Identification and classification of cassava plant leaf disease using deep learning technique

Abstract

Cassava is a woody shrub cultivated largely for its edible roots, the largest carbohydrate, and starch-rich vegetable. The plant leaf disease affects the productivity that primarily influences the farmers, as plant leaf diseases occur naturally. A Deep Convolutional Neural Network (DCNN) based leaf disease identification model is proposed. Automatic detection of cassava leaf disease using novel techniques assists humans in monitoring the huge farm of crops manually. Early detection of the disease symptoms on plant leaves is a better option to avoid spreading diseases throughout the farmland. The diseased leaf region's statistical features are extracted using Deep Learning (DL) technique and fed as input to the classifier network that adopts an open dataset pre-trained model with different classes of leaf disease. First, the given input image is identified as affected or healthy. Followed by classifying the type of leaf diseases based on the pre-trained database model feature extracted in the system.