Early Disease Detection in Plants using CNN

Abstract

The Indian economy benefits from early disease detection in plant leaves. According to reports, 10-30% of crops suffer harm from diseases that are not discovered during the curing process. Recent advancements in deep learning, particularly Convolutional Neural Networks (CNNs), have paved the way for transformative solutions in disease identification for agriculture. This study addresses the critical issue of early disease detection by harnessing the power of deep learning models, specifically CNNs. Different leaf disease detection technologies are used for different crops. The pre-trained deep learning model is used in this study to identify and categorize leaf diseases. A dataset of tomato, potato, and bell pepper leaf pictures from the plant village repository was employed for the current investigation. The developed model can detect 12 plant diseases in normal leaf tissue. The quantitative assessment of our CNN-based technique reveals an impressive accuracy rate of 86.21%. This notable accuracy underscores the efficacy of our approach in the challenging domain of plant disease detection. Our findings show potential for the larger agricultural environment beyond the immediate quantitative advantages. This technical advance not only paves the way for improved crop output and lower losses, but also brings in a new era of data-driven sustainability in agriculture. The bottom line is that our research "offers a tangible pathway for leveraging AI-powered solutions to address the long-standing challenges of plant disease detection, thereby significantly contributing to the well-being of farmers and the sustenance of the Indian agricultural sector.