A novel hybrid approach for rice plant disease detection based on stacked autoencoder and convolutional neural network model

Manoj Agrawal and Shweta Agrawal

International Journal of Services, Economics and Management, 2023, vol. 14, issue 3, 321-344

Abstract: Among all staple of foods, rice is most commonly used all over the world. A rice plant disease is a major concern that shows its negative effect on crop yields. If regular and proper diagnosis of these diseases were not carried out, then it may decrease its production, and ultimately rise the food scarcity. Manual diagnosis is quite time-consuming. Therefore, this paper is dedicated to using the benefits of deep learning for the automated detection of diseases. In this paper, a novel hybrid model termed a stacked autoencoder module with ResNet (SAMResNet) is presented by cascading of stacked autoencoder (SAE) and ResNet50 for automated rice plant disease detection and compared with other convolutional neural network (CNN) models such as basic CNN and VGG16. Among all, SAMResNet has achieved highest accuracy of 97%. The benefit of using SAE is that it minimises the learning parameters so that complexity reduces and ultimately improves the detection accuracy with reduced dimensionality.

Keywords: rice disease; deep learning; stacked autoencoder; convolutional neural network; CNN; VGG16; ResNet50; SAMResNet. (search for similar items in EconPapers)
Date: 2023
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.inderscience.com/link.php?id=132232 (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:ids:injsem:v:14:y:2023:i:3:p:321-344

Access Statistics for this article

More articles in International Journal of Services, Economics and Management from Inderscience Enterprises Ltd
Bibliographic data for series maintained by Sarah Parker ().