Improving Anthracnose Severity Grading in Green Beans through CNN-LSTM Integration

The present research is designed to build and modelized accurately the severity of anthracnose by integrating Convolutional Neural Networks (CNN) with Long Short-Term Memory (LSTM). Another fungal disease that affects agriculture adversely and leads to a considerable drop in production level and overall quality is known as anthracnose. Economic losses can be avoided by simply early and accurate monitoring of the illness intensity and adopting informed management strategies. The research used the high-resolution green bean leaf images dataset that was annotated using the three levels of infection with anthracnose disease as the drawings of the disease progression. The models were proposed following a hybrid approach between CNN and LSTM algorithms for feature extraction and processing. The model’s performance was evaluated by using standard metrics like accuracy, precision, recall, and F1 score. The results were compared with CNN and LSTM banded models on a standalone basis and state-of-the-art approaches were also considered. The CNN-LSTM model with an integrated structure that combines CNN’s ability to detect local features and LSTM’s for learning long-term dependencies, demonstrated a valuable performance with a 94.8% overall accuracy, in the grading of anthracnose severity. Such performance thus outperforms the results interpreted from the competing models by much, which is an indication of the employed integrated strategy being dominant for the understanding of anthracnose attributes and their changing patterns. The report after all recommends the application of CNN and LSTM models jointly though it aims to improve the accuracy and reliability of the agricultural diseases severity classification. The new model not only provides crop pathology evaluation in green beans anthracnose severity grading but also is scalable and hypothetically could be reused in defining the intensity of diseases in other crops taking into account their features. It is of a very high standard and among the paramount gains in precision agriculture because of the contribution it makes to disease management strategies and enhanced sustainable farming.