Name
Abstract | ||
|---|---|---|
We seek an efficient enhancement process that sharpens an image and yet smooths areas of low differentials and removes low-level impulse noise. The goal is to perform these antithetical processes to bring a fairly good image into better focus that is not sufficiently sharp. Upon adopting the perspective of a constructor beetle that explores each pixel of an image and examines its neighboring pixels, we find a small set of operational rules. If the neighbors have similar height (gray level) to the center pixel, then the beetle digs or fills a small amount to further level the neighborhood (smoothing). On the other hand, if the neighbors tend to be very different on the average so they appear as a barrier of holes or poles, then the beetle digs or fills to construct a barrier. The beetle also levels a single deep hole or high pole at the neighborhood center, which smooths impulse noise. |
Year | DOI | Venue |
|---|---|---|
2000 | 10.1006/rtim.1998.0154 | Real-Time Imaging |
Keywords | Field | DocType |
nonlinear rule-based convolution,impulse noise,rule based | Computer vision,Rule-based system,Nonlinear system,Computer science,Convolution,Image processing,Smoothing,Impulse noise,Pixel,Artificial intelligence,Small set | Journal |
Volume | Issue | ISSN |
6 | 1 | Real-Time Imaging |
Citations | PageRank | References |
5 | 0.85 | 9 |
Authors | ||
1 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Carl G. Looney | 1 | 198 | 21.58 |