Creation of an Android Application and the Use of Transfer Learning to Recognize Insect Species


Authors : Prabhat Kumar Singh; Pawan Kumar; Abshar Imam

Volume/Issue : Volume 7 - 2022, Issue 3 - March

Google Scholar : https://bit.ly/3IIfn9N

Scribd : https://bit.ly/3IBwJVG

DOI : https://doi.org/10.5281/zenodo.6386609

Abstract : Insect numbers are dwindling over the world, and some species have gone extinct in the past. Exploration of seldom seen Insect species has therefore become a difficult endeavor for Entomologists and Insect Watchers. We created an Android application based on deep learning to assist users in recognizing 280 different insect species, making insect categorization much more user-friendly. We employ Convolutional Neural Networks (CNN) pre-trained on ImageNet Dataset as freeze layers of the network in this article, then train the final output layer, which has 280 separate classes. The accuracy of CNN models such as EfficientNet-lite0, InceptionV3, Xception, ResNet-50, MobilenetV2, and InceptionResNetV2 has been evaluated, and the mobile app's operation has been discussed. Maximum train data accuracy of 99.81 percent and test data accuracy of 98.62 percent is accomplished.

Keywords : Transfer Learning, Classification of Bird Species, Deep Learning, Recognition of Image, CNN, Android Application

Insect numbers are dwindling over the world, and some species have gone extinct in the past. Exploration of seldom seen Insect species has therefore become a difficult endeavor for Entomologists and Insect Watchers. We created an Android application based on deep learning to assist users in recognizing 280 different insect species, making insect categorization much more user-friendly. We employ Convolutional Neural Networks (CNN) pre-trained on ImageNet Dataset as freeze layers of the network in this article, then train the final output layer, which has 280 separate classes. The accuracy of CNN models such as EfficientNet-lite0, InceptionV3, Xception, ResNet-50, MobilenetV2, and InceptionResNetV2 has been evaluated, and the mobile app's operation has been discussed. Maximum train data accuracy of 99.81 percent and test data accuracy of 98.62 percent is accomplished.

Keywords : Transfer Learning, Classification of Bird Species, Deep Learning, Recognition of Image, CNN, Android Application

Never miss an update from Papermashup

Get notified about the latest tutorials and downloads.

Subscribe by Email

Get alerts directly into your inbox after each post and stay updated.
Subscribe
OR

Subscribe by RSS

Add our RSS to your feedreader to get regular updates from us.
Subscribe