Diagnosis and Identification of Citrus Canker Growth Rate Using Machine Learning

Citrus canker is a bacterial disease caused by Xanthomonas citri pv citri. Mainly the research in the field of Citrus canker has been done on fruits; early detection of disease on leaves can help in adopting preventive measures at early stage. The present study focuses on the detection of Citrus canker in leaves and classification of six different growth stages/levels of disease: water soaking, yellow initiation, chlorosis, and blister formation, canker infection (50%), canker infection (100%). The primary dataset was developed by taking images of citrus leaves infected by X. citri for the specified time. The proposed strategy is based on the detection of the lesion spot on citrus leaves and the classification of citrus disease stages/level was done to evaluate the growth rate of citrus canker through six distinct growth levels and evaluate affected area of leaf with each passing day. The entire processing includes Pre-processing, Segmentation, Features extraction and perform Classification using four different classifiers which include SVM, KNN, Naive Bayes, and Neural Network. The efficiency of different classifiers was 96.77% by SVM, followed by KNN 90%, NB 84% and NN 82% respectively. The lowest execution time was achieved by NB 0.011s, followed by KNN 0.015s, SVM 0.021s, and N.N. 0.216s.