Banana crops are prone to damage by several types of pests and diseases. Once the pest or disease afflicting a crop is identified, swift and targeted action can reduce the extent of outbreaks and potentially save entire harvests.
CGIAR researchers at the Alliance of Bioversity International and CIAT have developed a digital tool to help farmers better protect their banana crops. The tool is a mobile application that merges expertise on banana genetic resources with artificial intelligence to quickly identify common afflictions that threaten bananas, allowing farmers and extension workers to act quickly and save their crops.
The smartphone app, called Tumaini – which means “hope” in Swahili – helps banana farmers scan plants for signs of five major diseases and one common pest. Farmers use the app to upload a photo of an affected crop, which is then scanned for symptoms of pests and diseases using image-recognition technology, drawing on a dataset of more than 50,000 images. Tumaini records the data, including geographic location, and feeds it into the database. The app then provides a diagnosis and recommends steps to address the affliction.
The Tumaini app has so far demonstrated a 90% success rate in detecting pests and diseases
Other existing crop disease detection methods focus primarily on leaf symptoms and can only accurately function when pictures contain detached leaves on a plain background. The novelty of Tumaini is that it can detect symptoms on any part of the crop – including the fruit, bunch or plant – and can read low-quality images, even those containing background noise, like other plants or leaves, to maximize accuracy.
Tested in Colombia, the Democratic Republic of the Congo, India, Benin, China, and Uganda, the Tumaini app has so far demonstrated a 90% success rate in detecting pests and diseases. The work is a step toward creating a satellite-powered, globally connected network to control disease and pest outbreaks, say the researchers who developed the technology.
Research on the app was published in 2019 in the journal Plant Methods and became one of the year’s most successful research papers for Bioversity International and CIAT. It also generated significant media attention and sparked inquiries from industry stakeholders regarding possibilities to expand the use of the app. To date, some 2,000 farmers are using the app in the field. A second version, released in 2020, allows for offline use, and researchers expect uptake of the app to increase as a result.
Header photo: A CIAT researcher demonstrates how to add a photo to the Tumaini app. Photo by CIAT.