글꼴 유사도 판단을 위한 Faster R-CNN 기반 한글 글꼴 획 요소 자동 추출

Abstract

Ever since media contents took over the world, the importance of typography has increased, and the influence of fonts has be n recognized. Nevertheles , the cur ent Hangul font system is very poor and is provided pas ively, so it is practical y impos ible to understand and utilize al the shape characteristics of more than six thousand Hangul fonts. In this paper, the characteristics of Hangul font shapes were selected based on the Hangul structure of similar fonts. The stroke element detection training was performed by fine tuning Faster R-CNN Inception v2, one of the de p learning object detection models. We also propose a system that automatical y extracts the stroke element characteristics from characters by introducing an automatic extraction algorithm. In comparison to the previous research which showed poor ac uracy while using SVM(Support Vector Machine) and Sliding Window Algorithm, the proposed system in this paper has shown the result of 10 % ac uracy to properly detect and extract stroke elements from various fonts. In conclusion, if the stroke element characteristics based on the Hangul structural information extracted through the system are used for similar clas ification, problems such as copyright wil be solved in an era when typography’s competitivenes becomes stronger, and an automated proces wil be provided to users for more convenience