Abstract | ||
---|---|---|
The max-tree is a data structure that represents all possible upper thresholds of an image, it has been successfully used in many image processing and analysis applications. The max-tree corresponding to a natural image usually has thousands of nodes, which makes unpractical to build a comprehensive graphical representation of its complete structure. In this paper, we propose a methodology that allows to build an interactive max-tree graphical representation that permits the user to navigate through the max-tree nodes, to visualize its connected components and to create node subsets. Our representation displays a simplified max-tree, but it allows the user to access all max-tree nodes using the interactive features. To the best of our knowledge, this is the first work that proposes an interactive graphical representation of the max-tree. We depict the potential of our max-tree visualization tool for interactive segmentation, connected filtering, and collection of training samples. |
Year | DOI | Venue |
---|---|---|
2015 | 10.1109/IPTA.2015.7367110 | 2015 International Conference on Image Processing Theory, Tools and Applications (IPTA) |
Keywords | Field | DocType |
max-tree visualization,interactive max-tree,connected filtering,interactive segmentation | Computer vision,Scale-space segmentation,Feature detection (computer vision),Computer science,Image texture,Visualization,Image processing,Image segmentation,Connected component,Artificial intelligence,Minimum spanning tree-based segmentation | Conference |
ISSN | ISBN | Citations |
2154-512X | 978-1-4799-8636-1 | 1 |
PageRank | References | Authors |
0.35 | 16 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Luis A. Tavares | 1 | 1 | 0.35 |
Roberto Souza | 2 | 72 | 9.78 |
Leticia Rittner | 3 | 82 | 12.95 |
Rubens Campos Machado | 4 | 50 | 3.17 |
Roberto de Alencar Lotufo | 5 | 572 | 53.61 |