Chris Fleizach Cse 260 Faster Connected Component Labeling
نویسنده
چکیده
The performance of parallelized connected component labeling can be improved in two ways: faster local labeling and faster interprocessor communication. Suzuki et al. describe one method of improving the speed of the local labeling procedure through the use of another table that remembered component labels. In speeding up global information dissemination, one may either allow each component to communicate with all it’s neighbors or centralize the effort in some manner. The goal of this project is to demonstrate the ability to speedup component labeling through different measures, while maintaining the fidelity of the labeling and eliminating bottlenecks.
منابع مشابه
Fleizach CSE 262 Readings : May 11 . 2006 Task Scheduling Strategies for Workflow based Applications in Grids
Many scientific computing tasks are too large to be processed quickly on one computer. Instead, the grid has become popular for a wide range of disciplines. Efficiently using the all the resources of the grid is a challenging, NP problem though. The authors introduced a new method for scheduling that accounted for data dependencies, a topic often ignored in job scheduling algorithms. They also ...
متن کاملCSE 262 Readings : May 9 . 2006 Multi - Step Processing of Spatial Joins
The spatial join combines spatial objects from a Cartesian product that satisfy certain geometric attributes. Unfortunately, to validate the operation the actual objects must be compared against each other. The authors discussed a new process for reducing the number of items that have to be retrieved from disk and compared through filtering and approximation. The three step pipeline started by ...
متن کاملTwo Strategies to Speed up Connected Component Labeling Algorithms
This paper presents two new strategies to speed up connected component labeling algorithms. The first strategy employs a decision tree to minimize the work performed in the scanning phase of connected component labeling algorithms. The second strategy uses a simplified union-find data structure to represent the equivalence information among the labels. For 8-connected components in a two-dimens...
متن کاملEfficient Connected Component Labeling on RISC Architectures
This article introduces two fast algorithms for Connected Component Labeling of binary images, a peculiar case of coloring. The first one, SelkowDT is pixel-based and a Selkow’s algorithm combined with the Decision Tree optimization technique. The second one called Light Speed Labeling is segment-based line-relative labeling and was especially thought for commodity RISC architectures. An extens...
متن کامل