MUMBAI, India, Feb. 13 -- Intellectual Property India has published a patent application (202641007469 A) filed by Velammal Engineering College, Chennai, Tamil Nadu, on Jan. 26, for 'system and apparatus for image enhancement and multi-class recognition.'
Inventor(s) include Dr. K. Thilagam; Dr. S. Suganthi Amudhan; Dr. R. Vinodhini; and Mrs. M. Manimegalai.
The application for the patent was published on Feb. 13, under issue no. 07/2026.
According to the abstract released by the Intellectual Property India: "The present disclosure relates to a method for operating an AI camera platform, particularly for surveillance in low-light environments. The method involves capturing raw frames and exposure metadata via an image sensor module disposed within an environmental housing. A compute processor applies a low-light megapixel enhancement module to the raw frames, adapting parameters based on the exposure metadata and scene statistics to generate enhanced frames. Subsequently, the processor applies a multi-class recognition engine, which uses shared feature extraction with separate task heads, to the enhanced frames to output face, vehicle, and behaviour inferences. A unified data analytics pipeline aggregates these inferences to generate tracks and events. A communications interface outputs video along with the tracks and events, for use with a smart-city monitoring system. The principal use is in automated multi-class object recognition for intelligent surveillance systems."
Disclaimer: Curated by HT Syndication.
