MUMBAI, India, June 30 -- Intellectual Property India has published a patent application (202641054395 A) filed by Hindusthan College Of Engineering And Technology on April 29, 2026, for Cashguard:a Deep Learning System For Banknote Authentication And Counterfeit Detection.
Inventors include Dr. K. Kalaiselvi; and Shafni.
The application for the patent was published on June 26, 2026, under issue no. 26/2026.
Abstract: The current disclosure shows that it is smart and automated using deep learning and computer insight to verify the authenticity of banknotes. This system is known as CashGuard and is designed to discriminately detect both genuine and counterfeit money using the skills of computer-generated images of the imaging machines. The printing and reprographic technology has been invented and the money forging quality has been improved and even enhanced, the ways of identifying money are no longer effective. Manual inspection systems and the hardware systems may not be scalable, fast or even adapt to changes that do occur, in fact in the real world, which involves lighting, orientation and quality of images. The suggested model is the convolutional neural network (CNN), which is trained on a banknote image dataset. The model acquires complex visual templates, such as texture, micro structure and structure plots, which allow it to make reliable classifications in the process of acquisition. The pipeline in which the system is constructed is systematic in nature, in which an image acquisition phase, preprocessing, feature extraction, classification, and the generation of results are performed. It is real-time and can be integrated into banking systems, retail payment systems, and retail payment systems, and mobile apps. The invention will not only increase the accuracy of detection, reduce manual verification, but also provide scaling to the new generation of financial security challenges.
Disclaimer: Curated by HT Syndication.
