We accept

Detection of Backup Move Forgery

J. Reethrose B. E. , Dr. J. P. Ananth M. E. , Ph. D. ,


Abstract-Digital images are easy to manipulate and revise using some editing and enhancing software. So that it is difficult to recognize the duplicate images. Copy-move manipulations are common form of local processing, where elements of a graphic are copied and reinserted into another area of the same image. The problem of detecting the copy-move forgery describes an efficient and reliable diagnosis and detects duplicate image parts. Most detection algorithm focused on pixel basis. Within this paper propose a new approach to detect forgery image such size, rotate, etc.

Keywords-copy-move forgery, SIFT, LSH, RANSAC.


Copy-move forgery is one of image tampering, were a part of the image is copied and pasted on another area of the same image. This copy-move forgery is easily done by some editing software such as Adobe Photoshop. Normally the human eye does not easily find out the copied region. The areas may be scaling or rotation type of manipulations. The purpose of copy-move forgery is detecting duplicate image areas. The most common image manipulation techniques involve the following

  • Removal of items from the image.
  • Addition of items in the image.
  • Change the items appearance in the image.

The most typical of these three manipulations is removal of undesired items from the image. Digital image forgery diagnosis techniques are labeled into effective and passive techniques. In active way, the digital image requires some pre-processing such as watermark embedding or personal generation during creating the image, which would limit of their application in practice. Furthermore, there are an incredible number of digital images in internet without digital personal or watermark. In such situation active approach cannot be used to get the authentication of the image. Unlike the watermark-based and signature-based methods; the passive technology does not need any digital personal produced or watermark inserted beforehand.

Fig 1. 1 Classification of Forgery detection techniques

  1. GENERAL Recognition PROCEDURE

Copy move manipulations lead to duplicate image locations, which practical forensic analyses take a look at in terms of powerful feature representations of parts of the image. Examining the image is vital before the preprocessing. After optional preprocessing (e. g. , color to grayscale change), the image is transformed to the feature space. Feature representation is finding the duplicate region. There are so many methods used to get the duplicate image such as DCT (Discrete Cousine Transform), DWT (Discrete Wavelet Transform), and PCD (Main Component Research). A set of feature vectors symbolizes local image characteristics and it is inspected for similarities in a coordinating procedure. That is achieved either by splitting the image into small blocks, which are then transformed individually, or by finding salient key points and extracting feature vectors founded thereon. The matching process is locating the similarity of duplicate image blocks. A number of the matching algorithms are k-d tree, Sorting, Nearest Neighbour Search, and Hashing. Similar feature vectors or their equivalent coordinates in the image airplane. Phony positives in the matching technique are pruned in your final error reduction step. The error decrease step is locating the duplicate image region.










Fig 2. 1 General copy move recognition pipeline


Accordingly, digital image forensics has emerged as a fresh research field that aims to uncover tampering functions in digital images. One common manipulation in tampering with digital images is known as region duplication, where a continuous portion of pixels is copied and pasted to another location in the same image. To make convincing forgeries, the duplicated areas are often created with geometrical or illumination changes. There are many method used in the existing system. DWT (Discrete Influx Transform) used to reduce dimensionality reduction. But it does not find the rotation and scaling. Lexicographic Sorting and Counting Bloom Filters are also found in the prevailing system. But it cannot find solution of scaling and rotation. It does not remove the noise. The Zernike instant is not hard way to get the duplicate (-rotate-) move forgery. This method is still weak against scaling or the other tempering based on Affine transform. Existing System gets the disadvantage of computational complexity and does not find correctness of the duplicate image parts.

In modern times, several methods have been suggested to find region duplication for the intended purpose of image forensics. These procedures derive from finding pixel blocks that are exact copies of one another in an image. Such methods are most effective for the diagnosis of region copy-move, where a region of pixels is pasted with no change to another location in the image. A form of digital tampering is Copy-Move forgery, in which a area of the image itself is copied and pasted into another area of the same image to conceal an important subject. As the copied part result from the same image, its important properties, such as noises, Shape, color and surface, will be compatible with all of those other image and thus could be more difficult to distinguish and detect.

In the preprocessing stage the RGB image is changed into grayscale image. Apply SIFT algorithm using to find the keypoints. SIFT Algorithm is employed to identify the keypoint localization. Good keypoints and features should signify distinct locations in an image, be productive to compute and solid to local geometrical distortion, sound, illumination versions and other degradations. Here, present SIFT features detection method to find the duplicate. Specifically, to find the locations, of potential duplicated parts, we first identify SIFT keypoints within an image. The found keypoints are matched up using hashing algorithm. We are able to use the matched SIFT keypoints to calculate the affine transform variables, but the obtained email address details are inaccurate because of the large numbers of mismatched keypoints. To learn the unreliable keypoints we use Random Test Consensus (RANSAC) algorithm then use the Affine transform. Finally find the duplicate region.

The pursuing diagram shows the way to find the backup move forgery. Organic image is considered as the forgery image. Normally the uncooked image is RGB image. That RGB image is converting into gray scale. This is actually the preprocessing stage. Noises removal also contains the preprocessing level. The steps involved with proposed method as follows.

  • First step to learn the keypoints using SIFT (Scale Invariant Feature Transform).
  • Find the keypoints then perform the complementing keypoints technique. Matching keypoints is using the Locality Private Hashing (LSH). Matching is not hard to discover the hash buckets. This hash is available the similar principles or keypoints.
  • Duplicate region is detected after corresponding. Find the duplicate region using the RANSAC (RANdom SAMple Consensus) algorithm.


Fig 3. 1 Stop diagram of forgery detection

A. Finding keypoints

In the preprocessing stage the RGB image is changed into grayscale image. Apply SIFT algorithm for finding the keypoints. SIFT algorithm contain the following levels:

  • Scale-space extrema detection
  • Keypoint localization
  • Orientation assignment
  • Generation of keypoint descriptors

Good keypoints and features should symbolize distinct locations in an image, be useful to compute and strong to local geometrical distortion, brightness variations, noises and other degradations. Here, to provide a fresh region duplication diagnosis methods predicated on the image SIFT features. Specifically, to detect the locations, of potential duplicated regions, first discover SIFT keypoints in an image. And compute the SIFT features for such keypoints. To guarantee the obtained feature vector invariant to rotation and scaling, the size of the neighborhood depends upon the dominant size of the keypoint, and all gradients within are aligned with the keypoints dominating orientation prominent orientation.

B. Matching keypoints

The similar keypoints are available out using Locality Sensitive Hashing (LSH) technique. Previous season a k-d tree algorithm used to identify the keypoint. That is taken more time search to compute the similar principles. Locality Private Hashing easy to detect the similar worth. Locality-sensitive hashing(LSH) is a way of executing probabilisticdimension reductionof high-dimensional data. The essential idea is tohashthe source items so that similar items are mapped to the same buckets with high probability (the number of buckets being much smaller than the universe of possible type items). This is different from the conventional hash functions, such as those used incryptographyas in cases like this the target is to maximize probability of "collision" of similar items rather than avoid collisions.

C. Duplicate Region

RANSAC algorithm used to detect the error. This implies SIFT produce the keypoints then Locality Private Hashing used to get the similar keypoints. Locality Private Hashing gets the bucket. Each bucket contains the index that index contain the prices of keypoints. RANSAC algorithm reduces the error. Instead of RANSAC using the Affine change. So it will easily to find out the mistake of size, rotation and change of backup move forgery recognition.


In particular the human eye does not easily find out the copied region. The areas may be scaling or rotation kind of manipulations. The goal of copy-move forgery is detecting duplicate image regions. Duplicate move forgery is difficult to identify the duplicate image region. SIFT can be used to identify the keypoints of given image. SIFT is Level Invariant Feature Transform. So that it focused to detect the Scale and transformation. Good keypoints and features should stand for distinct locations within an image, be effective to compute and strong to local geometrical distortion, brightness variations, sound and other degradations. Here, we present a new region duplication detection method based on the image SIFT features. Locality Private Hashing picks up the similar keypoints. Finally RANSAC algorithm used to get the duplicate image region.


[1] Rohini. R. Maind, Alka Khade, D. K. Chitre "Robust Image Backup move Forgery Detection" International Journal of Advanced and Innovative Research (IJAIR) ISSN: 2278-7844, Vol. 2, Concern 8, 2013.

[2] Yanjun Cao, Tiegang Gao, Li Fan, Qunting Yang "A solid recognition algorithm for copy-move forgery in digital images" Forensic Research International 214 (2012).

[3] Reza Oji "An Automatic Algorithm for Subject Recognition and Diagnosis PREDICATED ON ASIFT Keypoints" Signal & Image Processing: An International Journal (SIPIJ) Vol. 3, No. 5, October 2012.

[4] Pradyumna Deshpande, Prashasti Kanikar, "Pixel Founded Digital Image Forgery Detection Techniques" International Journal of Executive Research and Applications (IJERA) Vol-2, Concern 3, May-June 2012.

[5] B. L. Shivakumar, Dr. S. Santhosh Baboo, "Automated Forensic Way for Copy-Move Forgery Diagnosis predicated on Harris Interest Items and SIFT Descriptors" International Journal of Computer Applications (0975 - 8887) Quantity 27- No. 3, August 2011

[6] Xunyu Pan and Siwei Lyu, " Detecting Image Region Duplication Using Sift Features" IEEE, ICASSP, Dallas, USA 2010.

[7] Seung-Jin Ryu, Min-Jeong Lee, and Heung-Kyu Lee, "Detection of Copy-Rotate Move Forgery Using Zernike Moments" International Meeting on Information Concealing 2010.

[8] Saiqa Khan, Arun Kulkarni, "Reduced Time Complexity for Detection of Copy-Move Forgery Using Discrete Wavelet Transform" International Journal of Computer Applications (0975 - 8887) Level 6- No. 7, September 2010.

[9] Sevinc Bayram, Husrev Taha Sencar, Nasir Memon, "A COMPETENT and Robust Method for Detecting Copy-Move Forgery" International Discussion on Acoustics, Talk, and Signal Control - 2009.

[10] Tehseen Shahid, Atif Bin Mansoor "Copy-Move Forgery Diagnosis Algorithm for Digital Images and a fresh Reliability Metric" International Journal of Recent Fads in Executive, Vol 2, No. 2, November 2009.

[11] Aristides gionis, piote indyk, Rajeev motwani "Similarity search in high dimensions via hashing 1999.

[12] Prof. Unmukh Datta, Chetna Sharma "Examination of Copy-Move Image Forgery Detection" International Journal of Advanced Research in Computer Research and Electronics Engineering (IJARCSEE) Size 2, Issue 8, August 2013

[13] Frank Y. Shih and Yuan Yuan "AN EVALUATION Study on Copy-Cover Image Forgery Detection" The Open up Artificial Brains Journal, 2010, 4, 49-54 49

More than 7 000 students trust us to do their work
90% of customers place more than 5 orders with us
Special price $5 /page
Check the price
for your assignment