×
Home Current Archive Editorial board
Instructions for papers
For Authors Aim & Scope Contact
Original scientific article

A DEEP ENSEMBLE LEARNING AND INTELLIGENT HYBRID FEATURE EXTRACTION MODEL FOR THE PEST DETECTION AND CLASSIFICATION IN CLOUD COMPUTING

By
P. Bharathi Orcid logo ,
P. Bharathi

Research Scholar, Department of Computer Science, Kongunadu Arts and Science College , Coimbatore, Tamil Nadu , India

K. Dhanalakshmi Orcid logo
K. Dhanalakshmi

Associate Professor, Head, Department of Information Technology, Kongunadu Arts and Science College , Coimbatore, Tamil Nadu , India

Abstract

This study presents an advanced approach for Pest Detection (PD) through automatic identification of invasive insects using Deep Ensemble Learning (DEL) and intelligent Feature Extraction (FE) techniques. The process begins with Z-score normalization (ZSN) to eliminate noise and improve classifier performance. Improved Feature Weighted Fuzzy Clustering (IFWFC) model is presented to achieve effective data grouping and Henon Chaotic Map Encryption (HCME) model to encrypt the image is presented. The extraction of the feature step involves the application of Hybrid Enhanced Wild Horse Optimization (EWHO) and Gray-Level Co-occurrence Matrix (GLCM) to extract the features accurately. The feature selection (FS) is performed on a modified version of Cuckoo Search Algorithm (MCSA) to achieve a precision in the model. The DEL model combines several deep learning models such as ResNet 50, Enhanced DenseNet201 and Granular Neural Network (GNN), which are optimized using the Pelican Optimization Algorithm (POA), to achieve a better classification. The IP102 data set was used to test the system with a better score of 99.90 as accuracy, 99.92 as precision, and 99.91 as a recall. The proposed model is much better than the current pest recognition techniques in classification with respect to accuracy, thus making it useful in real-time pest detection. This study presents the prospect of DEL-based model in precision agriculture and provides an efficient method of pest identification which can lead to better crop production and less dependence on pesticides.

Citation

This is an open access article distributed under the  Creative Commons Attribution Non-Commercial License (CC BY-NC) License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 

Article metrics

Google scholar: See link

The statements, opinions and data contained in the journal are solely those of the individual authors and contributors and not of the publisher and the editor(s). We stay neutral with regard to jurisdictional claims in published maps and institutional affiliations.