Home  >  Article  >  Hardware Tutorial  >  AI almost doubles the number of known Nasca lines in just 6 months

AI almost doubles the number of known Nasca lines in just 6 months

Patricia Arquette
Patricia ArquetteOriginal
2024-09-29 06:03:30333browse

AI almost doubles the number of known Nasca lines in just 6 months

Nasca glyphs are ground images of people, animals or geometric shapes that were probably used by theNasca people for ceremonial and religious purposes. They can be found in the Nasca Desert in southern Peru, which stretches along the Pacific coast.

Until now, only 430 glyphs were known. Researchers had to carry out years of field studies and analyze aerial photographs to find them. A study published on September 23, 2024 at PNAS by Masato Saki, Akihisa Sakurai, Siyuan Lu and Marcus Freitag describes how the number of knownNasca lines was almost doubled within 6 months through the use of artificial intelligence.

AI almost doubles the number of known Nasca lines in just 6 months

The study used a neural network trained with natural photos, which was tuned to recognize relief-like Nasca glyphs.Given the limited number of training examples, the known geoglyphs were segmented into smaller parts to expand the training dataset. Using a gridded classification model, a detailed probability map with 5-meter resolution was generated, enabling the detection of even the most elusive geoglyphs. The conclusion of the study is:

The AI-assisted method is much more scalable, as evidenced by the number of new geoglyphs it has identified: It has assisted in detecting 303 new figurative relief-type geoglyphs, almost doubling the amount of known relief-type geoglyphs.

The enormous success that the use of AI has brought in the identification of Nasca lines provides a glimpse of what artificial intelligence could achieve in archaeological research and other scientific fields in the future.

The above is the detailed content of AI almost doubles the number of known Nasca lines in just 6 months. For more information, please follow other related articles on the PHP Chinese website!

Statement:
The content of this article is voluntarily contributed by netizens, and the copyright belongs to the original author. This site does not assume corresponding legal responsibility. If you find any content suspected of plagiarism or infringement, please contact admin@php.cn