Reading Ancient Coin Legends: Object Recognition vs. OCR

TL;DR

This paper compares object recognition and OCR techniques for reading ancient coin legends, demonstrating that object recognition-based methods outperform standard OCR in recognizing coin inscriptions.

Contribution

It introduces a scene text recognition method tailored for ancient coin legends and empirically compares its performance to standard OCR.

Findings

Object recognition-based method outperforms OCR on coin legend recognition

Proposed method achieves higher accuracy in character and word recognition

Experiments conducted on a dataset of 180 cropped coin words

Abstract

Standard OCR is a well-researched topic of computer vision and can be considered solved for machine-printed text. However, when applied to unconstrained images, the recognition rates drop drastically. Therefore, the employment of object recognition-based techniques has become state of the art in scene text recognition applications. This paper presents a scene text recognition method tailored to ancient coin legends and compares the results achieved in character and word recognition experiments to a standard OCR engine. The conducted experiments show that the proposed method outperforms the standard OCR engine on a set of 180 cropped coin legend words.