Next-Gen Hybrid Deep Learning for Accurate Pepper Leaf Mosaic Virus Classification

This research presents a new way to sort out the stages of pepper leaf mosaic virus illness using an advanced learning model. It combines Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks for this purpose. This study looks at how using space and time data can improve the way we find out about plant diseases, which is a better method than what we already have. The set of pictures has 10,500 illness images sorted into different health stages. These were handled and used to teach the model about disease by testing it out too. The suggested CNN-LSTM mix model uses the skill of CNN to get space details from leaf pictures and LSTM's ability to look at how these features change over time. This mix is made to make disease stage classification better and faster. The model got a total correctness of 96.5%, much better than single CNN and LSTM models. The model was very good when checked with new data. This shows it can work well in real life too. The results of this study are very important for precision farming and plant disease research. The fused CNN-LSTM model is very accurate and works well in most cases. This makes it great for finding plant diseases early, which can help manage them better to save more crops from getting sick. The study also shows where more research can be done in the future. This includes changing models to fit different environmental situations and sick plants, along with not depending too much on big sets of data.