Breaking New Ground: High-Accuracy Detection and Severity Analysis of Onion Smut Using AI

Finding and putting plant diseases in the right category is very important for farming ways. It directly affects how you look after your plants and the food they give. In this work, a mixed deep learning (DL) model is made and tested to directly deal with the problem of finding out onion smut disease. This new study puts together the strong power that Convolutional Neural Networks (CNN) have with the powerful ability to sort things out which Support Vector Machines (SVM) possess. For the sickness's seriousness, this method can place onion smut into six separate groups. When tested on 2600 real-time pictures, the model got an overall correctness of 92%. This clearly shows that it is good at finding diseases correctly and also figuring out how bad they are. The study shows that this mixed CNN-SVM model is much better than the old ways of making things right. In the future, farming and plant care will be a helpful tool for doing jobs very accurately. This job is helpful for the farming area of artificial intelligence (AI). It also shows how future studies can use a mix of machine learning and sorting techniques to group plant illnesses.