A Novel Approach to Fennel Variety Classification in India: Employing CNN and RNN to Distinguish Florence, Selma, and Perfection

The study presents a new mix of Convolutional Neural Networks (CNNs) and recurrent Neural Networks (RNNs) for more accurate fennel-kind sorting. It shows an amazing 92.64% accuracy in telling Florence, Selma, and Perfection styles apart. This model uses pictures and order data to tell apart the complex differences between these fennel types. Preparing the images included making them all 224x224 pixels, and then adding extra changes and adjustments to make sure they worked better with the computer model. There were over 10 thousand pictures in total. The model design combines CNNs for handling images and RNNs to analyze sequences, showing high precision and trustworthiness in sorting. Precision, recall, and F1 scores show how well the model can find and get examples of each fennel type right. Comparing it with other methods, the model is more accurate. This makes fennel-type classification a big step forward using modern techniques. The idea's results go beyond farming, showing how flexible it is in many ways of classifying things. Its power in farming gives farmers an automatic tool to guide their crops. This helps with growing methods, attracting markets, and ensuring quality control of the produce for sale. But, in the plan to make it work better for huge amounts of data and look at more ways to study information. This study is a big step forward in automatic sorting methods. It helps make advancements in smart farming and opens doors for high-tech machine learning uses on plant groupings.