Japanese / English

Detail of Publication

Text Language Japanese
Authors Akira Horimatsu, Masakazu Iwamura, and Koichi Kise
Title Fast Affine Invariant Character Recognition Using Unique Determination of Feature Points
Journal Trans. IEICE
Vol. J98-D
No. 4
Pages pp.651-660
Reviewed or not Reviewed
Month & Year April 2015
Abstract Scene text captured with a digital camera or a smartphone suffers from various degradation. Thus, it is desired that a character recognition technique is robust to the degradation. In this paper, we focus on perspective distortion and propose a character recognition method that is fast, invariant to affine distortion and free from layout constraint. For affine invariant matching, geometric hashing is a well-known method for object recognition. However, it is computationally too expensive. Thus, we propose a method to greatly reduce the computational cost.
DOI 10.14923/transinfj.2014IUP0016
URL http://search.ieice.org/bin/summary.php?id=j98-d_4_651&category=D&year=2015&lang=J&abst
Back to list