"How good is good enough?" Establishing quality thresholds for the automatic text analysis of retro-digitized comics

Abstract

Stylometry in the form of simple statistical text analysis has proven to be a powerful tool for text classification, e.g. in the form of authorship attribution. When analyzing retro-digitized comics, manga and graphic novels, the researcher is confronted with the problem that automated text recognition (ATR) still leads to results that have comparatively high error rates, while the manual transcription of texts remains highly time-consuming. In this paper, we present an approach and measures that specify whether stylometry based on unsupervised ATR will produce reliable results for a given da-taset of comics images.

Computational Approaches to Comics Analysis

“…Training simple CNNs on relatively small data sets significantly increases recognition, with Dey et al () achieving an accuracy rate of 97% on hand‐drawn symbols. A different approach is taken in Hartel and Dunst (). Using Tesseract 4, the authors show that even imperfectly recognized comics texts can be used for simple text analysis based on a bag‐of‐words model and yield similar results to manually‐transcribed data.…”