Fruit Classification and Defect Detection System Using Faster Region Convolutional Neural Network

Aziz, Amir Aizat (2019) Fruit Classification and Defect Detection System Using Faster Region Convolutional Neural Network. [Final Year Project] (Submitted)

[thumbnail of AMIR AIZAT_23010.pdf] PDF
AMIR AIZAT_23010.pdf
Restricted to Registered users only

Download (727kB)

Abstract

Malaysia is still a net importer of both fresh and refined fruits and the fresh fruit export price
is around USD 174 million. Various methods are presented to improve fruit and vegetable
production. Using the latest technologies and knowledge-based production systems,
conventional farms will be turned into sustainable farms. Since consumers use the appearance
of fruits to first evaluate the quality of fresh food, the presence of skin defects appears to be
one of the most influential factors in fresh food quality and price. For this purpose, packing
houses need suitable systems capable of detecting skin deficiencies in fruits. The problem
statement of this study is packing houses do not have a proper method that can identify the
deficiencies in fruits using computer vision. It also involved computer vision technology
approach and machine learning doing supervised learning that used Faster R-CNN model as
element to achieve the objective of the project. Hence, this study been conducted to develop a
method to classify the types of fruits and identify defect of the fruits based on their outer skin.

Item Type: Final Year Project
Subjects: Q Science > Q Science (General)
Departments / MOR / COE: Sciences and Information Technology > Computer and Information Sciences
Depositing User: Mr Ahmad Suhairi Mohamed Lazim
Date Deposited: 09 Sep 2021 19:57
Last Modified: 09 Sep 2021 19:57
URI: http://utpedia.utp.edu.my/id/eprint/20890

Actions (login required)

View Item
View Item