Image-Based Oriental Food Recognition and Calories Calculation

 




 

LIm, Chee Hong (2020) Image-Based Oriental Food Recognition and Calories Calculation. Final Year Project (Bachelor), Tunku Abdul Rahman University College.

[img] Text
Lim Chee Hong.pdf
Restricted to Registered users only

Download (3MB)

Abstract

Food is one of the important things that are needed by every people in their daily life. With increasing of the type of food, food recognition has been developed to aid in some applications in order for people to record their calorie intake. However, food recognition technology remains as a challenging task to the researchers due to the diversity of the nature of food. Especially for the oriental food, the variety style of the cooking style makes the oriental food to become lookalike but with different name due to its cooking style. Hence, there are scarce works on the recognition of oriental food by using Convolutional Neural Network (CNN) based on the food images. Therefore, this research is to develop an oriental food recognition and calories estimation system by using Convolutional Neural Network (CNN). In this research, a CNN model called VGG network with different number of layer will be introduced and will be used in the recognition process. The objective of this research is to validate the performance of the CNN model in oriental food recognition with different hyper-parameter setting. In a nutshell, the average accuracy obtained by the proposed CNN model for oriental food recognition with optimum hyper-parameter setting is 80%.

Item Type: Final Year Project
Subjects: Technology > Electrical engineering. Electronics engineering
Technology > Food Technology
Faculties: Faculty of Engineering and Technology > Bachelor of Electrical and Electronics Engineering with Honours
Depositing User: Library Staff
Date Deposited: 24 Apr 2020 15:37
Last Modified: 11 Apr 2022 07:24
URI: https://eprints.tarc.edu.my/id/eprint/14249