MUMBAI, India, July 7 -- Intellectual Property India has published a patent application (202641078355 A) filed by Podugu Devi Pradeep; Miracle Eductaional Society Group Of Institutions; Dr. Allu. Venkateswara Rao; and Dr. D. Madhavi on June 25, 2026, for An Adaptive Copy-Move Forgery Detection System In Digital Images Using Garic Deep Learning And Hybrid Heuristic Optimization.
Inventors include Dr. Allu Venkateswara Rao; and Dr. D. Madhavi.
The application for the patent was published on July 03, 2026, under issue no. 27/2026.
Abstract: The present disclosure relates to digital image forensics and, more particularly, to a system and method for adaptive copy-move forgery detection in digital images. The disclosed method receives an input image and performs pre-processing including grayscale conversion, noise reduction, and division of the image into overlapping blocks. The pre-processed image is further enhanced through deblurring using an mm'-type filter in combination with a Prewitt operator for edge-based gradient estimation. Feature extraction and feature matching are then performed using a Generalized Approximate Reasoning-Based Intelligence Control (GARIC) model comprising a neuro-fuzzy architecture configured to identify similar block patterns indicative of copy-move tampering. Similarity between feature vectors is evaluated using distance-based matching to detect duplicated regions. Post-processing is carried out using a hybrid optimization framework combining Artificial Bee Colony (ABC) optimization and African Buffalo Optimization (ABO) to eliminate false matches and refine localization of forged regions. The disclosed approach is applicable to digital forensic analysis and provides improved forgery detection performance under image manipulations including scaling, rotation, blurring, and related post- processing operations.
Disclaimer: Curated by HT Syndication.
