Abstract | ||
---|---|---|
GPUs (Graphics Processing Units) are specialized microprocessors that accelerate 3D or 2D graphics operations. Recent GPUs, which have many processing units connected with a global memory, can be used for general purpose parallel computation. To utilize the powerful computing ability, GPUs are widely used for general purpose computing. The main purpose of this paper is an ellipse detection algorithm with Hough transform. The feature of our algorithm is that to reduce computational time and space, the parameter spaces in the Hough transform are decomposed for each parameter and each parameter is computed in series. Also, we implemented our algorithm on a modern GPU system. The experimental results show that, for an input image with size of 2040$\times$2040, our GPU implementation can achieve a speedup factor of approximately 64 times over the sequential implementation without the GPU support. |
Year | DOI | Venue |
---|---|---|
2011 | 10.1109/ICNC.2011.61 | Networking and Computing |
Keywords | Field | DocType |
fast ellipse detection algorithm,powerful computing ability,gpu implementation,general purpose parallel computation,gpu support,ellipse detection algorithm,general purpose computing,modern gpu system,main purpose,hough transform,parameter space,recent gpus,parallel computer,computational geometry | Graphics,2D computer graphics,Computer science,CUDA,Parallel computing,Computational geometry,Algorithm,Hough transform,General-purpose computing on graphics processing units,Ellipse,Speedup | Conference |
ISBN | Citations | PageRank |
978-1-4577-1796-3 | 18 | 1.17 |
References | Authors | |
9 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yasuaki Ito | 1 | 511 | 60.47 |
Kouhei Ogawa | 2 | 18 | 1.17 |
Koji Nakano | 3 | 1165 | 118.13 |