EfficientNet based Convolutional Neural Network for Visual Plant Disease Detection

EfficientNet based Convolutional Neural Network for Visual Plant Disease Detection

Crops grown in tropical, subtropical, and temperate climates are subject to a variety of diseases and pests. Plant illnesses are complicated by interactions between the virus, the host plant, and the insect. In this paper, we present a robust transfer-learning-based detector for real-time plant dise...

Full description

Bibliographic Details
Published in:19th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology, ECTI-CON 2022
Main Author: Kumar D.; Ishak M.K.; Maruzuki M.I.F.
Format: Conference paper
Language:English
Published: Institute of Electrical and Electronics Engineers Inc. 2022
Online Access:https://www.scopus.com/inward/record.uri?eid=2-s2.0-85133381132&doi=10.1109%2fECTI-CON54298.2022.9795496&partnerID=40&md5=52772a29c111bf47eb0d0b80cb66bf83
id 2-s2.0-85133381132
spelling 2-s2.0-85133381132
Kumar D.; Ishak M.K.; Maruzuki M.I.F.
EfficientNet based Convolutional Neural Network for Visual Plant Disease Detection
2022
19th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology, ECTI-CON 2022


10.1109/ECTI-CON54298.2022.9795496
https://www.scopus.com/inward/record.uri?eid=2-s2.0-85133381132&doi=10.1109%2fECTI-CON54298.2022.9795496&partnerID=40&md5=52772a29c111bf47eb0d0b80cb66bf83
Crops grown in tropical, subtropical, and temperate climates are subject to a variety of diseases and pests. Plant illnesses are complicated by interactions between the virus, the host plant, and the insect. In this paper, we present a robust transfer-learning-based detector for real-time plant disease detection with PlantDoc datasets and plant village dataset. EfficientNetV2 architecture is more efficient, high speed and accurate in compared with EfficientNetV1. The EfficientNetV2 with the pretrained weights, Image net and plant village obtain the highest score in accuracy with 74%,F1-score=0.74 if compared to other benchmarked models. © 2022 IEEE.
Institute of Electrical and Electronics Engineers Inc.

English
Conference paper
author Kumar D.; Ishak M.K.; Maruzuki M.I.F.
spellingShingle Kumar D.; Ishak M.K.; Maruzuki M.I.F.
EfficientNet based Convolutional Neural Network for Visual Plant Disease Detection
author_facet Kumar D.; Ishak M.K.; Maruzuki M.I.F.
author_sort Kumar D.; Ishak M.K.; Maruzuki M.I.F.
title EfficientNet based Convolutional Neural Network for Visual Plant Disease Detection
title_short EfficientNet based Convolutional Neural Network for Visual Plant Disease Detection
title_full EfficientNet based Convolutional Neural Network for Visual Plant Disease Detection
title_fullStr EfficientNet based Convolutional Neural Network for Visual Plant Disease Detection
title_full_unstemmed EfficientNet based Convolutional Neural Network for Visual Plant Disease Detection
title_sort EfficientNet based Convolutional Neural Network for Visual Plant Disease Detection
publishDate 2022
container_title 19th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology, ECTI-CON 2022
container_volume
container_issue
doi_str_mv 10.1109/ECTI-CON54298.2022.9795496
url https://www.scopus.com/inward/record.uri?eid=2-s2.0-85133381132&doi=10.1109%2fECTI-CON54298.2022.9795496&partnerID=40&md5=52772a29c111bf47eb0d0b80cb66bf83
description Crops grown in tropical, subtropical, and temperate climates are subject to a variety of diseases and pests. Plant illnesses are complicated by interactions between the virus, the host plant, and the insect. In this paper, we present a robust transfer-learning-based detector for real-time plant disease detection with PlantDoc datasets and plant village dataset. EfficientNetV2 architecture is more efficient, high speed and accurate in compared with EfficientNetV1. The EfficientNetV2 with the pretrained weights, Image net and plant village obtain the highest score in accuracy with 74%,F1-score=0.74 if compared to other benchmarked models. © 2022 IEEE.
publisher Institute of Electrical and Electronics Engineers Inc.
issn
language English
format Conference paper
accesstype
record_format scopus
collection Scopus
_version_ 1809678158628126720

Similar Items