Abstract: Agriculture is a major source of income in India, and the country's economy is heavily reliant on it. To maximize agricultural productivity and profit, it is critical to diagnose plant leaf diseases at an early stage. Because naked eye observation of diseases does not always yield reliable results, especially during the early stages, an image processing technique is utilized to detect leaf diseases accurately. It consisted of five steps: image acquisition, preprocessing of the acquired image, feature extraction, disease classification, and display of the results.. This work presents a thorough examination of the categorization of agricultural illnesses using the Support Vector Machine classifier. Recently, many works have been inspired by the success of deep learning in computer vision for plant diseases classification. Unfortunately, these end-to-end deep classifier slack transparency which can limit their adoption in practice. In this paper, we propose a new trainable visualization method for plant diseases classification based on a Convolutional Neural Network (CNN) architecture composed of two deep classifiers. The first one is named Teacher and the second one Student. This architecture leverages the multitask learning to train the Teacher and the Student jointly.Then, the communicated representation between the Teacher and the Student is used as a proxy to visualize the most important image regions for classification. This new architecture produces sharper visualization than the existing methods in plant diseases context.
Keywords: Image processing , Save Model, CNN, Leaf Disease, Matlab, Feature Extraction, Deep Learning, Leaf Dataset.
|
DOI:
10.17148/IJARCCE.2022.11641
