An Approach to Identify Diseases in Betel Leaf Using Deep Learning Techniques

Despite its many benefits, such as vitamin C, thiamine, niacin, riboflavin, and carotene, we fail to preserve betel vine leaves against bacteria, resulting in a substantial crop loss worldwide. Because of the high level of damage, identifying leaf diseases are essential. To classify betel leaf diseases, several deep learning methods such as VGG16, VGG19, Resnet50, AlexNet, and InceptionV3 were applied using transfer learning, as well as some traditional machine learning approaches such as SVM and Logistic Regression. The dataset contains 5800 individual images of betel leave with two diseases (Stem Leaf and Bacterial Leaf Spot) that were captured using a mobile phone camera and preprocessed using Contrast limited adaptive histogram equalization (CLAHE) and Gaussian mixture model (GMM). Finally, the InceptionV3 reached the maximum accuracy of 94.83 % among the rest of the implemented algorithms, on an unseen test set with a total of 658 images.