Name
Abstract | ||
|---|---|---|
Radon transform is robust to noise, and its performance is independent on the calculation of pattern centroid. Based on these facts, Radon transform is employed to extract invariant features in this paper. Radon functional integral transform is proposed, and Radon functional integral descriptors are constructed. It is proved that these descriptors are invariants to translation, scaling and rotation. The performance of the proposed descriptors has been tested by some experiments on images from the coil-20 database. In comparison with some moment-based methods, the proposed method is more robust to noise. |
Year | DOI | Venue |
|---|---|---|
2016 | 10.1109/ICMLC.2016.7872984 | 2016 International Conference on Machine Learning and Cybernetics (ICMLC) |
Keywords | Field | DocType |
Radon functional integral transform,Radon functional integral descriptor,Radon transform,Feature extract | Pattern recognition,Radon,Robustness (computer science),Feature extraction,Invariant (mathematics),Artificial intelligence,Integral transform,Scaling,Radon transform,Centroid,Mathematics | Conference |
Volume | ISBN | Citations |
2 | 978-1-5090-0391-4 | 0 |
PageRank | References | Authors |
0.34 | 0 | 2 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yingkang Wang | 1 | 0 | 0.34 |
| Jianwei Yang | 2 | 58 | 12.73 |