AI to be used for producing rapid damage analysis of flooded areas.
When a flood hits a town or a village, satellites can capture images of the disaster, but when it comes to a quick assessment of the damage, there would be no better tool than artificial intelligence. It would help the relief workers to save more people in much less time. For the past 20 years, billions of people have been affected by disasters like floods. It is about time the governments stepped up their game in disaster management. On this ground, United Nations Institute for Training and Research (UNITAR) and the Operational Satellite Applications Programme (UNOSAT) have joined hands with UN Global Pulse to apply artificial intelligence for analyzing satellite imagery.
Quick response to disaster
The main motive behind the use of artificial intelligence in this method is to improve the response timeline of such disasters. Potential areas suffering from damage must be identified as soon as possible. To make the whole process of mapping completely automatic, the algorithms of machine learning would be pre-programmed to download the images of flooded areas. In no time, disaster maps would be produced as outputs to help the humanitarian counselors in providing aids to the affected people. Experts of UNITER have revealed that the system would also have the facility of live streaming the disaster maps.
Making of the AI model
To maintain a high level of accuracy, the AI model has been initially fed with manual flood maps of about 15 major historic floods, from UNOSAT’s extensive collection. Later, the model was tested in a previously unknown flooded area to check if it can work properly. The accuracy rate has turned out to be more than 97%. This brings in a lot of hope for a better future in disaster management. However, the model is yet to be implemented in real-life situations.
Share This Article
Do the sharing thingy
About Author
More info about author
Source: https://www.analyticsinsight.net/artificial-intelligence-is-set-to-save-more-people-from-floods/
