科学家利用卫星和人工智能在秘鲁发现了143幅巨大的古代图画

科学家利用卫星和人工智能在秘鲁发现了143幅巨大的古代图画

Humanity has come a long way with its numerous discoveries throughout history. Everything from new locations to new species to new technologies, history is chock-full of notable findings and breakthroughs.

人类历史上有过无数的发现，走过了漫长的道路。从新地点到新物种再到新技术，历史上充满了显著的发现和突破。

However, the universe is infinitely vast, so the journey never stops and there’s always something new to find. In fact, you’d be surprised how tremendously little we know about our own planet. The fact that we’ve only explored around 5% of our oceans is proof of how much is still left to be discovered.

然而，宇宙是无限广阔的，所以旅程从未停止，总是有新的东西要发现。事实上，你会惊讶地发现我们对自己的星球知之甚少。事实上，我们只探索了大约5%的海洋，这证明了还有多少东西有待发现。

Scientists discovered over 140 new Nazca Lines, including this one—a bird

科学家们发现了140多条新的纳斯卡线条，包括这一条——一只鸟

Humanoid

人形

The Nazca Geoglyphs (also known as Nazca Lines) are a group of very large glyphs impressed into the soil of the Nazca Desert in southern Peru. These depict a variety of animals, people, plants, among other things. The glyphs date back to some time between 500BC and 500AD and is a designated UNESCO World Heritage Site.

纳斯卡地文字(也被称为纳斯卡线条)是一组非常大的文字，刻在秘鲁南部纳斯卡沙漠的土壤上。这些画描绘了各种各样的动物、人、植物等。这些象形文字的历史可以追溯到公元前500年到公元500年之间，是联合国教科文组织指定的世界遗产。

Fish

鱼

One of the more interesting facts about the Nazca Lines is that they are extremely huge. The individual figures can span anywhere from 0.4 to 1.1 kilometers across, amounting to approximately 1,300 kilometers of lines covering around a 50-square-kilometer area.

关于纳斯卡线条的一个更有趣的事实是，它们非常巨大。单个数字的跨度从0.4公里到1.1公里不等，相当于大约1300公里的线路覆盖了50平方公里的面积。

Human Neck

人类的脖子

Besides carrying out extensive fieldwork and other activities, Sakai’s team analyzed a number of high-resolution 3D data. They have also managed to discover one more glyph by developing an AI model based on the AI server IBM Power System AC922 configured with the deep learning platform IBM Watson Machine Learning Community Edition. This led to the first geoglyph discovery by an AI in history.

除了进行广泛的实地调查和其他活动，酒井法子的团队还分析了一些高分辨率的3D数据。他们还开发了一个基于IBM Power System AC922的人工智能模型，该模型配置了IBM Watson机器学习社区版的深度学习平台。这导致了历史上第一个人工智能发现的地理符号。

Fox & Bird

狐狸和鸟

A collaboration with IBM Japan will allow the science team to analyze and manage large complex data sets using AI. The new tech and efforts invested in the study will inevitably promote a greater understanding of the Nazca Lines in their entirety, and will boost the speed at which research of this heritage site is done.

与IBM日本的合作将允许科学团队使用人工智能分析和管理大型复杂数据集。在研究中投入的新技术和努力将不可避免地促进对纳斯卡线条的整体理解，并将加快对这一遗产遗址的研究速度。

Image credits: Yamagata University