Evaluating Watermelon Mosaic Virus Seriousness with Hybrid RNN and Random Forest Model: A Five-Degree Approach

The study brings a mixed deep learning model for the exact measure of Watermelon Mosaic Virus (WMV) seriousness levels inside watermelons. The model uses a mix of Recurrent Neural Networks (RNNs) and Random Forests to get 99.8% overall good results when looking at more than 16,000 images with different levels of WMV problems. The model is strong because of accuracy, remembering well, and F1 score. These measurements show it works the same way at all levels of seriousness. The mixed design carefully cleans up and teaches with WMV-infected pictures. It can find complex patterns well in them. The model's skill at correctly sorting five levels of seriousness is shown by performance evaluation measures and an analysis called the Confusion Matrix. This also shows where there may be room for improvement. This study is a big step forward in the field of farming viruses. It gives an accurate way to measure how bad WMV can get. The model's accuracy gives people a good tool to find out problems early and make smart choices about how they manage diseases. This helps protect watermelon plants from threats linked with WMV at the farm too. Additionally, this research shows the possible use of mixed deep-learning models in farming situations. It sets up future improvements and wider applications to fight crop diseases more effectively. This study helps to understand WMVs better and makes watermelon farming stronger against viruses. It supports keeping crops healthy for longer.