Identification of Potato Plant Diseases using Deep Neural Network Model and Image Segmentation

Abstract

This research proposes an approach for classifying diseases affecting potato leaves using DL and image segmentation techniques. Collecting data, preprocessing the data, segmenting an image, extracting features, and classifying the picture are the five main components of the suggested technique. The primary objective is to use image processing methods to identify diseased potato plants within the Plant Village Dataset (PVD) of photos. Diseased areas in images of potato leaves are separated using the K-Means clustering technique. A deep neural network (DNN) model with Adam and categorical cross-entropy hyperparameters categorises PLD. With the suggested approach, the classification accuracy attained is 98% for PLD detection. Our model successfully detected and classified leaf diseases in potato plants, as evidenced by experimental findings.