Paddy leaf disease detection using a deep transfer learning approach

Abstract

Paddy leaf diseases pose a significant threat to crop fields and agricultural productivity, necessitating efficient detection methods for timely intervention. This study provides a comprehensive overview of recent advancements in paddy leaf disease detection, introducing an innovative approach utilizing deep learning models. Several integrated models for image classification were evaluated in the research, including VGG-19, VGG16, MobileNetV2, DenseNet, ResNet-50, Xception, Inception, and the customized Inception-V3. The findings of the study revealed that MobileNetV2 emerged as the top performer, achieving the highest accuracy of 95%. This model demonstrated exceptional performance in accurately identifying paddy leaf diseases. On the other end of the spectrum, Xception exhibited the lowest accuracy at 85%. The remaining models, including VGG-19, VGG16, ResNet-50, DenseNet, Inception, and the customized Inception-V3, showcased varying degrees of accuracy falling between these two extremes. This research underscores the potential of transfer learning models, including the customized Inception-V3, in enhancing disease detection accuracy. It emphasizes the significance of continuous innovation and improvement in disease detection methodologies to support sustainable and cost-effective farming practices. With the customized Inception-V3 model achieving an accuracy of 93%, this study highlights its promising performance in contributing to the advancement of agricultural technology.