A Study on the Variations in Chinese Character Libraries: A Case Study of Song Typeface

Abstract

Background: Documents play an essential role in societal operations, where ensuring their authenticity and integrity is paramount. Despite this, documents are susceptible to alterations, such as forgery or tampering, which can undermine their reliability. Forensic document examination is critical in detecting these modifications, thus safeguarding legal and procedural standards. Aims and Objectives: Conventionally, such examinations have concentrated on elements such as printing devices, document formatting, and print steganography, with limited attention given to the textual content itself. This study seeks to bridge that gap by focusing on the typeface variations in printed documents, using the Chinese Song typeface as a case study. Materials and Methods: For the first time, this research systematically explores the differences in typeface versions across printed documents, concentrating on a dataset of 3500 commonly used Chinese characters. We statistically analyzed and compared 29 different versions of four mainstream Chinese typefaces currently available on the market. Results: Leveraging Python-based tools, we replaced the traditional printing process with digital image rendering to conduct our analyses, employing both traditional overlay comparison methods and image differentiation techniques. Furthermore, we introduced an innovative metric to quantify character variation. Conclusion: Based on these findings, we identified a subset of frequently used characters with significant forensic examination value, offering pioneering insights into the practical application of typeface version analysis in document forensics.

Keywords

• character differences
• common characters
• difference value
• document examination
• printed documents
• typeface versions