Using machine learning to tackle Fall Armyworm

In response to the severe impact of the Fall Armyworm (FAW) on maize crops in Africa, particularly in Uganda where agriculture forms a significant part of the economy, a small team of Ugandan developers devised a machine learning solution to aid local farmers. Leveraging TensorFlow, they developed the Farmers Companion app, which uses smartphone images and TensorFlow Lite to detect FAW damage in maize crops and suggest solutions. Despite challenges such as limited resources and data collection difficulties, the team has continued to expand their dataset and improve the model's accuracy. Their efforts have gained recognition through features on national TV and international events, and the app is continually updated to support more crops and integrate cloud services for better performance. The initiative aims to enhance food security and reduce hunger by employing advanced technology in agriculture.