AI software can automatically identify ancient cuneiform tablets, and researchers have made a breakthrough

News on November 24: A team from Martin Luther University Halle-Wittenberg (MLU), Johannes Gutenberg University Mainz and Mainz University of Applied Sciences developed a New artificial intelligence software can now decipher illegible cuneiform script. Unlike previous methods that relied on photos, this AI system utilizes 3D models of cuneiform tablets, delivering results that are more reliable than previous methods. This enables searches by comparing content across multiple slices, paving the way for entirely new research questions.

In the new study, the researchers used 3D models of nearly 2,000 cuneiform tablets, including about 50 from the MLU collection. It is estimated that there are about a million such tablets around the world, many of which are more than 5,000 years old and are among the oldest written texts in mankind. The films cover a wide range of topics, from shopping lists to court decisions, and shed light on humanity's past thousands of years ago. However, because these cuneiform tablets were unfired blocks of clay with words imprinted on them, they became extremely difficult to decipher, even to the trained eye

The content that needs to be rewritten is: Picture source note: This picture is generated by artificial intelligence and provided by the authorized service provider Midjourney

To solve this problem, Assistant Professor Hubert Mara of MLU proposed the idea of ​​developing an artificial intelligence system based on 3D models. The new system decrypted text better than previous methods. In principle, this AI system works similarly to optical character recognition (OCR) software, which converts words and images of text into machine-readable text. This has many advantages, once converted to computer text, the text can be more easily read or searched

Ernst Stötzner, member of the MLU, explains: "OCR is usually used for processing photographs or scans. With ink on paper or parchment, this is not a problem. However, with cuneiform tablets, the situation becomes It's even more complicated because light and viewing angle can greatly affect the recognition of certain characters." He developed this new AI system as part of his master's thesis. The team trained the new AI software using 3D scans and other data, much of it provided by Mainz University of Applied Sciences, which is responsible for the large-scale project of 3D models

The AI ​​system then successfully identified the symbols on the chip. The researchers were surprised to find that the system worked well even on essentially poor image material, such as photographs. The work of the researchers at Haller and Mainz provides new access to hitherto relatively exclusive material and opens many new research directions. Currently it is only a prototype capable of reliably identifying symbols in both languages, however a total of twelve cuneiform languages ​​are known to exist. In the future, the software may also help decipher damaged inscriptions, such as three-dimensional cuneiform in cemeteries.

The content that needs to be rewritten is: [Quote: Webmaster’s Home]

The above is the detailed content of AI software can automatically identify ancient cuneiform tablets, and researchers have made a breakthrough. For more information, please follow other related articles on the PHP Chinese website!