Pepper Disease Detection Model Based on Convolutional Neural Network and Transfer Learning

Abstract

We used a deep learning approach based on convolutional neural networks to perform plant disease detection and diagnosis using leaves images of healthy and diseased plants. The model was trained using images from a data set of 2,478 chilies, consisting of 1,478 healthy leaves (19% from the field environment) and 1,000 infected leaves (10% from the field environment). The detection model is trained based on transfer learning, and the best performance reaches 99.55% accuracy when identifying diseased or healthy plants. The model can be applied to the early warning of pepper diseases, and the method can be further extended to support the identification of crop diseases under actual cultivation conditions.