Numerical Simulation and Design of Copy Move Image Forgery Detection Using ORB and K Means Algorithm

Abstract

Copy-move is a common technique for tampering with images in the digital realm. Therefore, image security authentication is of critical importance in our society. So copy move forgery detection (CMFD) is activated in order to identify the forged portion of a photograph. A combination of the Scaled ORB and the k-means++ algorithm is used to identify this object. The first step is to identify the space on a pyramid scale, which is critical for the next step. A region's defining feature is critical to its detection. Because of this, the ORB descriptor plays an important role. Extracting FAST key points and ORB features from each scale space. The coordinates of the FAST key points have been reversed in relation to the original image. The ORB descriptors are now subjected to the k-means++ algorithm. Hammering distance is used to match the clustered features every two key points. Then, the forged key points are discovered. This information is used to draw two circles on the forged and original regions. Moment must be calculated if the forged region is rotational invariant. Geometric transformation (scaling and rotation) is possible in this method. For images that have been rotated and smoothed, this work demonstrates a method for detecting the forged region. The running time of the proposed method is less than that of the previous method.