By Joan Ahumuza
Fall armyworm, the larval life stage of a fall armyworm moth, impacts maize crops worldwide but particularly in countries like Uganda, where agricultural businesses employ 70% of the population.
Studies show the potential impact is between 8.3 and 20.6 million tons per year, with the fallout amounting to between USD2.48m and USD6.19m per year.
The threat of devastating losses prompted developers participating in a Google Developer Group in Mbale to create an Android app — FlatButter — that identifies signs of fall armyworm damage in maize crops.
It’s been featured on a national TV station in Uganda and highlighted by the Food Agricultural Organization of the United Nations, as well as by Google in a short film published today.
“The vast damage and yield losses in maize production, due to FAW, got the attention of global organizations, who are calling for innovators to help,” wrote Hansu Mobile and Intelligent Innovations CEO Nsubuga Hassan, who led the team that developed the app.
“It is the perfect time to apply machine learning. Our goal is to build an intelligent agent to help local farmers fight this pest in order to increase our food security.”
The app developers first collected data samples from nearby fields, using their smartphones to capture images after they categorized these before tapping Google’s TensorFlow to retrain an AI model with transfer learning, a technique where a model suited to one task is re-purposed on a second, related task.
Using TensorFlow Converter, a tool that takes a TensorFlow model and generates a lightweight version, they integrated a trained image classifier into the aforementioned FlatButter app.
Currently, the team’s training data set numbers nearly 4,000 data samples, but they say it’s growing as they continue to snap images of affected maize.
In any case, the interface is locked in tight: Users focus their smartphone camera on a crop and capture an image, after which the app’s model analyzes the plants in view for damage and suggests possible solutions.
The team says it’ll next tackle coffee and cassava diseases as it shifts to cloud services and Google’s Firebase development platform. “Our plan is to collect more data and to scale the solution to handle more pests and diseases,” said Hassan.
“With improved hardware and greater localized understanding, there’s huge scope for Machine Learning to make a difference in the fight against hunger.”
They’re not the first to apply AI to ecology. Microsoft recently highlighted a Santa Cruz-based startup — Conservation Metrics — that’s leveraging machine learning to track African savanna elephants.
Separately, a team of researchers developed a machine learning algorithm trained on Snapshot Serengeti that can identify, describe, and count wildlife with 96.6% accuracy.
Intel’s TrailGuard AI system prevents poaching by detecting motion with an embedded camera using an offline, on-device AI algorithm.
And scientists at Queensland University used Google’s TensorFlow machine learning framework to train an algorithm that can automatically detect sea cows in ocean images.