Internet of Things Enabled Pomegranate Leaf Disease Detection and Classification using Cuckoo Search with Sparse Auto Encoder

Authors

  • P. Sindhu, G. Indirani

DOI:

https://doi.org/10.17762/msea.v71i3.252

Abstract

Recent technological advancements in the field of Internet of Things (IoT) and computer vision (CV) enable proper identification of plant diseases, which is a major challenge in agricultural productivity. Since plant leaf diseases mainly affect crop productivity and quality, earlier recognition of diseases becomes essential. The latest developments of deep learning (DL) models enable us to effectually categorize the existence of pomegranate leaf diseases. Therefore, this article develops an IoT Enabled Pomegranate Leaf Disease Detection and Classification using Cuckoo Search with Sparse Autoencoder (PLDDC-CSSAE) model. The presented PLDDC-CSSAE model determines appropriate class labels of the pomegranate leaf diseases accurately and rapidly. The PLDDC-CSSAE model initially employs Gaussian filtering-based noise removal with Shannon entropy based image segmentation. Next, NasNet model is exploited to produce high level deep features and finally, cuckoo search (CS) algorithm with SAE model is utilized for classification. The CS algorithm aids in the appropriate parameter choice of the SAE model and consequently resulting in enhanced performance. The PLDDC-CSSAE model shows better result over other methods by using  benchmark dataset..

Downloads

Published

2022-06-09

How to Cite

P. Sindhu, G. Indirani. (2022). Internet of Things Enabled Pomegranate Leaf Disease Detection and Classification using Cuckoo Search with Sparse Auto Encoder. Mathematical Statistician and Engineering Applications, 71(3), 904 –. https://doi.org/10.17762/msea.v71i3.252

Issue

Section

Articles