Defect green coffee bean detection using image recognition and supervised learning

Shafian Izan Sofian (2022) Defect green coffee bean detection using image recognition and supervised learning. Universiti Malaysia Sabah. (Unpublished)

	Text DEFECT GREEN COFFEE BEAN DETECTION USING IMAGE RECOGNITION AND SUPERVISED LEARNING.24pages.pdf Download (338kB)
	Text DEFECT GREEN COFFEE BEAN DETECTION USING IMAGE RECOGNITION AND SUPERVISED LEARNING.pdf Restricted to Registered users only Download (2MB)

Abstract

Addressing the quality of green coffee bean is an important process to define its quality and market price for any industry that processing it. Normally, the evaluation that is carried out in determining the quality of green coffee is by visual inspection where it has limitations, and it is prone to error. Therefore, in this research project, the process will be conducted by using an image classifier with the model of a machine learning algorithm which the candidates comprise of Support Vector Machine, k-Nearest Neighbour and Decision Tree. k-nearest neighbour has the highest F1-score (0.51) than the other two algorithms (Support Vector Machine: 0.50, and Decision Tree: 0.48). The model was integrated as web application with Flask where user can upload the image and the system will return result with precision and prediction. This integrated web application is tested with functionality test and integration test which it succeeded both successfully fulfilling each criterion tested.

Item Type:	Academic Exercise
Keyword:	Green coffee bean , Quality , Image recognition
Subjects:	Q Science > QA Mathematics > QA1-939 Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science > QA76.75-76.765 Computer software
Department:	FACULTY > Faculty of Computing and Informatics
Depositing User:	DG MASNIAH AHMAD -
Date Deposited:	18 Jul 2022 20:00
Last Modified:	18 Jul 2022 20:00
URI:	https://eprints.ums.edu.my/id/eprint/33344

Actions (login required)

View Item