Scientists have trained an AI algorithm to count flowers on fruit trees using smartphone images. The system is being trained to predict harvest size months in advance, saving time, money, and water for farmers. Researchers from Chile, Spain, and the UK’s National Robotarium developed the algorithm in an attempt to address a global problem. The lead researcher stated that farmers often rely on manual methods to estimate their yields, resulting in significant error margins. Agriculture uses 70% of the world’s freshwater and wastes nearly half. And, half of the fruit and vegetables grown for human consumption are also lost!
The blossom-counting AI could reduce this waste and improve farmers’ allocation of water and fertilizer. The AI was tested in an orchard in Spain, where it predicted flower counts with 90% accuracy, compared to the 50-70% range of manual counts. The system recognizes unique flower patterns, shapes, and colors even when they overlap or are partially obscured. In September, the researchers will validate the AI’s predictions against a real peach harvest, hoping it can be adapted for other important fruit crops worldwide. Currently farming is one of the least efficient industries, but farmers are increasingly adopting technology like AI, drones, and robots to make their jobs easier, greener, and more profitable. For instance, German startup Constellr has developed a satellite-based crop monitoring system, and Belgium’s Robovision has created a platform for managing AI-based vision systems for farms.
The ainewsarticles.com article you just read is a brief synopsis; the original article can be found here: Read the Full Article…
