Face classification using curvature-based multi-scale morphology
US7123783B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Jan 21, 2003 |
| Grant date | Oct 17, 2006 |
| Priority date | — |
| Expiry date | Jan 6, 2025 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06V40/168
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
An image classification system uses curvature-based multi-scale morphology to classify an image by its most distinguishing features. The image is recorded in digital form. Curvature features associated with the image are determined. A structuring element is modulated based on the curvature features. The shape of the structuring element is controlled by making it a function of both the scaling factor and the principal curvatures of the intensity surface of the face image. The structuring element modulated with the curvature features is superimposed on the image to determine a feature vector of the image using mathematical morphology. When this Curvature-based Multi-scale Morphology (CMM) technique is applied to face images, a high-dimensional feature vector is obtained. The dimensionality of this feature vector is reduced by using the PCA technique, and the low-dimensional feature vectors are analyzed using an Enhanced FLD Model (EFM) for superior classification performance.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.