MUMBAI, India, Aug. 24 -- Intellectual Property India has published a patent application (202621085121 A) filed by Dkte Society'S Textile And Engineering Institute; Mrs. Smruti Alankar Pore; Ms. Triveni Ashok Dhekale; Ms. Vaishnavi Sanjay Khot; Ms. Rohini Dadaso Upadhye; and Ms. Vedika Rajendra Shinde on July 11, 2026, for Portable Ai-Based Ocr System For Recognizing Codes On Metal Components.
Inventors include Mrs. Smruti Alankar Pore; Ms. Triveni Ashok Dhekale; Ms. Vaishnavi Sanjay Khot; Ms. Rohini Dadaso Upadhye; and Ms. Vedika Rajendra Shinde.
The application for the patent was published on August 21, 2026, under issue no. 34/2026.
Abstract: ABSTRACT: Portable AI-Based OCR System for Recognizing Codes on Metal Components This invention describes a portable embedded artificial intelligence-based optical character recognition system for recognizing alphanumeric codes formed on metallic components. The system captures images of engraved, stamped, etched, impressed, laser-marked, printed, painted, or otherwise marked codes on reflective, curved, low-contrast, scratched, oily, oxidized, uneven, or industrially worn metal surfaces. Controlled illumination and reflection suppression enhance visibility of the marked region, while computer vision preprocessing performs noise reduction, contrast enhancement, thresholding, contour extraction, angular correction, skew compensation, and region-of-interest localization. Segmented character regions are processed using a trained deep learning recognition model to generate a predicted alphanumeric code. The recognized code is validated against predetermined rules or stored industrial records, stored with associated recognition data, and displayed in real time. The system improves metal-component traceability, reduces manual inspection dependency, and supports industrial quality control, inventory verification, dispatch inspection, and process automation.
Disclaimer: Curated by HT Syndication.
