Name
Abstract | ||
|---|---|---|
This paper presents a non-parametric method to extract a very short feature vector from the curvature function of a planar shape. Curvature is adaptively calculated using a new procedure that removes noise from the contour without missing relevant points. Then, its Fourier transform is projected onto a set of vectors, which have been chosen to be as representative as possible, to obtain the similarity between the input object and each vector of the set. These similarity values are the elements of the feature vector. The proposed method is very fast and classification has proven that the representation is good. |
Year | DOI | Venue |
|---|---|---|
2002 | 10.1016/S0031-3203(01)00041-3 | Pattern Recognition |
Keywords | Field | DocType |
Base projection,Adaptive curvature function,Vectorial subspace | Feature vector,Curvature,Pattern recognition,Fourier transform,Nonparametric statistics,Planar,Artificial intelligence,Mathematics | Journal |
Volume | Issue | ISSN |
35 | 1 | 0031-3203 |
Citations | PageRank | References |
23 | 0.96 | 16 |
Authors | ||
3 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| C. Urdiales | 1 | 251 | 33.14 |
| A. Bandera | 2 | 160 | 23.70 |
| Francisco Sandoval | 3 | 237 | 22.18 |