Development of Automated Leaf Disease Detection in Pomegranate Using Alexnet Algorithm

Abstract

Agriculture plays a crucial part in the overall growth of the country. Smart agriculture provides today's farmers a help in decision making. Farmers or professionals typically monitor crops for disease detection and identification by using traditional methods. Many techniques are designed to improve the productivity and quality of the crops, but the disease prediction of the existing techniques leads to loss of productivity and quality. Overcoming these issues this paper aims to develop a system to detect Bacterial Blight and Alternaria using Alexnet algorithm at early seedling stage. The main objective is to detect two major diseases Bacterial Blight and Alternaria using Alexnet algorithm. The dataset of pomegranate leaf images total 1245 was created which was unavailable for these diseases, 80% of dataset is used for the training part another 20% is used for the testing part. For evaluating the performance of Alexnet algorithm, the performance metrics such as accuracy, precision, recall was considered. Results showed a high accuracy rate of 97.60% and this developed pomegranate leaf diseases detection system is better than other algorithm in terms of accuracy, loss, recall. Developed system has loss of 0.1(scaled between 0-1) which is very less comparative to other similar models. Finding of this paper is Dataset was created which was unavailable and proposed approach have high accuracy than others through which we can detect diseases at very early seedling stage.