EN ENA new research, born from the collaboration between the Institute of Applied Physics “Nello Carrara” of the CNR, the Department of Information Engineering of the University of Florence and the CAMNES Study Center, has highlighted a new field of application for Artificial Intelligence.
The scholars of the University of Florence and CNR-Ifac have indeed developed an AI method to recognize and classify hieroglyphs. It is not the first time that AIs have been used to study languages no longer in use, but the authors of the study have managed to create for the first time a semi-automated tool, capable of autonomously reading and deciphering Egyptian symbols.
The Egyptian hieroglyphs are complex linguistic signs, composed of two elements: the semagram, that is, the graphic symbol that has a meaning related to the image itself, and the phonogram, that is, the phonetic value of the symbol. Given the complexity of the language and the variability of the supports on which they are found, it was necessary to find a method that was able not only to classify the individual ideograms, but also to support a more sophisticated system of recognition and translation of the documents available.
The solution identified is linked to “Deep Learning”, a branch of Machine Learning that involves the use of a set of algorithms based on DNN (deep neural networks), that is, deep neural networks. The scholars first tested CNNs (Convolutional Neural Networks) specialized in recognizing images and then dedicated themselves to the development of Glyphnet, a custom neural network that was able to identify and interpret hieroglyphs with greater accuracy.
The results were excellent: Glyphnet opens the doors not only to the automatic translation of ancient Egyptian sources but to the broader use of AIs in Egyptology.
The application of Deep Learning is gaining ground in more and more sectors, from automatic translations to medicine and pharmacological analyses, allowing real-time document interpretation, rapid tumor detection in CT scans, or predicting toxic effects of chemicals in food or medicines.
At Aidia we also use the Deep Learning approach in our solutions. If you want to learn more, write to info@aidia.it or fill out the form in the contact us section and schedule a free consultation with us.
source: A. Barucci, C. Cucci, M. Franci, M. Loschiavo and F. Argenti, “A Deep Learning Approach to Ancient Egyptian Hieroglyphs Classification,” in IEEE Access, vol. 9, pp. 123438-123447, 2021, doi: 10.1109/ACCESS.2021.3110082.
Marketing Specialist at AIDIA, graduated in International Studies in Florence, passionate about history, economics, and the bizarre things of the world.
At Aidia, we develop AI-based software solutions, NLP solutions, Big Data Analytics, and Data Science. Innovative solutions to optimize processes and streamline workflows. To learn more, contact us or send an email to info@aidia.it.
