Abstract
This article considers edge detection using neural network approach, and describes the integration of neural networks into a committee. The novelty of the proposed technique comes from two ideas: firstly, since the training set describes a solution, the proposed algorithm is able to detect multidirectional edges on noised image. Secondly, to increase the accuracy of detection, neural networks were combined into a committee.
Original language | English |
---|---|
Pages | 66-69 |
Number of pages | 4 |
Publication status | Published - 2018 |
Externally published | Yes |
Event | 28th International Conference on Computer Graphics and Vision, GraphiCon 2018 - Tomsk, Russian Federation Duration: 24 Sep 2018 → 27 Sep 2018 |
Conference
Conference | 28th International Conference on Computer Graphics and Vision, GraphiCon 2018 |
---|---|
Country | Russian Federation |
City | Tomsk |
Period | 24.9.18 → 27.9.18 |
Keywords
- Committee classification
- Digital image
- Edge detection
- Neural network
- Neural network committee
ASJC Scopus subject areas
- Computer Graphics and Computer-Aided Design
- Computer Vision and Pattern Recognition