The U.S. Department of Defense is utilizing a visual computing artificial intelligence system called xView2 to assist with the ongoing disaster response efforts in the wake of the devastating earthquake in Turkey and Syria on February 6 which has caused tens of thousands of lives.
Developed in 2019 by the Pentagon's Defense Innovation Unit and Carnegie Mellon University's Software Engineering Institute, xView2 is an open source project still in its early development stages. However, it has already been deployed to help rescue missions in Turkey.
The AI system combines satellite imagery with machine learning to quickly identify building and infrastructure damage in the affected area and detect the severity of the damage.
Some specialists explain that the damage assessment process which normally takes days, weeks, or even months, can now be completed in a matter of hours or minutes, depending on the scale of the disaster, thanks to AI systems like xView2.
The algorithms of xView2 use a pixel-based object identification technique on satellite images of the region to determine the severity of the damage, which is represented by the darkness of the red color that indicates the condition of the debris. Moreover, AI systems such as xView2 may also reduce the need for eyewitness reports and denunciations.
Despite the advantages of xView2, there are some obstacles to its use, such as the need for clear satellite images and the inability to detect damage to the edges of buildings due to its aerial perspective.
Microsoft, the University of California, and Berkeley are among some of the institutions and organizations that are helping with the development of xView2.
Comments