Edible Bird Nest Grade Classification

 




 

Lim, Xiu Kai (2019) Edible Bird Nest Grade Classification. Final Year Project (Bachelor), Tunku Abdul Rahman University College.

[img] Text
Lim Xiu Kai.pdf
Restricted to Registered users only

Download (4MB)

Abstract

Edible Bird Nests (EBN) is extracted from saliva of swiftlets and it is consumed by Chinese community for its rich nutritional and medicinal values. Accordingly, production of Edible Bird Nests is high in South East Asia. However, the process of grading EBN based on image is complicated since the images can be negatively impacted by impurities. In bird nest industry, there is still no automated and standardized application to grade EBN. Classification of EBN is based on manual inspection by the swiftlet farmers and the experts. Several state-of the art methods are mainly used to evaluate quality and differentiate grade of agriculture products according to its low level features. In this work, Convolutional Neural Network (CNN) is introduced to identify grade of Edible Bird Nests (EBN). High level approach such as VGG-16, VGG-19 and modified VGG-19 are used to classify the EBN into different grades. Comparisons of hyper-parameters are made to obtain highest testing accuracy of model. At the end of project, the accuracy obtained by proposed algorithm is 69%.

Item Type: Final Year Project
Subjects: Technology > Electrical engineering. Electronics engineering
Faculties: Faculty of Engineering > Bachelor of Engineering (Honours) Electrical and Electronics
Depositing User: Library Staff
Date Deposited: 31 Jan 2020 02:35
Last Modified: 31 Jan 2020 02:35
URI: https://eprints.tarc.edu.my/id/eprint/13052