Empowering Smart Farming with Machine Intelligence: An Approach for Plant Leaf Disease Recognition

Abstract

The growing global demand for sustainable agriculture has led to increased interest in leveraging machine intelligence to address critical challenges in modern farming practices. This paper introduces an innovative approach for plant leaf disease recognition in smart agriculture using the Vision Transformer (ViT) model. The proposed framework combines the power of self-attention mechanisms and transformer-based architectures to capture intricate relationships between image patches, enabling accurate and efficient disease identification. Leveraging the widely recognized PlantVillage dataset as a case study, our experiments demonstrate the efficacy of the ViT model in achieving superior disease recognition performance. The results highlight the model's ability to generalize across diverse crops and diseases, making it a promising tool for empowering farmers with timely disease detection and management. Additionally, the paper emphasizes inclusivity, ensuring the accessibility and effectiveness of the approach for farmers across diverse regions, backgrounds, and resources. Through this work, we contribute to the advancement of smart farming practices and pave the way for sustainable agriculture in the era of machine intelligence