학술논문
Optical character recognition based on phase features
Document Type
Conference
Author
Source
2015 International Conference on Computing Systems and Telematics (ICCSAT) Computing Systems and Telematics (ICCSAT), 2015 International Conference on. :1-5 Oct, 2015
Subject
Language
Abstract
A new method for recognition of content-less characters in low quality images using the phase congruency and local energy model is proposed. New phase features invariant to nonuniform illumination and slight geometric distortions are obtained. With the help of computer simulation the performance of the proposed method for character recognition in degraded images is presented and compared with that of common featurebased techniques in terms of recognition metrics.