food production in collaboration with large agribusiness firms, independent farmers, and rural communities.
- Last year, the Linux Foundation launched Agstack, a partnership among universities, nonprofits, and IBM. The effort provides code, data, and frameworks to developers of open-source AI projects for agriculture.
- MIT’s now-defunct OpenAg included models that predicted how plants would grow under various environmental conditions.
Why it matters: The emerging practice of precision agriculture, which seeks to take into account not only entire fields but also local conditions down to the level of individual plants, could help farmers sow seeds, grow crops, fight pests, and harvest produce more efficiently. Off-the-shelf systems may not serve farmers who work in different parts of the world or grow niche crops. Open-source projects can expand their options effectively and inexpensively.
We’re thinking: Farmers tend to welcome innovations that improve yields and cut costs. They’re also famously self-sufficient, performing repairs and installing upgrades to their equipment. As self-driving tractors and precision-ag systems take root, they’re great candidates to become early adopters of industry-focused platforms that make it easy for anyone to build useful AI applications.