After uncovering a new Nasca Line formation with IBM Watson Machine Learning Accelerator on IBM Power Systems, Yamagata University will deploy IBM PAIRS in the hopes of further discoveries with AI
Archaeologists first discovered the mysterious Nasca Lines in southern Peru while traveling on foot in the late 1920s. Over nearly a century, new technologies such as satellite-based or drone-based remote hyperspectral sensing and imagery have helped researchers discover hundreds of these figures in an area covering about 500 square kilometers. Researchers from Japan’s Yamagata University alone have discovered more than 100 new geoglyphs since 2006. These scientists are now looking to work more efficiently and improve their ability to find and study new geoglyphs through the use of the IBM PAIRS platform and AI.
Created between 500 BC and 500 AD and best seen from an aerial view, the Nasca Lines depict shapes of varying complexity – from simple geometric shapes and plants to zoomorphic designs of animals — some several hundreds of meters in length, etched into the terrain. The exact purpose of the geoglyphs, designated a UNESCO World Heritage Site in 1994, is still unknown. Many theories have been surfaced about the reasons why the ancient Nasca cultures created them, from marking solstice points to offering art to deities in the sky. By uncovering more of these mysterious formations, archaeologists hope to piece together clues about their existence.
In the hopes of discovering new formations, better understanding the complexity and culture of the Nasca Lines, and raising awareness of their existence, IBM Research and Yamagata archaeologists are jointly deploying IBM PAIRS Geoscope, IBM’s cloud-based AI technology for scaling geospatial analytics to large and very complex data sets.
IBM made PAIRS widely available in February, and businesses have already begun to use PAIRS to help improve how multiple sources of data can be integrated to benefit large-scale operations. For example, it is now being used for crop identification and irrigation management, as well as monitoring vegetation growth around assets such as power lines to reduce the risk of outages.
For Yamagata’s research, PAIRS brings the unique ability to analyze massive and disparate geospatial and temporal datasets from a number of sources, including layers of LiDAR data — which is used to sense and examine the surface of the Earth — alongside drone images, satellite visuals and geographical survey information, to help reveal new lines and formations. Such integration is typically a difficult challenge given the scale and heterogeneity of these data sources. Using traditional approaches, it would require a significantly longer time to integrate these types of data volumes, potentially adding months to the discovery process. With PAIRS, these same tasks and analyses are expected to take minutes.
Prior to Yamagata’s decision to officially adopt and deploy PAIRS, researchers from the university and IBM spent the past few months investigating AI’s feasibility to help locate and better understand new formations. Uncovering new Nasca Line formations has historically been difficult due to the amount of “white noise” surrounding them, including roads and flood trails. The researchers’ initial goal was to explore if AI could help sift through tremendous amounts of data to pinpoint relevant clues that could lead to unearthing new figures.
To test this theory, IBM and Yamagata used IBM Watson Machine Learning Accelerator (WMLA) on IBM Power Systems to help researchers quickly analyze drone and satellite images to identify possible new geoglyphs. After training a deep neural network to identify Nasca Lines, the researchers fed the system more images to see if the AI could spot markings the researchers had missed. The process was a success, as the AI model discovered a humanoid-like figure in those images that researchers had been unable to detect, resulting in the uncovering of a new, undiscovered Nasca Lines formation.
New geoglyph Nasca lines, Peru
Working together, IBM Research and Yamagata look forward to integrating multi-modal, disparate and large data sets using PAIRS. Their hope is that by training AI and deep learning models on these volumes of unique and unstructured information, they will be able to gain valuable insights that could lead to greater discoveries and information surrounding the Nasca Lines.
The fourth-quarter issue of the IBM Journal of Research & Development is dedicated to the exploration and deployment of hardware for AI systems. It contains 10 contributions from leading authorities in the fields that summarize the latest state of the art and share new research results.
In a new study in the journal Nature, an IBM Research-led collaboration describes an exciting breakthrough in a 140-year-old mystery in physics — one that enables us to unlock the physical characteristics of semiconductors in much greater detail and aid in the development of new and improved semiconductor materials.