Researchers at International Rice Research Institute (IRRI) have developed an Artificial Intelligence (AI)-based image analysis model that estimates expected rice yield.
Rice is one of the most important crops supplying up to 20 per cent of the world’s food energy. Estimating the yield in rice using the traditional method is time and labour-consuming. It requires harvesting, transportation, drying, separating grains from other parts, and weighing.
The process makes it difficult and quasi-impossible to evaluate rice productivity of large numbers of rice farmers on the experimental station in real-time. It is also challenging to evaluate the productivity on farmers’ fields at regional scale.
According to Plant Phenomics journal researchers from IRRI and Okayama University Japan have developed this cheaper and faster method of estimating rice yields that is expected to revolutionise the practice of growing rice.
To start, the research team collected a set of rice canopy images and corresponding grain yield across seven countries. The database consisted of over 20,000 images and 400 farmers with various levels of grain yield, growth environments and plant shape. The AI model was then trained by using this database.
“The results showed that the developed AI model accurately predicts rice grain yield by using only images of the rice canopy at harvest,” said Dr Kazuki Saito, a researcher at IRRI.
The new technology is expected to help smallholder farmers in Africa and other developing countries to make decisions and plans based on expected rice yield from their plots. Researchers note that the new technology does not require sophisticated equipment- only needing a digital image of the rice canopy, which can be easily obtained by using a digital camera or a smartphone.