NORMA eResearch @NCI Library

A Comparative study between Traditional Machine Learning and Deep Learning Models to classify Rice Types

Mohamed Ali, Zainab Maqsud Ghulam Hussain (2023) A Comparative study between Traditional Machine Learning and Deep Learning Models to classify Rice Types. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (1MB) | Preview
[thumbnail of Configuration manual]
PDF (Configuration manual)
Download (956kB) | Preview


Rice is one of the most commonly cultivated grains on the planet. When purchasing rice packets at the market, the physical appearance is the first thing that comes to mind for buyers. There are also rice varieties with numerous distinguishing traits. Typically, these attributes consist of shape, texture, and colour. On the basis of these distinguishing properties of the various types of rice, classification and identification of quality are possible. Machine learning models help facilitate the recognition of trends and patterns. Traditional machine learning models like Support Vector Machines (SVM), Random Forest and Decision Tree are compared with deep learning models like Convolutional Neural Network when it comes to classifying different types of rice (CNN). Common types of rice used in this experiment include Arborio, Basmati, Ipsala, Jasmine, and Karacadag. Over 75,000 images of rice grains are included in the dataset, with 15,000 depicting each kind. Images were preprocessed to prepare them for feature extraction. Values for quantifying feature set performance were produced using these models. Based on the results, we can see that the Random Forest model had the highest classification accuracy at 92.66%, followed by the SVM model at 89.33%, the Decision Tree model at 77.33%, and the CNN model at 50.66%. The classification increases with the inclusion of each new feature. On the basis of the obtained performance measurement results, it is feasible to conclude that the study was successful in identifying rice types.

Item Type: Thesis (Masters)
Hasanuzzaman, Mohammed
Uncontrolled Keywords: Rice Classification; Machine Leaning; Random Forest; Support Vector Machines; Decision Tree; Convolutional Neural Network; Image Processing; Deep Learning
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
S Agriculture > SB Plant culture
Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Tamara Malone
Date Deposited: 23 May 2023 12:46
Last Modified: 23 May 2023 12:46

Actions (login required)

View Item View Item