MUMBAI, India, Oct. 12 -- Intellectual Property India has published a patent application (202641118540 A) filed by Vishnu Institute Of Technology on October 04, 2026, for Machine Learning-Based Detection Of Duplicated Regions In Tampered Digital Images.
Inventors include Mr. Kotta Kiran; Mrs. Ch Ch R S Mounika; Beedilli Ashirith; Emani Pallavi; Chellaboina Abhinav Kumar; and Bhukya Venkateswarlu Naik.
The application for the patent was published on October 09, 2026, under issue no. 41/2026.
Abstract: The present invention discloses a machine learning-based digital image forensic system for detecting and localizing duplicated regions in tampered images. The system receives a digital image, performs image preparation, and extracts statistical and colour-based features from image regions. Linear Discriminant Analysis (LDA) classifies image blocks into pristine and tampered samples. Speeded-Up Robust Features (SURF) are extracted and compared to identify duplicated regions based on feature correspondence. The extracted features are quantized into a Bag of Features representation and processed using a Support Vector Machine (SVM) classifier. A tampering possibility map is generated to provide localized identification of potentially forged image regions.
Disclaimer: Curated by HT Syndication.
