Detecting and Estimating Severity of Leaf Spot Disease in Golden Pothos using Hybrid Deep Learning Approach

Abstract

The proposed study uses a hybrid model of convolutional neural networks (CNN) and long Short-Term Memory (LSTM) for the classification of healthy and leaf-spot diseased images of the Golden Pothos plant. A dataset of 8000 images was collected and pre-processed before being used for training and testing the model. The images were first classified into binary categories of healthy and leaf spot diseased and then into four different severity levels of the disease. The performance of the model was evaluated using various performance parameters, including accuracy, precision, recall, and F1-score. The model achieved an overall accuracy of 95.4% and 97.5% for binary and multi-class classification, respectively. The proposed model outperformed other state-of-the-art models for disease classification in plants, making it a promising solution for detecting plant diseases. Our study provides insights into the potential of using hybrid models in plant disease diagnosis and paves the way for further research in this area.