An Automated Approach for Detection and Classification of Plant Diseases
The results of this study show that the EfficientNet B1 model suggested worked remarkably well, with a 99% accuracy rate, which emphasizes how well the model can classify plant illnesses, which could be helpful for farmers and other agriculture specialists.
Abstract
Plants, including many different foods, from grains to fruits and vegetables, are vital sources of nutrition and essential parts of numerous economies and ecosystems. This project uses the PlantVillage dataset to address automated plant disease identification demand. Using transfer learning, we proposed the EfficientNet B1 model, which has demonstrated encouraging outcomes in earlier studies. The dataset included 20,639 photos that were divided into 15 different classifications. The model was executed using the default learning rate across 40 epochs with a batch size 40. The results of this study show that the EfficientNet B1 model we suggested worked remarkably well, with a 99% accuracy rate. This result emphasizes how well the model can classify plant illnesses, which could be helpful for farmers and other agriculture specialists. Early disease identification made possible by this technology can promote food security and sustainable agriculture. The results of this study have possibilities for the agricultural industry, as quick and accurate disease detection can reduce crop loss and increase productivity. It also emphasizes how advanced deep learning techniques can be used in practical applications.