A Novel Classifier for Plant Health Monitoring: A Focus on Banana Leaf Disease Detection Using Deep Learning

Abstract

Crop productivity and food security in modern agriculture are largely dependent on efficient monitoring and early identification of plant diseases. With a focus on the identification of leaf diseases, this study investigates the use of machine learning algorithms in the context of plant health monitoring. To extract pertinent characteristics from photos of plant leaves, the study makes use of sophisticated image processing techniques. This results in a large dataset that can be used to train and assess machine learning models. The proposed approach makes use of a comparison study to assess how well different machine learning algorithms recognise and categorise distinct leaf disease kinds. High-resolution photos of plant leaves displaying disease signs are collected as part of the approach, which also include preprocessing the data to improve feature extraction and utilising labelled datasets to train machine learning models. After that, the models are evaluated on hypothetical data to see how well they generalise and perform in actual situations. In this paper an Ensembled CNN Leaf Disease Detection Classifier (ECNNLDD) used to predict the healthiness of a leaf which classifies whether is healthy or unhealthy which is a 2-class problem and compared with the existing classifier like Decision Tree and SVM algorithm in which the proposed classifier Ensembled CNN Leaf Disease Detection Classifier outperformed when compared with the existing classifiers. The findings of proposed classifier gave the best accuracy in agriculture by offering an intelligent and automated solution for early detection and diagnosis of plant leaf diseases. The integration of Deep learning into plant health monitoring systems holds the potential to revolutionize farming practices, enabling farmers to adopt timely and targeted interventions, thereby minimizing crop losses, and promoting sustainable agriculture.