Green check
Link copied to clipboard

AI in Archaeology Paves the Way for New Discoveries

See how AI is helping archaeologists uncover hidden sites, restore ancient artifacts, and bring new insights into our understanding of ancient history.

Archaeology, the study of ancient history through the excavation and analysis of artifacts, often faces challenges such as vast areas to explore and the fragile nature of artifacts. Traditional methods can be time-consuming and even cause hidden sites or treasures to be skipped over. Artificial intelligence (AI) can step in to help with this. For instance, a recent breakthrough in the Yucatan Peninsula used deep learning to uncover over 60,000 previously unknown ancient Maya archeological sites that were hidden beneath dense vegetation.

In particular, computer vision can help archaeologists analyze images to detect artifacts and precisely map out ancient sites. Image analysis can help uncover hidden patterns in landscapes or even piece together broken artifacts to speed up discoveries and preserve even the smallest details of our history and culture. Archaeologists are unlocking new AI-driven ways to explore our past that are faster, more precise, and more insightful than ever before.

The Growth of AI-Driven Archaeology Techniques 

AI started making waves in archaeology towards the end of the 20th century, thanks to advancements in data analysis. By the 1990s, archaeologists began using machine learning and data mining to analyze the vast amounts of information they gathered. One of the first big successes was using AI to predict where archaeological sites might be based on historical and geographical data patterns, transforming how excavations were planned.

An early example of AI’s growing role in archaeology is the work of Iconem, a Paris-based company founded in 2010. Iconem uses drones and AI to create 3D digital models of historic landmarks threatened by war, conflict, and natural decay. By capturing thousands of images, Iconem has documented sites in over 20 countries, including the ancient ruins of Pompeii, Assyrian cities in Iraq, and Buddhist monasteries in Afghanistan. These digital preservations are valuable resources for education and research, especially for sites that are hard to reach or have been damaged.

Fig 1. Iconem’s 3D Model of Palmyra, Syria, Before It Was Destroyed in 2015.

By 2021, AI was even being used to restore ancient artifacts. Projects like the RePAIR initiative in Italy used robots to carefully piece together fragile items from places like Pompeii. Today, AI's role in archaeology continues to expand, helping in tasks like deciphering ancient texts and analyzing satellite images. AI has become an essential tool in uncovering and preserving our history.

How Computer Vision is Used in Archaeology

Computer vision is a branch of AI that enables computers to understand and interpret visual data. In archaeology, computer vision helps researchers analyze artifacts, map out ancient sites, and even reconstruct historical structures. By processing images and videos using deep learning, computer vision can extract insights that make preserving and interpreting archaeological findings easier.

Analyzing Artifacts With Computer Vision

Archaeologists often find thousands of fragments like pottery, bones, and other materials during digs. Traditionally, identifying and cataloging these pieces takes a lot of time and expert knowledge. However, with computer vision, this process can be automated. For instance, the Arch-I-Scan project at the University of Leicester uses image recognition and machine learning to identify and record details of pottery fragments automatically. The system can analyze photos from handheld devices and classify artifacts by size, shape, design, and texture.

Fig 2. Arch-I-Scan Can Help Recognise and Record Details of Pottery Remains.

Vision-based Mapping of Archaeological Sites

AI is especially useful in the early stages of archaeological projects. It can help identify potential archaeological sites before any digging begins and reduces the risk of damaging historical areas. By processing aerial or satellite images, computer vision models like Ultralytics YOLOv8 can use object detection to detect buried sites. YOLOv8 can be trained to detect subtle changes in a landscape that could hint at the presence of buried structures, such as ancient walls, roads, or buildings. In fact, slight variations in soil color, vegetation growth, or ground texture can even indicate underground remains.

Fig 3. Using AI to Identify Locations to Dig.

A good example of this is a case study in Portugal’s Alto Minho region. Researchers developed an AI model trained on annotated burial mounds using a technique called data augmentation, which expands the dataset by creating new similar training examples. The AI system can suggest potential new burial mound locations. Despite the challenge of false positives due to the aerial perspective of LiDAR data, this approach has achieved a 72.53% success rate after expert validation.

3D Reconstruction of Ancient Artifacts

Archaeologists often find broken pieces rather than intact artifacts during excavations. Traditionally, piecing these fragments together is a slow and meticulous task. However, AI-assisted 3D reconstruction is much quicker. Using digital 3D models, algorithms can analyze the shapes of fragments and accurately rebuild the original artifacts. This process saves time and provides precise measurements and a better understanding of the artifacts that can offer deeper insights into the cultures that created them.

Fig 4. 3D Reconstruction of Fragments.

AI in Underwater Archaeology

One exciting area of AI in exploring the past is underwater archaeology. In Baiae, a submerged Roman city off the coast of Italy, AI is helping to monitor and preserve the ancient ruins beneath the sea. WSense, a startup specializing in underwater communication, has developed a network of acoustic modems and sensors powered by AI. These systems adapt to changing sea conditions in real time and transmit environmental data. They make it possible to remotely monitor the site, help divers communicate and navigate underwater, and could even be available soon to tourists through interactive smart tablets.

Fig 5. Divers explore the underwater ruin at Baiae in Naples, Italy.

Using AI to Understand Ancient Texts

AI is also making strides in deciphering ancient texts. DeepMind's Ithaca, a cutting-edge tool, uses deep learning to restore damaged inscriptions, determine their origins, and date them accurately. Trained on a massive dataset of Greek inscriptions, Ithaca can restore texts with 62% accuracy and date them within 30 years of their true age. AI is redefining historical research by allowing scholars to reexamine key periods in Greek history with greater precision. When combined with human expertise, AI systems can significantly improve the accuracy of text restorations.

Fig 6. An example of a restored inscription that dates back to 485/4 BCE.

Pros and Cons of Using AI in Archaeology

AI offers many benefits to the field of archaeology and can help archaeologists uncover patterns and insights that might otherwise be missed. Here are some of the key benefits of AI in archaeology:

  • Improves accuracy in identifying and classifying artifacts.
  • Reduces human error, leading to more precise interpretations.
  • Ensures findings are based on solid data rather than personal biases.
  • AI helps create detailed records of artifacts that can be shared worldwide
  • Smart innovations make it easier for researchers to collaborate and preserve valuable cultural history for future generations

Despite the many benefits, there are also challenges related to relying on AI in archaeology. Here are some of the key challenges  of AI in archaeology:

  • Digital tools might overshadow traditional archaeological methods.
  • The hands-on skills and techniques developed over centuries could lose their importance.
  • Ethical concerns, such as protecting cultural heritage and indigenous rights, need careful consideration.
  • Overdependence on technology might reduce the critical thinking and creativity that human expertise adds to archaeology.

Overall, it's important to strike a balance that supports rather than replaces traditional archaeological practices. Open communication between AI specialists and archaeologists is essential to maintaining a balance.

What Factors Into Implementing AI in Archaeology?

Integrating AI into archaeology requires careful planning to make sure it complements rather than replaces traditional methods. To successfully bring AI tools into archaeological research, it's important to follow a few key steps. First, set clear objectives for using AI, whether to speed up data analysis, improve artifact classification, or assist in mapping sites. Having specific goals helps you choose the right AI tools for your research needs.

Then, while selecting AI tools, consider how well they fit into your archaeological goals. The tools should ideally be scalable, able to handle increasing amounts of data, and user-friendly so that archaeologists can use them effectively without needing extensive technical knowledge. It's also important to consider cost-effectiveness, as AI solutions can be expensive to implement and maintain.

Wrapping Up Our Findings

AI is changing the face of archaeology, making it quicker, more precise, and more accessible. By analyzing large amounts of data, uncovering hidden sites, and restoring ancient artifacts, AI is enhancing how we explore and understand the past. Its ability to swiftly and accurately process information allows archaeologists to make discoveries that once took years to achieve with traditional methods. As AI continues to evolve, its potential to reveal even more about our history is unlimited. The future of archaeology looks promising, with AI playing a key role in preserving our cultural heritage and deepening our understanding of ancient civilizations.

To learn more about AI, visit our GitHub repository and engage with our community. Explore AI applications in manufacturing and agriculture on our solutions pages. 🚀

Facebook logoTwitter logoLinkedIn logoCopy-link symbol

Read more in this category

Let’s build the future
of AI together!

Begin your journey with the future of machine learning