Issuu

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395-0056

Volume: 10 Issue: 05 | May 2023

p-ISSN: 2395-0072

www.irjet.net

A Sustainable Decipher of Egyptian Hieroglyphs Karthik Variath Divakaran1, Shiyas Ahamed S1, Tushar Renji Thoonkuzhy1, Yadul Manoj1, Mr. Ajith S.2 1B.Tech student, Department of Computer Science and Engineering, Rajagiri School of Engineering & Technology,

Kerala, India

2Assistant Professor, Department of Computer Science and Engineering, Rajagiri School of Engineering &

Technology, Kerala, India ---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract - This paper presents various approaches for

Documentation has played a crucial part in the progress of civilization, allowing us to learn from the past and build upon the achievements of previous generations. Throughout history, documentation has taken several forms, ranging from stone tablets and papyrus scrolls to books and electronic media. In recent years, the proliferation of digital technology has brought about a new era of documentation, with huge amounts of data being created and stored digitally every day [17]. As we continue to produce and consume an ever-increasing amount of information, it is more important than ever to develop efficient documentation systems that can manage and maintain this knowledge for future generations. Our work is an important component of this effort, as it seeks to produce a robust and effective system for recognizing and translating ancient Egyptian hieroglyphs. By accurately documenting and translating hieroglyphs, our work can help preserve and share the knowledge and culture of one of the oldest civilizations in human history. This is especially crucial given the fragile nature of many hieroglyphic artifacts and the limited number of experts who are able to study and translate them [17]. By leveraging the recent advancement in machine learning and image recognition, our paper aims to make hieroglyphic translation more accessible and accurate than ever before, contributing to the broader goal of preserving and sharing knowledge across generations.

recognizing Egyptian hieroglyphs using machine learning techniques. Egyptian hieroglyphs represent a complex system of writing used in Ancient Egypt for over 3,000 years, which poses a substantial challenge for recognition due to the huge number of symbols and their variability in form and form. Documentation of the culture and language of different civilizations has been a critical component of human history, allowing us to preserve and communicate knowledge across time and space. Our system, which utilizes machine learning and image recognition to recognize and interpret ancient Egyptian hieroglyphs, is an important part of this ongoing movement to preserve and share knowledge for future generations. Our approach leverages the ability of neural networking to accurately recognize and classify hieroglyphs based on their visual features. Our system is able to recognize hieroglyphs from hand-drawn images/doodles of hieroglyphs and a live video feed. The proposed system is created entirely from scratch, with no usage of any pre-existing models or any pre-existing datasets. We trained and evaluated our model on a large dataset of manually created hieroglyphic images. We follow our own architecture of the neural network for training purposes. Experiments show that our model achieves a covetable accuracy for hieroglyph recognition. Our application has the potential to revolutionize the field of Egyptian hieroglyphs decipherment and make them more accessible to scholars and the general public.

Neural networks are a type of machine learning that has become increasingly popular in recent years. They are designed to mimic the way that the human mind works, and are especially effective at recognizing patterns in data [1][5]. This makes them ideal for image classification tasks, such as identifying hieroglyphs [6]. The use of neural networks for image classification has opened up new possibilities for understanding the hieroglyphic script and its significance. The study of hieroglyphs has a long and rich history. The hieroglyphic script is a complex system of writing that combines both phonetic and ideographic elements. The script consists of over 700 individual signs, each of which have its own significance and significance [7]. The script was first deciphered in the early 19th century by Jean-Francois Champollion, who used the Rosetta Stone to unlock the secret of the ancient script. Since then, scholars have made large strides in understanding the grammar and vocabulary of hieroglyphs. However, there is still much that

Key Words: Ancient Egyptian hieroglyphs, Image recognition and classification, Machine learning, Convolutional Neural Networks, JavaScript, p5.js, ml5.js.

1. INTRODUCTION Egyptian hieroglyphs are a system of writing that was used in ancient Egypt from about 3200 BCE until the end of the 4th century CE. The hieroglyphic script was used to compose a variety of texts, including administrative documents, religious texts, and literature [7]. Despite the script being ancient, there is still much to be learned about it, and advances in machine learning have created new opportunities for understanding and interpreting these symbols.

Impact Factor value: 8.226

ISO 9001:2008 Certified Journal

Page 827