Classification of Cotton Leaf Diseases Using AlexNet and Machine Learning Models
Premkumar Borugadda *
Department of Computer Science, Pondicherry University, Pondicherry, India.
R. Lakshmi
Department of Computer Science, Pondicherry University, Pondicherry, India.
Surla Govindu
Department of Computer Science, Pondicherry University, Pondicherry, India.
*Author to whom correspondence should be addressed.
Abstract
Computer vision has been demonstrated as state-of-the-art technology in precision agriculture in recent years. In this paper, an Alex net model was implemented to identify and classify cotton leaf diseases. Cotton Dataset consists of 2275 images, in which 1952 images were used for training and 324 images were used for validation. Five convolutional layers of the AlexNet deep learning technique is applied for features extraction from raw data. They were remaining three fully connected layers of AlexNet and machine learning classification algorithms such as Ada Boost Classifier (ABC), Decision Tree Classifier (DTC), Gradient Boosting Classifier (GBC). K Nearest Neighbor (KNN), Logistic Regression (LR), Random Forest Classifier (RFC), and Support Vector Classifier (SVC) are used for classification. Three fully connected layers of Alex Net provided the best performance model with a 94.92% F1_score at the training time of about 51min.
Keywords: Cotton disease detection, machine learning, deep learning, F1_score.