Monitoring war destruction from space using machine learning.
Hannes Mueller1,2, Andre Groeger3,4, Jonathan Hersh5
1Institute of Economic Analysis, Spanish National Research Council (CSIC), 08193 Bellaterra, Spain; andre.groger@uab.es h.mueller.uni@gmail.com.
Summary
This study introduces an automated deep-learning method to measure building destruction in conflict zones using satellite imagery. This approach provides more comprehensive and frequent data for conflict analysis and humanitarian efforts.
Area of Science:
- Remote Sensing
- Artificial Intelligence
- Conflict Studies
Background:
- Current methods for assessing building destruction in conflict zones (eyewitness reports, manual detection) are scarce, incomplete, and biased.
- Reliable destruction data is crucial for media, humanitarian aid, human rights monitoring, reconstruction, and academic research on violent conflict.
Purpose of the Study:
- To develop and validate an automated method for measuring building destruction using high-resolution satellite imagery.
- To overcome the limitations of existing data collection methods in conflict zones.
Main Methods:
- Utilized deep-learning techniques combined with label augmentation.
- Incorporated spatial and temporal smoothing to exploit data structure.
- Applied the method to assess destruction in Syrian cities during the civil war.
Main Results:
- Demonstrated an automated approach to generate destruction data with unprecedented scope, resolution, and frequency.
- Successfully reconstructed the evolution of damage in major Syrian cities.
Conclusions:
- The developed deep-learning method offers a significant advancement in monitoring conflict-induced destruction.
- This approach enhances the availability and reliability of data for various applications, including humanitarian response and academic research.
Related Concept Videos
01:30Design Example: Identifying the Locations of Monuments in the Field Using Global Positioning System Device
216
Surveyors use Global Positioning System (GPS) technology to measure the precise location and elevation of points on Earth. In a recent survey, GPS receivers were used to determine the coordinates and elevations of two park monuments. The process involved careful mission planning, data collection, and correction to ensure accuracy. The survey began with mission planning to identify optimal satellite visibility and minimize Position Dilution of Precision (PDOP). A geodetic control point...
216
01:10Detection of Black Holes
2.3K
Although black holes were theoretically postulated in the 1920s, they remained outside the domain of observational astronomy until the 1970s.
Their closest cousins are neutron stars, which are composed almost entirely of neutrons packed against each other, making them extremely dense. A neutron star has the same mass as the Sun but its diameter is only a few kilometers. Therefore, the escape velocity from their surface is close to the speed of light.
Not until the 1960s, when the first neutron...
Their closest cousins are neutron stars, which are composed almost entirely of neutrons packed against each other, making them extremely dense. A neutron star has the same mass as the Sun but its diameter is only a few kilometers. Therefore, the escape velocity from their surface is close to the speed of light.
Not until the 1960s, when the first neutron...
2.3K