Finding the Nazca Lines Using an AI App

A team of archaeologists from Yamagata University, IBM Japan, and the Panthéon-Sorbonne University of Paris 1 used artificial intelligence applications to discover a previously unknown Nazca line. In their research archaeological journalthe group used a variety of data sources to teach AI apps.

The Nazca Lines are famous all over the world for their gigantic scale. The painting was created by the ancients by carving a hollow in the desert floor and can only be fully appreciated from an aircraft. They were made on a dry plateau that stretches for about 80 km in one direction and about 400 km in another.

Since its rediscovery by Spanish explorers, the geoglyphs have inspired many research efforts. Some attempt to understand what they mean and why they were made, while others simply try to find them all. It is unknown how many were made, so archaeologists and historians continue to look for more pieces. In this new effort, the researchers used artificial intelligence to assist in the search.

The research team noted that a large amount of data has been collected on geoglyph areas, including photographs taken from airplanes, drones, satellites, and LiDAR. The most useful were aerial photographs that mapped the entire region at a resolution of 0.1 meters per pixel.

The team trained an AI application on the geoglyph data and set it up to search for new geoglyphs from detailed photographs. They discovered four new geoglyphs of his. One looks like a man with a club, a pair of legs, a bird, and a fish. The human geoglyph was five meters long. The fish was 19 meters, the bird was 17 meters, and the legs were 78 meters.

The researchers suggest that the approach could be refined to find more geoglyphs in the area, or modified for use in other archaeological image retrieval applications.