Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

HieroGlyphTranslator: Automatic Recognition and Translation of Egyptian Hieroglyphs to English

Ahmed Nasser, Marwan Mohamed, Alaa Sherif, Basmala Mahmoud, Shereen Yehia, Asmaa Saad, Mariam S. El-Rahmany, Ensaf H. Mohamed

TL;DR

The paper presents HieroGlyphTranslator, a four-stage pipeline that automatically recognizes Egyptian hieroglyphs in imagery, maps symbols to Gardiner codes, and translates them into English. It leverages a hybrid segmentation approach (Contour Detection + Detectron2), a ResNet50-based classifier, and a Transformer-based transliteration/translation module via OpenNMT, evaluated on the Morris Franken and EgyptianTranslation datasets. The model achieves a BLEU score of 42.22 for translation, outperforming a relevant baseline, and demonstrates strong classification performance, indicating a significant advancement in image-to-English hieroglyph translation. This work offers a practical, end-to-end tool that could enhance accessibility to hieroglyphic texts for scholars and the public.

Abstract

Egyptian hieroglyphs, the ancient Egyptian writing system, are composed entirely of drawings. Translating these glyphs into English poses various challenges, including the fact that a single glyph can have multiple meanings. Deep learning translation applications are evolving rapidly, producing remarkable results that significantly impact our lives. In this research, we propose a method for the automatic recognition and translation of ancient Egyptian hieroglyphs from images to English. This study utilized two datasets for classification and translation: the Morris Franken dataset and the EgyptianTranslation dataset. Our approach is divided into three stages: segmentation (using Contour and Detectron2), mapping symbols to Gardiner codes, and translation (using the CNN model). The model achieved a BLEU score of 42.2, a significant result compared to previous research.

HieroGlyphTranslator: Automatic Recognition and Translation of Egyptian Hieroglyphs to English

TL;DR

The paper presents HieroGlyphTranslator, a four-stage pipeline that automatically recognizes Egyptian hieroglyphs in imagery, maps symbols to Gardiner codes, and translates them into English. It leverages a hybrid segmentation approach (Contour Detection + Detectron2), a ResNet50-based classifier, and a Transformer-based transliteration/translation module via OpenNMT, evaluated on the Morris Franken and EgyptianTranslation datasets. The model achieves a BLEU score of 42.22 for translation, outperforming a relevant baseline, and demonstrates strong classification performance, indicating a significant advancement in image-to-English hieroglyph translation. This work offers a practical, end-to-end tool that could enhance accessibility to hieroglyphic texts for scholars and the public.

Abstract

Egyptian hieroglyphs, the ancient Egyptian writing system, are composed entirely of drawings. Translating these glyphs into English poses various challenges, including the fact that a single glyph can have multiple meanings. Deep learning translation applications are evolving rapidly, producing remarkable results that significantly impact our lives. In this research, we propose a method for the automatic recognition and translation of ancient Egyptian hieroglyphs from images to English. This study utilized two datasets for classification and translation: the Morris Franken dataset and the EgyptianTranslation dataset. Our approach is divided into three stages: segmentation (using Contour and Detectron2), mapping symbols to Gardiner codes, and translation (using the CNN model). The model achieved a BLEU score of 42.2, a significant result compared to previous research.

Paper Structure

This paper contains 15 sections, 4 equations, 12 figures, 8 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Materials and Methods
  4. Datasets
  5. The Proposed Pipeline Architecture
  6. Preprocessing
  7. Segmentation Module
  8. Classification Module
  9. Transliteration and Translation Module
  10. Results and Discussion
  11. Evaluation Metrics
  12. Experimental Results
  13. Classification Module
  14. Translation Module
  15. Conclusion

Figures (12)

  • Figure 1: Example of Egyptian literals Wiesenbach2019
  • Figure 2: The structure of an Egyptian hieroglyph: (a) Semagram (Owl) and phonogram /m/; (b) Semagram (Nose), determinative (nose, face, and associated actions), and phonogram; (c) Semagram (Mouth), phonogram /r/ and phonetic complement (R).
  • Figure 3: Example piece of Unas pyramids wall
  • Figure 4: Data distribution before and after augmentation
  • Figure 5: The proposed pipeline architecture
  • ...and 7 more figures