Plant Disease Recognition Using Different CNN Models
Shivam Gupta, Shrey Gilotra, Shivam Rathi, Tanupriya Choudhury, Ketan Kotecha, Tanupriya Choudhury
Abstract
The research paper highlights the remarkable threats that plant diseases pose to global agricultural productivity and food security. The early detection and accurate recognition of these diseases are crucial for the effective management of plants. Deep learning models, such as CNN, Resnet34, Resnet50, and VGG16, have shown propitious results in computer vision tasks, including plant disease recognition. The study explores the effectiveness of these models using a comprehensive dataset consisting of images of healthy and diseased plants. The pretrained versions of the models are fine-tuned using transfer learning techniques with performance evaluation based on metrics like accuracy, precision, recall and f1 score. Comparative analysis also identifies each model's strengths and weaknesses in classifying plant diseases, considering factors like model architecture depth and complexity. The overall findings of this research contribute towards the development of automated plant disease recognition systems integrated into real-time monitoring systems.