Cauliflower Leaf Disease Detection Using Computerized Techniques

Abstract

Crop cultivation is an important role in the agriculture industry. Presently, food loss is primarily caused by sick crops, which affects growth rate and increase. High yield depends a lot on its growth. However, now crop leaf disease has become a significant problem in agriculture as a result of which the quality and growth rate of yield in agriculture is declining day by day. Through this paper, we have tried to diagnose the leaf disease of a particular crop (cauliflower). In this study, we collected data from each cauliflower leaf and created a dataset, and we divided our study was split into two sections: leaf disease detection and machine learning techniques. To detect leaf disease, we have used Matlab and various algorithms of classic machine learning such as Nav Bayes (NB), Decision Tree (TT), Random Forest (RF), Support Vector Machine (SVM), Secular Minimal Optimization (SMO). For each, we determined the accuracy, retraction, F-measurement, accuracy, and ROC values classification. This paper describes a simple but effective technique for reviewing leaf disease detection evidence and applying it to image processing and computer vision.