The ever-increasing problem of population expansion and limited food supply and resources has ballooned into the defining issue of our generation. As the population rate heads steadily towards the 2050 estimates of $10 billion, people are racing to figure out what to do about our limited supplies.
Perhaps technology, which got us into this mess, will also get us out of it. After all, many humanitarian efforts have been bolstered by the advancement of technology. The initiative of the ongoing green revolution has been significantly bolstered by the development of impressive robots and machine learning algorithms. Among these helpful inventions are things like satellites that automatically detect drought patterns, tractors that eyeball plants and kill the sick ones, and an AI-powered smartphone app that can tell a farmer what disease has crippled their crop.
A lot of farmers have found that they can make use of deep learning, a computing method by which programmers don’t need to tell a computer what to do but instead can train it to recognize patterns on its own. Using this technology, you can feed a computer photos of diseased and healthy plant leaves and a computer can eventually learn the difference between the two and be able to determine which is which on its own.
With this exact system, biologist David Hughes and epidemiologist Marcel Salathe taught a computer to identify 26 diseases within 14 crops. While the technology still needs revision (poor lighting and strange backgrounds in images taken from the internet can cause the computer to make mistakes), Hughes and Salathe hope to use AI in this way to power their app, PlantVillage.
PlantVillage enables farmers to upload pictures of their ailing plants which are then diagnosed by experts. Smartening their AI means making it that much easier for farmers all around the world to diagnose and treat plant diseases before they become regional famines. That’s why Hughes and Salathe need to feed their AI system “More and more images from various sources, in terms of how the pictures were taken, time of year, location, and so on… And the algorithm can just pick up on that and learn.”
It’s not just disease that will impede a plant’s growth. Even more common are outside conditions that are less than ideal for the plants:
“Most diseases that hamper growers are physiological stresses, so not enough calcium or magnesium or too much salt or too much heat,” Hughes explained. “People often think it’s a bacterial or fungal diseases.”
Unfortunately for farmers, a misdiagnosis can not only cause a plant to die, but it can make a farmer unwittingly waste money and time on pesticides or herbicides that are also harmful to the surrounding environment. With AI technology, farmers may be able to more quickly and accurately pinpoint the problems keeping their farms from producing to their full potential.
The UN has seen lot of potential value in this realm of advancement, stating that such technology is a “useful tool” for crop management, but that there’s no real replacement for an expert eye. Perhaps with enough AI advancement, there will be.