As global population rises — mostly in the developing countries — and demand for food increases, the food supply challenge is renewed for the agricultural sector. At the same time, the agriculture sector grapples with internal problems such as insufficient farm yield, inefficient production methods and pests and weeds. While many methods are being applied to optimize output, the application of Artificial Intelligence (AI) in agriculture represents a paradigm change in optimizing resources and supplying food. AI can potentially help in data collection, yield protection from diseases and pest and yodel production. AI technologies is already being applied though not on a large scale and uniformly. The application of AI also faces certain challenges in the form of data availability and connectivity that are unique to the agricultural sector.
Fast and relevant data collection is important in helping the agriculture sector meet the challenges. Many different data collection technologies are coming to the market. For example, the high-speed variable rate planting equipment provides accurate “as planted” information and yield monitors provide harvest data. Such data are useful in building predictive algorithms.
Pest and plant disease detection
Pests and diseases can destroy yield. AI technologies can detect plant pests and diseases more quickly and accurately than a trained human being. One such technology has been the image recognition algorithm by Resson, a Monsanto Growth Ventures (MGV) portfolio company with offices in Canada and San Jose.
AI is used to develop accurate yield predictions. Companies such as Orbital Insights, Descartes Labs, Gro Intelligence, and Tellus Labs have been developing yield prediction algorithms based on satellite imagery, weather information, and historical yield data. They claim the predictions are more accurate then United States Department of Agriculture (USDA) reports.
The same AI technology can be potentially applied across the globe irrespective of farm, soil, climatic or other conditions. AI technology adapts based on unique regional conditions. For example, Slantrange, a San Diego-based start-up, found its plant counting algorithm incompatible with the South African farm conditions. So, it adapted its technology accordingly.
AI can help in yield improvement by providing products for development of new seeds, fertilizers, or crop protection. In fact, yield improvement products may be preferred over some other agricultural AI products such as data collection because data collection has AI been going on for a long time, even without AI and yield improvement products have more financial value.
Probably it is going to take a while before AI is embraced in agriculture on a large scale. AI needs to sort out the challenges before that. The biggest contribution of AI will most probably be optimizing resources.
Author Bio: Kaushik is a passionate technical writer and a software architect by profession. He is interested in Big Data, AI, Machine Learning and Enterprise Application Development.