A new elastic image matching (EM) technique based on a category-dependent deformation model is proposed. In the deformation model, any deformation of a category is described by a linear combination of eigen-deformations, which are frequent deformation directions of the category and can be estimated statistically from the actual deformations. Experimental results on handwritten characters show that the proposed technique can attain higher recognition rates than conventional EM techniques based on the affine deformation model, which is a typical category-independent deformation model. The results also show the superiority of the proposed technique over those conventional EM techniques in computational efficiency.
All Science Journal Classification (ASJC) codes
- Theoretical Computer Science
- Information Systems
- Hardware and Architecture
- Computational Theory and Mathematics