ReAssess (REcovery ASSESSment)
Category
Prototype
Team info

info Mohammadreza Sheykhmousa

info Farzaneh Firoozyar

We are looking for team members
solid knowledge of deep learning algorithms.

Contact us!
Share
Recent votes
Voting has finished

ReAssess (REcovery ASSESSment)

The challenge

Post-disaster recovery is a part of Disaster Risk Management (DRM) which has been neglected in disaster-related studies. Large natural disasters tend to receive huge financial and technological supports within the recovery process. Yet, there is no comprehensive method to assess and monitor the integrated recovery process, and as a result, the recovery output is not quantifiable. The conventional recovery assessment methods (i.e., household survey) are expensive, time-consuming, prone to subjectivity, and hard to communicate among stakeholders, while also lacking accuracy, transparency, and reliability. This survey types also are limited to (at maximum) urban areas, and they are not able to cover rural recovery assessment. Thus, there is a need to develop a method which is capable of quantifying the integrated recovery process over large areas.

The solution

Remote Sensing and satellite imagery has been proven to be a great tool in other domains of DRM. Disaster and subsequent recovery process affect lands mainly through Land Cover and Land Use (LCLU). Changes due to LCLU has been widely studied in other RS disciplines. However, the value of LCLUC in the recovery has not been explored yet. Therefore, to characterize and quantify the recovery process there is a need to develop an RS-based methodology which suits the special properties of the recovery process and image-data characteristics. In this study measuring recovery was done by developing recovery maps based on tracking specific trajectories of LCLU changes which are called Transition Patterns (TP). TPs provides a nuanced definition of recovery, and they characterized recovery process in five categories: Slightly Positive (SP), Positive (P), Neutral (N), Negative (N), Slightly Negative (SN). The results of this study provide a novel methodology which helps stakeholders involved in the recovery process to avoid an expensive recovery assessment and provide a quantifiable recovery assessment, which covers physical and functional recovery. The developed framework is generic meaning that can be applied in different disaster-stricken areas and for different disaster events.

Video's

This project is being coached by