Scribd
Upload
10 views1 page

AI Disaster Response System: SankatSuraksha

The paper introduces SankatSuraksha, an AI-based disaster monitoring and response system that utilizes advanced deep learning techniques to detect floods from satellite imagery. It features a Disaster Response Coordination and Monitoring System (DRCMS) with user-friendly portals for reporting and resource allocation, enhanced by a credibility scoring mechanism for prioritizing responses. The system shows improved accuracy and efficiency in disaster management and aims to expand to other hazards in the future.

Uploaded by

splksingh958410
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
10 views1 page

AI Disaster Response System: SankatSuraksha

The paper introduces SankatSuraksha, an AI-based disaster monitoring and response system that utilizes advanced deep learning techniques to detect floods from satellite imagery. It features a Disaster Response Coordination and Monitoring System (DRCMS) with user-friendly portals for reporting and resource allocation, enhanced by a credibility scoring mechanism for prioritizing responses. The system shows improved accuracy and efficiency in disaster management and aims to expand to other hazards in the future.

Uploaded by

splksingh958410
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Abstract: This paper presents SankatSuraksha, an AI-driven multi-hazard disaster

monitoring and response system, integrated within a scalable Disaster Response


Coordination and Monitoring System (DRCMS). SankatSuraksha uses a state-of-the-art
deep learning pipeline—incorporating U-Net and DeepLabV3+ architectures—to detect
floods from satellite imagery, leveraging public datasets such as Sentinel-1/2 for robust
segmentation. The DRCMS complements this monitoring with dynamic user and admin
portals that allow disaster reporting, volunteer and NGO registration, and interactive
dashboards. Central to the system is a novel credibility scoring mechanism that assigns trust,
severity, and urgency scores to reports and actors, enabling prioritized, data-driven resource
allocation and response. The platform integrates real-time satellite analytics with
crowdsourced data to enhance situational awareness and operational efficiency. A modular
Streamlit interface supports demo and deployment, featuring geospatial mask visualization
and export capabilities. Evaluation on synthetic and real-world flood data sets demonstrates
improved prediction accuracy and quicker, more reliable coordination compared to
conventional, manual approaches. Future work includes expansion to other hazards and
nationwide scalability.
Keywords: Disaster response coordination, Flood detection, Satellite imagery analysis, Deep
learning, Credibility scoring, Trust assessment, Resource allocation, Multi-hazard
monitoring, Crowdsourced reporting, Streamlit visualization

You might also like