MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202621028119 A) filed by Prof. Rupali Mahajan; Prajwal Vishnu Bahekar; Riddhi Sandeep Balapure; Purvesh Kishor Didpaye; Dikshitha Sadarla; and Aarya Thombare, Pune, Maharashtra, on March 10, for 'real-time low-power rescue detection system using spiking neural networks and event-based vision..'
Inventor(s) include Prof. Rupali Mahajan; Prajwal Vishnu Bahekar; Riddhi Sandeep Balapure; Purvesh Kishor Didpaye; Dikshitha Sadarla; and Aarya Thombare.
The application for the patent was published on May 1, under issue no. 18/2026.
According to the abstract released by the Intellectual Property India: "The present invention proposes a real-time low-power rescue detection system using event-based neuromorphic vision and Spiking Neural Networks. This system processes live video feeds and converts them into spike representation, which is then analyzed using a trained model of an SNN to detect human motion and gestures indicating distress. Compared to existing CNN-based computer vision techniques, the proposed system is more efficient in terms of processing power and power consumption since it only processes pixel changes over time. This system can be implemented on drones and robots to assist rescue operations in critical situations by detecting and alerting rescue teams about the presence of people in distress."
Disclaimer: Curated by HT Syndication.
