NORMA eResearch @NCI Library

Hierarchical Classification of Insects using a Combination of Resnet and VGG Networks

Patil, Priyal Narendra (2022) Hierarchical Classification of Insects using a Combination of Resnet and VGG Networks. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (843kB) | Preview
[thumbnail of Configuration manual]
PDF (Configuration manual)
Download (1MB) | Preview


Image processing had been exponentially developed in the past few decades. Alongside the need and applications had been also rising. In our research, we approached the agricultural sector for image processing pests. As pests are most harmful to the crops and plants at any stage of the plant it is essential to detect them early as well as identify them in early stages which can help to tackle them with adequate measures. Our approach showcased the identification of the pest among 102 classes using VGG16 and ResNet mod- els which had attained accuracies of around 64 and 65.44% respectively. And one further investigation an unique method of training of model involving the AugMix technique and random noise in the training data helped to improve the accuracy of the ResNet50 and VGG16 hybrid model. With the proposed methodology model was able to achieve the accuracy of the 75% on the test dataset of the IP102.

Item Type: Thesis (Masters)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
Q Science > QL Zoology
H Social Sciences > HD Industries. Land use. Labor > Specific Industries > Agriculture Industry
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Tamara Malone
Date Deposited: 27 Feb 2023 17:14
Last Modified: 27 Feb 2023 17:14

Actions (login required)

View Item View Item