Abstract
This paper presents a concurrent implementation of a previously developed Dual-Input Max-Tree algorithm that implements anti-extensive attribute filters based on second-generation connectivity. The paralellization strategy has been recently introduced for ordinary Max-Trees and involves the concurrent generation and filtering of several Max-Trees, one for each thread, that correspond to different segments of the input image. The algorithm uses a Union-Find type of labelling which allows for effcient merging of the trees. Tests on several 3D datasets using multi-core computers showed a speed-up of 4.14 to 4.21 on 4 threads running on the same number of cores. Maximum performance of 5.12 to 5.99 was achieved between 32 and 64 threads on 4 cores.
| Original language | English |
|---|---|
| Title of host publication | Mathematical Morphology and its Application to Signal and Image Processing |
| Subtitle of host publication | Proc. 8th International Symposium for Mathematical Morphology 2007 |
| Publisher | MCT/INPE |
| Pages | 449-460 |
| Number of pages | 12 |
| ISBN (Print) | 978-85-17-00032-4 |
| Publication status | Published - 2007 |
Keywords
- shared memory
- parallel computing
- attribute filter
- Dual-Input Max-Tree
- second-generation connectivity