A Prototypical Network-Based Approach for Low-Resource Font Typeface Feature Extraction and Utilization

Kangying Li,Akira Maeda,Biligsaikhan Batjargal

doi:10.3390/data6120134

Abstract

This paper introduces a framework for retrieving low-resource font typeface databases by handwritten input. A new deep learning model structure based on metric learning is proposed to extract the features of a character typeface and predict the category of handwrittten input queries. Rather than using sufficient training data, we aim to utilize ancient character font typefaces with only one sample per category. Our research aims to achieve decent retrieval performances over more than 600 categories of handwritten characters automatically. We consider utilizing generic handcrafted features to train a model to help the voting classifier make the final prediction. The proposed method is implemented on the ‘Shirakawa font oracle bone script’ dataset as an isolated ancient-character-recognition system based on free ordering and connective strokes. We evaluate the proposed model on several standard character and symbol datasets. The experimental results showed that the proposed method provides good performance in extracting the features of symbols or characters’ font images necessary to perform further retrieval tasks. The demo system has been released, and it requires only one sample for each character to predict the user input. The extracted features have a better effect in finding the highest-ranked relevant item in retrieval tasks and can also be utilized in various technical frameworks for ancient character recognition and can be applied to educational application development.

Highlights

The Greek philosopher Aristotle explained things, concepts, and symbols in his ‘OnInterpretation’
Our research considers the efficient use of font typeface image resources and implements an oracle bone script offline handwriting-recognition framework based on the oracle bone script in the ‘Shirakawa font’
The main contributions of this paper can be considered as follows: (1) From a technical aspect, we present a new model structure based on metric learning to use a low-resource ancient character typeface dataset

Summary

Introduction

The Greek philosopher Aristotle explained things, concepts, and symbols in his ‘On. Interpretation’. He emphasized the writing system and the complicated relationship among things, concepts, languages, and cultures. There are different ancient characters in the world. The ancient Egyptian hieroglyphs lost their meaning in the fourth century AD and became a mysterious writing system. Protecting existing records of language and writing systems has become the goal for humanities studies. If these scarce records can be archived, the system that provides retrieval or identification can be accessed publicly. More scholars can see clearer vectorized glyph resources, and more people will be able to taste these fading historical–cultural ambiances

Objectives

Methods

Results

Discussion

Conclusion