American University of Beirut: Protecting citizens' privacy while assessing damages after the Beirut Explosion

Challenge

After the Beirut Explosion in Summer 2020 thousands of buildings and public assets were destroyed or badly damaged. Due to the massive size of the affected area, damage assessment was tedious.

The American University of Beirut (AUB) started to support disaster relief operations. Thousands of high-resolution street-level images from affected areas were collected with vehicle-mounted 360° cameras, in order to better understand the effect of the blast on buildings, and to develop a rapid post-disaster damage assement solution using machine learning.

These images contained personal information of Beirut's citizens, which needed to be protected in order to comply with data protection laws. To achieve this, the AUB turned to Celantur.

Solution

After providing a first image sample set to Celantur, the desired anonymization quality was successfully verified. Subsequently the complete set consisting of thousands of images was uploaded to Celantur's cloud service.

The scalable cloud service blurred faces and license plates on the images in a matter of few hours. The American University of Beirut finally received their anonymized images and could start working on their RnD project in a privacy-compliant manner.

Celantur's contribution