Using CNN-DT Model to Classify Indian Cauliflower Varieties in New Ways

Improving farming relies on correctly sorting crop types. This research introduces a new way to classify different types of cauliflower in India. It uses something called Convolutional Neural Networks (CNN) and Decision Trees (DT). The dataset comprises four distinct cauliflower varieties prevalent in India: Pusa Deepali, Pusa Snowball K-1, and Pusa Meghna are types of snowball. Strict data cleaning made the datasets more equal and added variety. The CNN-DT model is very good. It has a high 96.7% total correctness in sorting these types into their proper categories without mistakes. Looking at it in recent ways shows that this model is better than others in being accurate and working well. It's a big step forward for categorizing types of plants. The mixed model not only does well in getting things right but also allows clear choices to be made. This is important for farming uses. We use CNNs and DT together to understand important differences between cauliflower types. This helps us see how they look different from each other. This study shows ways for farm technology, to help people to manage crops better with useful information. The model shows amazing precision and easy understanding, but future improvements will focus on fixing the problem of different amounts in classes. It also needs to add more variety to its information. In the end, this study creates a new way to classify cauliflower types. It will help make farming better with new ideas and improve how we manage crops more efficiently.