MUMBAI, India, May 29 -- Intellectual Property India has published a patent application (202641042267 A) filed by Srm University-Ap, Amaravati, Andhra Pradesh, on April 2, for 'an adaptive currency recognition system using multi-frame prediction and fragment-based analysis.'
Inventor(s) include Sanjay Kumar; Daksh Chaurasia; and Suhail Mohammad.
The application for the patent was published on May 29, under issue no. 22/2026.
According to the abstract released by the Intellectual Property India: "The present invention relates to an assistive artificial intelligence-based method and system (100) for identifying denominations of predefined banknotes and providing audio feedback to a user. The method comprises capturing image data of a banknote using an image capture unit (102), preprocessing the image using an image preprocessing module (104), and processing the preprocessed image using a deep learning recognition module (106) to extract visual features. The method further includes classifying the banknote into a denomination using a classification module (108), generating a confidence score using a confidence evaluation module (110), and providing voice-based output through an audio output module (112). The method additionally comprises aggregating predictions across consecutive frames using a multi-frame prediction module to obtain a stabilized result, and performing fragment-based analysis when the confidence score is below a threshold. The method enables reliable real-time denomination recognition under variable condition."
Disclaimer: Curated by HT Syndication.
