AI-Enabled IoT mobile application for early maize plant disease detection; case study: NGOMA District

Abstract

Maize crop has become significant food security and income-generating crop for small-scale farmers in Rwanda. Unfortunately, maize farmers are still experiencing significantly lower yields due to several diseases. Diseases affect the quality of maize crops and reduce the efficiency of agriculture production resulting in a significant loss to the farmers. Maize plant health conditions play a vital role in earning good profit to farmers; moreover, plant health conditions should be monitored at different stages of plants growth for early treatment of plant diseases. Currently, the techniques used by Rwandan farmers to diagnose maize diseases depend on naked eyes observation requires being well trained and experienced, as some plant diseases are very hard to be recognized. Another technique is sending samples to the lab for testing; this is expensive and time-consuming. To overcome limitations presented by these techniques IoT and AI technologies are great imperative technologies for making farming more efficient; these technologies can mitigate measures to help farmers avoid losses and ensure good food security in the different sides of the country. In this thesis, for early maize plant disease detection, an AI-enabled IoT mobile application was prosed to help farmers automatically detect maize plant diseases at an early stage of plant growth. For detecting plant disease, plant image is captured through the camera and uploaded to the local server using an android application, the plant image undergoes various image processing algorithms at the server for determining the disease, and detected disease is sent back to the farmer's mobile application with remedies. Various performance indicators, such as classification accuracy and processing time, were used to evaluate our system. The model has an overall classification accuracy of 80% when it comes to distinguishing the three most common disease groups that damage maize leaves.