Modified Binary Large Object (BLOB) Algorithm Integrated with Speeded Up Robust Feature (SURF) and Oriented FAST and Rotated BRIEF (ORB) Algorithm for Improvement Feature Detection

The increasing prevalence of digital and traditional media has heightened concerns about image authenticity, particularly in sensitive contexts like criminal investigations and media reporting. Digital image forgery, which involves deliberately manipulating images to distort or conceal information, presents significant challenges for fields such as journalism and law enforcement. As image editing tools become more accessible, detecting these forgeries has become increasingly complex. Various techniques are utilized for identification, including active methods like digital watermarks and passive methods such as pixel-based and geometry-based approaches. However, despite technological advancements, traditional blob algorithms still struggle with forgery detection, particularly when applied to large-scale datasets. This research aims to explore the integration of established image feature extraction techniques, specifically SURF, BLOB, and ORB, to enhance the effectiveness of forgery detection. The study reveals that combining these techniques leads to improved accuracy and processing speed, especially through the implementation of a modified Binary Large Object (BLOB) detector alongside advanced feature matching methods like FGM and FLANN. These findings not only address existing gaps in literature but also lay the groundwork for further advancements in the field of image forgery detection.