Authors:
Ian Williams
1
;
David Svoboda
2
and
Nicholas Bowring
3
Affiliations:
1
Birmingham City University, United Kingdom
;
2
Masaryk university, Czech Republic
;
3
Manchester Metropolitan University, United Kingdom
Keyword(s):
Edge detection, Grey-scale measure, Performance measure, Connectivity, Figure of merit.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Biomedical Engineering
;
Biomedical Signal Processing
;
Computer Vision, Visualization and Computer Graphics
;
Data Manipulation
;
Feature Extraction
;
Features Extraction
;
Health Engineering and Technology Applications
;
Human-Computer Interaction
;
Image and Video Analysis
;
Image Filtering
;
Image Formation and Preprocessing
;
Informatics in Control, Automation and Robotics
;
Methodologies and Methods
;
Neurocomputing
;
Neurotechnology, Electronics and Informatics
;
Pattern Recognition
;
Physiological Computing Systems
;
Sensor Networks
;
Signal Processing, Sensors, Systems Modeling and Control
;
Soft Computing
;
Statistical Approach
Abstract:
This paper will discuss grey-scale edge detection evaluation techniques. It will introduce three of the most common edge comparison methods and assess their suitability for grey-scale edge detection evaluation. This suitability evaluation will include Pratt’s Figure Of Merit (FOM), Bowyer’s Closest Distance Metric (CDM), and Prieto and Allen’s Pixel Correspondence Metric. The relative merits of each method will be discussed alongside the inconsistencies inherent to each technique. Finally, a novel performance criterion for grey-scale edge comparison, the Grey-scale Figure Of Merit (GFOM) will be introduced which overcomes some of the evaluation faults discussed. Furthermore, a new technique for assessing the relative connectivity of detected edges will be described and evaluated. Overall this will allow a robust and objective method of gauging edge detector performance.