This paper first proposes a new quality metric for edge evaluation, based on six physical characteristics. These physical characteristics that affect human edge quality evaluation are continuity, smoothness, thinness, localization, detection and noisiness. The final edge quality score is a weighted linear combination of the quantified measures of these six edge quality attributes. The other feature of this edge evaluation metric is its adjustability. Through some training procedures, it can be adjusted to suit different user and application needs. In the latter part of this paper, an edge detector performance predictor is proposed. By a few initial measurements of image parameters, the performance of certain edge detectors can be predicted. Finally, the performance of several popular edge detectors is compared, under different variations of SNR, blurting and power spectrum.
| Identifer | oai:union.ndltd.org:arizona.edu/oai:arizona.openrepository.com:10150/277249 |
| Date | January 1990 |
| Creators | Chang, Dunkai Kyle, 1962- |
| Contributors | Strickland, Robin N. |
| Publisher | The University of Arizona. |
| Source Sets | University of Arizona |
| Language | en_US |
| Detected Language | English |
| Type | text, Thesis-Reproduction (electronic) |
| Rights | Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author. |
Page generated in 0.0019 seconds