Improving Citrus Farming Through Efficient and Accurate Diagnosis of Lemon Citrus Canker Disease with Deep Learning

Lemon citrus canker disease is a devastating bacterial disease that affects citrus trees and causes significant economic losses for citrus growers worldwide. The early detection and accurate classification of citrus canker disease are crucial for disease management and control. Within the scope of this research article, we suggest a hybrid model that combines a CNN and SVM for the multi-classification of lemon citrus canker diseases. The proposed model is trained on a dataset of six severity levels of citrus canker, ranging from healthy to severely infected, collected from citrus farms in Punjab, India. On the testing set, the suggested model achieved an overall accuracy of 94.03%, which was higher than the performance of other state-of-the-art models. The model that was presented has the potential to increase the accuracy and efficiency of citrus canker detection and classification, which can assist citrus producers and inspectors reduce the spread of the disease and minimize economic losses.