Tools
Tools
Shyamkuwar, Pranay (2019) Non-negative matrix factorization for classifying the defects on steel surface using Convolutional Neural Network. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (1MB) | Preview |
Preview |
PDF (Configuration manual)
Download (730kB) | Preview |
Abstract
Classification of defects on steel surface in a steel industry can significantly improve the production which can increase in profit. This has become a concern for all the companies worldwide increasing profit and reducing the production error. For a certain period of time machine-based inspection of the defects on steel surface has widely received attention. Due to the limitation of a human eye for manually recognizing the flaw is a very slow process. This research mainly focuses on classifying six different types of defects usually occurred during production of steel slab with the help of machine learning algorithm. This study focuses on using Deep CNN with Gaussian blur and Non-negative matrix factorization. First, we implemented by applying Gaussian blur on images with kernel size of (3x3 & 5x5) with Non-negative matrix factorization and 44.4% of accuracy is achieved when applying gaussian blur for kernel size of 3x3. However, accuracy is reached to 45.6% for kernel size of 5x5. Whereas for second experiment NMF is applied excluding gaussian blur and accuracy is reached to 93.1%.
Keywords: Gaussian Blur, Deep CNN, Non-negative matrix factorization.
| Item Type: | Thesis (Masters) |
|---|---|
| Subjects: | Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QA Mathematics > Computer software T Technology > T Technology (General) > Information Technology > Computer software T Technology > TH Building construction |
| Divisions: | School of Computing > Master of Science in Data Analytics |
| Depositing User: | Dan English |
| Date Deposited: | 09 Jul 2020 14:22 |
| Last Modified: | 09 Jul 2020 14:22 |
| URI: | https://norma.ncirl.ie/id/eprint/4325 |
Actions (login required)
 |
View Item |