Abstract | ||
---|---|---|
In this paper, we apply the 2-D real discrete chirplet transform to image compression. Instead of the pixel-based image representation, an image can be represented as only a few 2-D real discrete chirplet atoms if we use the 2-D real discrete chirplet transform based on matching pursuit. To reduce the computational complexity of the direct computation of 2-D real discrete chirplet transform, we develop a separable 2-D real discrete chirplet atom. Experimental results demonstrate that both a high compression ratio and a good quality of the reconstructed image can be achieved. Compared with JPEG, the chirplet transform has no blocking artificial effect at high compression ratio. |
Year | Venue | Keywords |
---|---|---|
2005 | Vision '05: Proceedings of the 2005 International Conference on Computer Vision | image compression |
Field | DocType | Citations |
Computer vision,Texture compression,Computer science,Artificial intelligence,Data compression,Image compression | Conference | 0 |
PageRank | References | Authors |
0.34 | 0 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jing Yang | 1 | 4 | 0.76 |
Hongxing Zou | 2 | 139 | 13.17 |