Keypoint based comprehensive copy‐move forgery detection

Abstract

Verifying the authenticity of a digital image has been challenging problem. The simplest of the image tampering tricks is the copy-move forgery. In copy-move forgery copied portion of the image is pasted on another part of the same image. Geometrical transformations are used on the copied portions of the image before pasting it for the tampered image to look realistic and visually convincing. To make it more complex, other processing approaches may also be applied in the forged region for hiding traces of forgery. These processings are the scale, rotation, JPEG compression, and AWGN. In this paper, an approach based on features of the CenSurE keypoint detector and FREAK descriptor is proposed. This combination has novelty in itself as it has never been used for this purpose before to the best of authors' literature studies. CenSurE detectors are fast and give stable and accurate output even in the case of rotated images, which we club with binary descriptor FREAK. Hierarchical clustering and Neighbourhood search is applied in such a way that it can locate and detect multiple copy-move forgeries. The authors are hopeful that the proposed approach may be used in real-time image authentication and copy-move forgery detection.

Advancing Copy-Move Manipulation Detection in Complex Image Scenarios Through Multiscale Detector

“…In Figure 6, a comparative graph illustrates the results of the proposed method for JPEG compression, alongside existing approaches such as Zhu [26], Bi [39], Li [40], and Diwan [21]. The results of the proposed approach showcase superior performance when compared to recent published works.…”

“…The findings exhibit a consistent level of performance across various datasets. The approach we present demonstrates superior performance compared to the standard key-point-based approach [21], [39] across multiple datasets including GRIP, CMFD, CA-SIA II.0, CoMoFoD, COVERAGE, MICC-600, and MICC-220. However, the results in the GRIP dataset exhibit a modest decrease.…”

“…Keypoint-based: Key-point-based approaches are employed to extract and compare distinctive key-points to identify instances of image manipulation when a block-based approach fails to detect [7]. Anjali et al [16] In this paper, the SuperPoint detector for detecting copy-move manipulations in complex assaults is described. This approach employs a trained model with self-supervised learning and a quick detection time.…”

“…Smooth images may have fewer detectable keypoints than images with more textured regions. This may result in a decreased detection rate for keypoint-based techniques for smooth images [21].…”

Image copy-move forgery detection and localization based on super-BPD segmentation and DCNN

“…To reduce the computational complexity of block-based CMFD methods, the keypoint-based methods are proposed, using features of key points to replace that of blocks. The main key features are scale invariant feature transform (SIFT) 12 , speed-up robust feature (SURF) 1 , Harris 13 , accelerated-KAZE (A-KAZE) 14 , oriented FAST and rotated BRIEF (ORB) 15 , fast retina keypoint (FREAK) 16 , etc. However, most keypoints extraction methods extract few key points in the smooth regions, resulting in some forgeries in the smooth regions being ignored easily.…”