Being able to identify crop problems early can make the difference between saving a crop and losing it, but high-tech solutions can be costly. An interdisciplinary team of researchers are launching an inexpensive camera system that can monitor crop stress remotely-and it costs less than the average smartwatch.
Developed by NC State computer vision and machine learning expert Paula Ramos-Giraldo, the StressCam system is based around a Raspberry Pi, a tiny, inexpensive and easily-programed computer. Solar powered and WiFi-enabled, it has a camera that takes pictures of a field and records signs of droughts such as wilting or curling leaves. The tiny computer runs a machine learning algorithm on the photos to analyze them for indications of drought stress. Then it sends this information to a web platform for resilience researchers, plant breeders and eventually farmers.
Undergraduate and graduate students were involved in designing the system’s machine learning algorithm as well as building the web-based interface for different groups of users.
“I think one of the coolest things about engineering is applying science and technology to an industry that could benefit from it, because otherwise it’s just math on paper,” said Artem Minin, an NC State Electrical and Computer Engineering senior who worked on the StressCam. “I really enjoyed working in the agricultural industry, and I think there’s a lot of applications for technology in the agricultural industry that haven’t been fully explored yet.”
The eventual goal is to roll out the StressCams to the Precision Sustainable Agriculture research network, a network of farms and research stations in 22 states supported in part by a USDA Agriculture and Food Research Initiative grant for the development of resilient agricultural systems.
This project is one of many coming out of the N.C. Plant Sciences Initiative, which aims to redefine team science and activate data-driven solutions that solve the world’s grand agricultural challenges.