Arecanut Disease Detection Using CNN and SVM Algorithms
Mamatha Balipa, Pallavi Shetty, Arhath Kumar, B. R. Puneeth, Adithya
Abstract
Areca nut is a tropical crop which is also known as betel nut. India is second in both areca nut production and consumption worldwide. A variety of diseases affect it throughout its life cycle from root to fruit. Cultivation classification is among the most crucial processes in crop management. Classification might be advantageous for different grades. Several textural features are extracted from the areca nut using Wavelet, Gabor, Gray Level Difference Matrix (GLDM), and Gray Level Co-Occurrence Matrix (GLCM). Currently, disease detection is done purely by visual observation, and farmers must carefully analyze each crop on a regular basis to detect diseases. A picture is entered into a convolutional neural network (CNN), a deep learning algorithm, which then assigns learnable weights and biases to various objects in the image, and then, based on the outcomes, learns to differentiate between them. We created a dataset of over 180 images of healthy and diseased areca nuts to train and test the CNN model. Support vector machine is also used to distinguish between healthy and diseased areca nuts.