A New Dawn for Tomato-spotted wilt virus Detection and Intensity Classification: A CNN and LSTM Ensemble Model

Tomato-spotted wilt virus (TSWV) is a severe plant disease that causes significant economic losses in tomato production worldwide. Early detection and intensity classification of TSWV-infected tomato plants is critical for effective disease management. This study proposes a novel TSWV detection and intensity classification approach based on a convolutional neural network (CNN) and a long short-term memory (LSTM) network ensemble model. A dataset comprising 30,000 images of tomato plants infected with TSWV was gathered and annotated with six intensity levels, ranging from 0 (indicating no symptoms) to 5 (indicating severe symptoms). A framework approach was developed, with aiming to enhancing the model’s performance r proposed approach achieved an overall accuracy of 97.37% on the test set, outperforming several state-of-the-art approaches. We also performed a statistical analysis of the inter-intensity level variability of the classification accuracy and found that the accuracy increased with the intensity level. Our results suggest that the proposed approach has the potential to be used in the early detection and intensity classification of TSWV-infected tomato plants, which could aid in the timely application of preventive measures and reduce the economic losses caused by TSWV.