Abstract | ||
---|---|---|
This paper presents three new algorithms for lambda-connected segmentation and fitting. It deals with a discrete system in which the elements are connected. The connectivity (known as the degree of connectedness) have the property, of gradual variation. The first algorithm proposed is a direct segmentation method for quadtree represented images. The algorithm does not decode the original quadtree code to restore the compressed image before segmentation. The second algorithm, called the lambda-band-connected search, is designed for noised image segmentation. It reserves a band width for the search agent to surpass, so that the search agent will not stop when it encounters a small noise. The third algorithm adds gradients and smoothes in lambda-connected fitting. |
Year | DOI | Venue |
---|---|---|
2004 | 10.1109/ICSMC.2004.1400884 | 2004 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN & CYBERNETICS, VOLS 1-7 |
Keywords | Field | DocType |
lambda-connectedness, segmentation, fitting, algorithm, graph | Scale-space segmentation,Image texture,Range segmentation,Computer science,Segmentation-based object categorization,Image segmentation,Artificial intelligence,Region growing,Connected-component labeling,Machine learning,Minimum spanning tree-based segmentation | Conference |
ISSN | Citations | PageRank |
1062-922X | 0 | 0.34 |
References | Authors | |
1 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Li Chen | 1 | 0 | 0.68 |
Osei Adjei | 2 | 14 | 4.20 |