Onyango Rollins Otieno and Aggrey Shitsukane Shisiali

Abstract

In the agricultural sector, the prevalence of challenges such as plant diseases and pest infestations result in significant crop losses, thereby impacting global food production. Traditional disease identification methods exhibit inefficiency and delay, necessitating the pursuit of innovative solutions. The integration of image processing and deep learning technologies emerges as a promising avenue for expeditious and accurate detection of plant diseases and pests. Augmented by embedded systems, this integration facilitates real-time responses, effectively bridging the temporal gap between identification and intervention. The system will operate by identifying and classifying plant health statuses based on images of leaves or crops, generating comprehensive reports encompassing causes, recommendations, and treatment strategies. Boasting a user-friendly interface and automated processes, the system is designed to furnish farmers with timely and precise information for efficient plant health management. The transformative potential of this system extends beyond the immediate context, promising to revolutionize agricultural practices, curtail crop losses, and contribute to the global adoption of sustainable and efficient farming methodologies. The main objective is to empower farmers with actionable insights, enabling swift responses to potential plant health concerns and ultimately elevating agricultural productivity.

Keywords — CNN, Deep Neural Networks, Machine Learning.