Plant Disease Identification Using Convolution Neural Network

Abstract

In the present day, the prevalence of plant diseases is emerging as a significant factor contributing to reduced food production and increased losses for farmers. Hence, it is imperative to employ a method capable of delivering swift and precise results. The recent expansion of deep learning has proven instrumental in addressing both conventional and unconventional challenges more effectively. The Convolutional Neural Network (CNN) has emerged as a cutting-edge approach for state-of-the-art identification and detection. To tackle the issue of plant diseases, we have established a comprehensive dataset featuring 37 different plants and crops for training and validating our model. Our implementation involves Resnet (Residual Neural Network), a specific CNN architecture. We capture images of diseased plant leaves and employ CNN-based classification for disease detection. Our model demonstrates superior accuracy compared to numerous previously utilized techniques for disease detection.