Jager, Brian de (2023) Edge Detection with Inhibition for Vectorization. Bachelor's Thesis, Computing Science.
|
Text
bCS_2023_JagerB.pdf Download (52MB) | Preview |
|
![[img]](https://fse.studenttheses.ub.rug.nl/style/images/fileicons/text.png) |
Text
toestemming.pdf Restricted to Registered users only Download (135kB) |
Abstract
We propose using a computational model of simple cells with push-pull inhibition as a replacement for Canny’s edge detector for the purpose of an image vectorization pipeline. The new image vectorization pipeline is based on a pre-existing image vectorization pipeline. We demonstrate that using the new edge detection method in an image vectorization pipeline we can improve the vectorized result, achieve better reliability, and generate simpler meshes. We show this improvement with the use of a handpicked data set of 20 noisy and 20 regular images from the ImageNet Object Localization Challenge - a subset of a widely used data set used for advancing computer vision and deep learning research. The image vectorization pipeline we propose is a contribution to the field of image processing as it provides solid evidence that we can improve the image vectorization process by using different methods of edge detection. As a result, noisy images get vectorized into simpler meshes and will be easier to work with. Additionally, the mean squared error of the vectorized result will be significantly lower.
| Item Type: | Thesis (Bachelor's Thesis) |
|---|---|
| Supervisor name: | Azzopardi, G. and Kosinka, J. |
| Degree programme: | Computing Science |
| Thesis type: | Bachelor's Thesis |
| Language: | English |
| Date Deposited: | 25 Oct 2023 06:59 |
| Last Modified: | 25 Oct 2023 06:59 |
| URI: | https://fse.studenttheses.ub.rug.nl/id/eprint/31556 |
Actions (login required)
 |
View Item |