Machine learning is solving challenging problems that impact everyone around the world. See how researchers at PlantVillage of Penn State University and the International Institute of Tropical Agriculture (IITA) are using ML and TensorFlow to help farmers detect diseases in Cassava plants.
Attending the summit from Penn State were PlantVillage team members Amanda Ramcharan, postdoctoral fellow in entomology, and Peter McCloskey, research technologist in entomology.
PlantVillage and its mobile app — called "Nuru," which is Swahili for "light" — uses artificial intelligence and machine learning to train computers to recognize disease symptoms. When deployed on a smartphone, the app couples with the device's camera to capture images of diseased plants and provides the user with a preliminary diagnosis with a high degree of accuracy. The user also can get disease-management information and advice.
The program incorporates TensorFlow, open source software for numerical computation using data flow graphs. Originally developed by Google's Machine Intelligence research organization for the purpose of conducting machine learning and deep neural networks research, TensorFlow is general enough to be applicable in a wide variety of other domains as well, according to the TensorFlow website.
"With partners at the International Institute of Tropical Agriculture, we're deploying our app to help African farmers, who typically don't have access to the kind of resources — such as land-grant agricultural scientists and extension specialists — that American farmers have," Hughes said. "But they do have smartphones in increasing numbers."
The Google video featuring PlantVillage focuses on the research group's work in Tanzania, demonstrating how local farmers are using the app to diagnose disease in cassava, an important crop that helps to feed 500 million Africans every day, Hughes noted.
"In the last 160 years, we've largely solved the food security problem in the United States, and now we want to be a 21st century land-grant institution," he said. "We won't do that with bricks and mortar and establishing extension offices in developing countries."
With mobile devices as the platform, researchers can democratize access to artificial intelligence, according to Ramcharan.
"Now a smallholder farmer can have access to the latest technology and human expertise in the palm of her hand," she said. "This is crucial for female farmers who do most of the farming in the developing world but face the most barriers to accessing technology and knowledge."
Hughes added that the research team is looking to implement these technologies on a global scale to enhance food security and improve people's livelihoods.
"Google's spotlight on PlantVillage at the TensorFlow Summit is a testament to the difference we can make with the creative use of the right technology," he said.