,
Coimbatore Institute of Technology , Coimbatore , India
,
Coimbatore Institute of Technology , Coimbatore , India
,
Coimbatore Institute of Technology , Coimbatore , India
,
Coimbatore Institute of Technology , Coimbatore , India
Coimbatore Institute of Technology , Coimbatore , India
As incidents of gendered violence and child abuse increase, there is a need for intelligent, real-time safety systems that autonomously detect distress and initiate emergency action without user activation. Q-SAFE (Quantum AI for Real-Time Women & Child Safety) is an open-source, multi-modal mobile app, utilizing quantum-enhanced machine learning models to detect panic situations based on audio cues (spoken and non-spoken) and motion. The system automatically generates an emergency broadcast that shares live location, messages alerts, and takes videos (in secure cloud storage) in the following ways: automatically detects predetermined distress words using existing Quantum /Indeterminate Natural Language Processing (NLP) classifiers, their corresponding visual module pending; and via accelerometer-based gesture recognition. Most importantly, due to employment of post-quantum cryptographic protocols, the emergency broadcasts use secure communication that does not depend on the users activating any manual processes. Q-SAFE departs from self-activated panic button approaches used in the past noting that fully autonomous systems that require no user activation have been absent for decades. Q-SAFE was developed using scalable technologies, while also conducting tests using multiple scenarios for future robustness, and emerges as an intelligent and privacy-respecting sentinel and caretaker for those that are most vulnerable.
This is an open access article distributed under the Creative Commons Attribution Non-Commercial License (CC BY-NC) License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
0
The statements, opinions and data contained in the journal are solely those of the individual authors and contributors and not of the publisher and the editor(s). We stay neutral with regard to jurisdictional claims in published maps and institutional affiliations.