A New Geometric Descriptor for Symbols with Affine Deformations
Qi Jia,Xin Fan,Zhongxuan Luo,Yu Liu,He Guo
DOI: https://doi.org/10.1016/j.patrec.2013.11.003
IF: 4.757
2014-01-01
Pattern Recognition Letters
Abstract:Great attention has been devoted to the development of shape descriptors that is the key to object recognition. Previous works have great success on either relatively simple symbols or limited transformations, e.g., translation, rotation and scaling. We propose a new affine invariant, named characteristic ratio (CHAR) that includes more points for complex symbols with rich inner structures. Moreover, we build a novel shape descriptor with CHAR values calculated on collinear points that cover the convex hull of a shape. Dynamic Time Warping algorithm is employed to compare the similarity of spectrum. The performance of the proposed descriptor is validated by the experiments compared with the classical SIFT descriptor, Shape Context (SC), recently developed Cross Ratio Spectrum (CRS) and Circular Blurred Shape Model (CBSM) on three kinds of symbols, i.e., alphanumeric characters, television networks logos and traffic signs with a wide range of transformations (2016 images in total). The results indicate a high recognition rate to severe affine deformations, and a good discriminating ability to similar symbols. We also perform the experiments on the GREC database corrupted by noise with different degrees, showing the robustness of our descriptor to noise.