Region Duplication Forgery Detection Technique Based on SURF and HAC

Abstract

Region duplication forgery detection is a special type of forgery detection approach and widely used research topic under digital image forensics. In copy move forgery, a specific area is copied and then pasted into any other region of the image. Due to the availability of sophisticated image processing tools, it becomes very hard to detect forgery with naked eyes. From the forged region of an image no visual clues are often detected. For making the tampering more robust, various transformations like scaling, rotation, illumination changes, JPEG compression, noise addition, gamma correction, and blurring are applied. So there is a need for a method which performs efficiently in the presence of all such attacks. This paper presents a detection method based on speeded up robust features (SURF) and hierarchical agglomerative clustering (HAC). SURF detects the keypoints and their corresponding features. From these sets of keypoints, grouping is performed on the matched keypoints by HAC that shows copied and pasted regions.

Image Region Duplication Forgery Detection Based on Angular Radial Partitioning and Harris Key-Points

“…Our method reduces the false positive rate while still maintaining a high true positive rate. As seen from Table 4, the execution time for block based methods: Fridirch et al [14] and Popescu and Farid [39] are high compared with keypoint based methods: [18,37,38] and our method. The time required to detect forgery in our method is faster than Pan and Lyu's method [38] and Amerini et al [18].…”

“…Key-points-based techniques [18,22,37,38] and image regions based methods [14,39]. The true positive rates, false positive rates, Precision and Recall are evaluated on the tampered and authentic images from the …”

Keypoint Based Authentication And Localization Of Copy-Move Forgery In Digital Image

“…SIFT) have recently been proposed for locating copy-pasted areas [12,16]. Unlike block generation-based methods, keypoint-based methods consider features extracted from the entire image (i.e., without subdivision), and the extracted features are compared with one and another to locate similar keypoints [26].…”

“…Specifically, the method based on SIFT keypoints [15] and another based on SURF with hierarchical agglomerative clustering [12] are considered as the representative conventional methods. The results in terms of TPR and FPR are recorded in Table 4.…”

“…Although a number of copy-move forgery detection methods are already in existence [8,[12][13][14][15][16][17][18], the proposed detection method demonstrates an improvement in detection rate, and it is more robust against noise, rotation, scale, as well as JPEG compression. The proposed method, which is based on SIFT, can authenticate a digital image while locating the tampered areas, if there are any.…”