Extending Gardiner’s code for Hieroglyphic recognition and English mapping

Abstract

“Knowledge is power”… Writing is the main way to preserve the humanity knowledge across the ages. Therefore, an automatic and accurate mapping from ancient scripts to modern live language is a must to have. Such a system will support knowledge transfer, tourism and education. This article presents a new algorithm to segment and recognize the ancient Egyptian Hieroglyphs from images and produce the corresponding English meaning. The algorithm used image processing along with Optical Character Recognition (OCR). Then, the meaning behind the image containing Hieroglyphs is interpreted based on the context of the mapped English sentence. Gardiner’s sign list is a standard list used to classify the Hieroglyphics symbols such that similar shapes are grouped in the same category. Hence, Hieroglyphics script is mapped to the English language. Hieroglyphics could be either read from left to right or from right to left based on the face orientation of the Hieroglyphic symbol. However, Gardiner’s code does not help to automate the reading direction of the Hieroglyphics script. In this work, an extension of the list is proposed to resolve the ambiguity in the reading direction. The results of the segmentation and recognition of Hieroglyphics symbols are demonstrated and compared to similar work for Chinese character recognition. Moreover, the results obtained from the state-of-the-art used in Hieroglyphic character recognition compared to the results obtained from the proposed algorithm on Hieroglyphic character are addressed. The mapped English sentence is then compared to some defined patterns. When a match is found, the input gets structured and reformatted accordingly. Tests on the defined patterns were conducted, and the results were successful, however, some additional results are generated that show equivalent synonyms to the input English sentence.