Precision Agricultural Classification of Indian Turnip Varieties: CNN and Naive Bayes Methodologies

This study uses picture-identifying Convolutional Neural Networks (CNN) and Naive Bayes models to tell three big Indian turnip kinds apart - Pusa Swarnima, Purple Top White Globe, and Snowball. It focuses on different things they look like. The study uses about 10,500 high-quality pictures taken by the researcher to look at ways of identifying different types of images. The CNN design made for sorting different types of turnips did very well. It was 96.95% right overall in what it could do. Measuring success by looking at accuracy, recall and F1-score shows how well the model finds small differences between types of turnips. A good comparison chart shows that the CNN model is better than other methods for correctly sorting different types of turnips. While showing great results, know there might be problems with picture quality and surroundings that can limit what's possible. Ideas for things to do in the future are expanding data collection with various environments and using transfer learning. This helps improve model flexibility. This study shows how the CNN model helps farming by choosing turnip kinds accurately. This is good for saving plant diversity and improving ways of growing crops in India.