MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202641086667 A) filed by Sr University on July 15, 2026, for Real-Time Pedestrian Detection System.
Inventors include Deepa G; Dr. Johnson Kolluri; and Dr. Rahul Mishra.
The application for the patent was published on July 17, 2026, under issue no. 29/2026.
Abstract: ABSTRACT Disclosed herein is a real-time pedestrian detection system (100) that comprises an RGB camera unit (102), a depth sensor (104), an infrared sensor (106), and a processing unit (108). The processing unit (108) includes a multimodal data acquisition module (110), a data pre-processing module (112), an environmental condition analysis module (114), a sensor confidence estimation module (116), a feature extraction module (118), an explainable adaptive fusion engine (120), a pedestrian detection and localization module (122), an explainability generation module (124), a lightweight adaptive inference module (126), and an output generation module (128). The system (100) further comprises a memory unit (130), a communication interface (132), and a display/user interface (134). The system dynamically estimates sensor confidence based on environmental conditions, adaptively fuses multimodal features for pedestrian detection and localization, generates visual confidence maps and decision traces, and provides computationally efficient real-time pedestrian detection with explainability for diverse operating environments.
Disclaimer: Curated by HT Syndication.
