Abstract | ||
---|---|---|
The first step in a general 3-D vision system is the segmentation of a digital image into a number of regions that correspond to physical scene surfaces. To compensate for the difficulties associated with edge and region based segmentation methods, we present an integrated approach for range image segmentation. In the first stage we detect jump in the range image. The second stage, involves detecting fold edges using the absolute value of the residual and surface normals. We have used a Bayesian approach for the region growing part of our algorithm. Markov Random Filed is used to model the apriori knowledge information in the Bayesian formulation. A number of experimental results show that this approach is very effective in segmenting a wide variety of range images. |
Year | Venue | Keywords |
---|---|---|
1990 | MVA | bayesian approach,digital image,vision system,region growing |
Field | DocType | Citations |
Computer vision,Scale-space segmentation,Pattern recognition,Range segmentation,Image texture,Segmentation,Segmentation-based object categorization,Image segmentation,Artificial intelligence,Region growing,Mathematics,Minimum spanning tree-based segmentation | Conference | 1 |
PageRank | References | Authors |
0.36 | 3 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Arun K. Sood | 1 | 65 | 10.81 |
Ezzet Al-hujazi | 2 | 44 | 2.94 |