Penn State A.I. Can Spot Plant Disease

Artificial intelligence neural network developed by Penn State can see plant disease using a smartphone

Immediate need in food insecure Africa relying on Cassava vegetable

From Axios:

Cassava root plant is grown in tropical climates

Researchers at Penn State University and the International Institute for Tropical Agriculture in Tanzania have trained a neural network running on a smartphone to spot cassava disease with 93% accuracy.

Why it matters: Diseases and pests threaten cassava, a root vegetable that is critical for food security in sub-Saharan Africa, and can lead to losses of over a $1.2 billion each year.

The vision: "In the same way we have personalized medicine, I think we can have personalized agriculture on scale for hundreds of millions of people in real time," Penn State's David Hughes told Wired. He envisions farmers and gardeners using artificial intelligence to help them diagnose their crops.

How it works: Neural networks are typically trained with millions of images. Instead of starting from scratch, the researchers took advantage of something called transfer learning in which an existing network is retrained with new data. In this case, they used 2,756 images of cassava leaves with tell-tale signs of disease.

The cassava: the toxic potato of the tropics 

The cassava is a root plant grown in tropical environments and is called a cassava in Africa and yuca in South America, though they are different varieties. It is a source of carbohydrates and a staple in the diets of Sub Saharan Africans, much like the potato is in the US. 

It must be carefully prepared because the more bitter varieties emit toxins that can be lethal. Often these varieties are avoided but will be eaten during times of scarcity and are sometimes preferred as a way to ward of animal and human theft from food supplies. Source