Details
Presenter(s)
![Abdullah Al-Amaren Headshot](https://confcats-catavault.s3.amazonaws.com/CATAVault/ieeecass/master/files/styles/cc_user_photo/s3/user-pictures/21201.jpg?h=c27cc264&itok=gaiMSLOI)
Display Name
Abdullah Al-Amaren
- Affiliation
-
AffiliationConcordia University
- Country
-
CountryCanada
Abstract
Edge detection plays a very important role in many image processing and computer vision applications. In this paper, we propose a new non-deep learning edge detection technique based on three binary images with three different thresholds instead of gray or colored images. The proposed edge detection technique extracts the majority of the true edges. Extensive experiments are carried out and the results show that the proposed technique provides a performance ratio better than that of the existing non-deep learning techniques.