A Comparative Performance Evaluation of Various Classification Models for Detection and Classification of Flying Insect

Nithin Kumar (), Nagarathna, Vijay Kumar L and Francesco Flammini
Additional contact information
Nithin Kumar: Vidyavardhaka College of Engineering, Mysuru, India
Nagarathna: PES College of Engineering, Mandya, India
Vijay Kumar L: University of Agricultural Sciences, Bangalore, India
Francesco Flammini: University of Applied Sciences and Arts of Southern Switzerland, Manno, Switzerland

Interdisciplinary Description of Complex Systems - scientific journal, 2023, vol. 21, issue 1, 52-68

Abstract: Agriculture has long been a part of Indian culture. It is known as the Indian economy’s backbone. Agriculture contributes to 17 % of the Indian GDP, but still, farmers confront several problems in growing their crops, one among them is insect pests. “Computational Entomology” is a branch of data mining that assists farmers in overcoming the challenges of damaging insect pests by utilizing appropriate sensors and methodologies for pest classification and application of the pesticides at the right time. The authors used various machine learning and deep learning algorithms to classify insects and examine the influence of classification performance on multiple classes of insects often found in Indian agricultural fields with varying numbers of data and classification models. The study found that proposed CNN based classification model performs better than other classification models in insect categorization, with a classification accuracy of 94,6 %. The research work done till now in the field of computational entomology deals with the insects grown in laboratory colonies or well-developed insects grown in the same geographic region and condition, but we have evaluated the performance of different classification models using random images available over the internet to select the well-suited classification model to classify flying insects. Applications with precise insect classification using machine learning and deep learning algorithms would have significant implications for entomological research. It is necessary to develop an automated insect classification techniques to provide a foundation for future research in the field of computational entomology.

Keywords: computational entomology; insect pest; machine learning techniques; classification; deep learning techniques (search for similar items in EconPapers)
JEL-codes: C88 (search for similar items in EconPapers)
Date: 2023
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.indecs.eu/2023/indecs2023-pp52-68.pdf (application/pdf)

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:zna:indecs:v:21:y:2023:i:1:p:52-68

Access Statistics for this article

More articles in Interdisciplinary Description of Complex Systems - scientific journal from Croatian Interdisciplinary Society Provider Homepage: http://indecs.eu
Bibliographic data for series maintained by Josip Stepanic ().