MUMBAI, India, June 30 -- Intellectual Property India has published a patent application (202641073916 A) filed by Sona College Of Technology on June 15, 2026, for System And Method For Non-Contact Valuation Of Gold Items Using Image Processing And Machine Learning.
Inventors include V. Sudha; D. P. Sangeetha; D. Sharmiladevi; Afrah Fatima S; and Harsini M K.
The application for the patent was published on June 26, 2026, under issue no. 26/2026.
Abstract: ABSTRACT The present invention relates to a system and method for image-based estimation of weight, purity, authenticity, and monetary value of gold items using a consumer-grade imaging device such as a smartphone camera. The invention enables non-contact valuation of gold items without requiring physical weighing scales, XRF testing devices, density testing kits, or expert manual assessment. The system captures multiple guided images of a gold item from different angles and generates a three-dimensional model using photogrammetry to determine volumetric parameters. A purity estimation module analyzes colour, shine, reflectance, texture, hallmark markings, and wear indicators to estimate karat value and authenticity. Based on the estimated purity, a density value corresponding to the alloy composition is assigned, and the approximate weight of the gold item is determined. A market data integration module retrieves real-time gold market rates, and a valuation module calculates estimated monetary value by considering weight, purity, and market adjustment factors. FIGURE 1&2
Disclaimer: Curated by HT Syndication.
