Detection of Insects and Pests in Agriculture field using MobileNet

Abstract

Abstract: The Indian economy heavily relies on agriculture, with high-quality crop production playing a pivotal role. However, frequent pest attacks pose significant threats by reducing crop yields and compromising food safety through nutrient depletion. This adversely impacts the economy, leading to substantial losses for farmers and risking lives. Timely monitoring of crops is imperative to combat pests effectively, necessitating the use of appropriate pesticides. Pest detection technologies can aid in early intervention, preventing crop damage and pesticide overuse. Artificial intelligence (AI) emerges as a crucial tool in addressing agricultural challenges. This research focuses on utilizing the MobileNetV2 algorithm for pest classification, leveraging image reshaping and feature extraction techniques. Results indicate MobileNetV2 outperforms other pre-trained models, achieving a higher accuracy of 0.95. By enhancing pest detection capabilities, AI-based technologies offer promising solutions to bolster agricultural production and mitigate economic losses.