Deep Learning to Mosquitoes Species Pattern Recognition

Conference proceedings article

Authors/Editors

Strategic Research Themes

Publication Details

Author list: Boonprakong P., Chamnongthai K., Patchoo W., Mongkalangoon P.

Publisher: Taiwan Association for Aerosol Research

Publication year: 2021

Start page: 464

End page: 468

Number of pages: 5

ISBN: 9781665411974

ISSN: 1680-8584

eISSN: 2071-1409

URL: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85125166275&doi=10.1109%2fICSEC53205.2021.9684591&partnerID=40&md5=035b048295b28a2b78c6118ae56a7568

Languages: English-Great Britain (EN-GB)

View on publisher site

Abstract

The epidemic from mosquito-borne pathogens is difficult to control and monitor and is considered as a silent threat that cannot be realized. Importantly, therapeutic vaccines are developed to increase immunization to avoid the danger of life. We found these problems in daily life. Mosquitoes live with people in society and the most common species are Aedes Aegypti, Aedes Albopictus, Anopheles minimus, Culex Quinquefasciatus and Armigeres subalbatus. In general, they carry germs to people. In this paper, we are interested in classifying 5 types of mosquitoes. So, we apply LeNet 5. The results show that training network for is each species has the accuracy of 96-99%. Testing network for is each species has the accuracy of 96.25% The results are based on mosquitoes image dataset of 1, 200 images, divided as 840 training images and 120 validating images, and 240 testing images. © 2021 IEEE.

Keywords

LeNet5, Mosquitoes species