ISSN No.

: 2454- 2024 (online)

International Journal of Technical Research & Science

Digital Image Forgery and Techniques of

Forgery Detection: A brief review
Amandeep Kaur*, Jyoti Rani**
Email id: [email protected]
Dept. of Comp Sci. and Engg. , Giani Zail Singh PTU Campus Bathinda

Abstract-Digital images are all around us-from our mobile phones to the pages of online websites. Digital
images are used in almost every field whether it is information forensic, journalism, criminal and forensic
investigations or medical fields and many more. Because of the large availability and popularity of user-friendly
image editing tools and software it become easy to alter the images but such modified images pose some serious
dangers or problems in some fields where the genuineness of image has a prime important and in such fields it
become very difficult to verify the authenticity and probity of digital images. Digital image forgery is the
process of tampering contents of an image that is changing the meaning of image without leaving any detectable
clues. In this paper, we present a review of various types of digital image forgery and forgery detection
techniques.
Index Terms- Digital Image Forensics, Image Forgery, Forgery Detection

1. INTRODUCTION
Digital image forgery deals with digital image. The process of creating fake image has been tremendously
simple with the introduction of powerful computer graphics editing software such as Adobe Photoshop, GIMP,
and Corel Paint Shop, some of which are available for free. There are many cases of digital image forgery. All
of these cases can be categorized into three major groups, based on the process involved in creating the fake
image. The groups are Image Retouching, Image Splicing, and Copy-Move Attack, Morphing.

Image Retouching can be considered to be the less harmful kind of digital image forgery. Image retouching
does not significantly change an image, but instead, enhances or reduces certain feature of an image. This
technique is popular among magazine photo editors. It can be said that almost all magazine cover would employ
this technique to enhance certain features of an image so that it is more attractive; ignoring the fact that such
enhancement is ethically wrong. Fig. 1.1 shows an original image of lady’s face whereas fig. 1.2 shows the
same face with enhanced effects applied to it.

Fig. 1.1 (a) Original image Fig. 1.1 (b) Enhanced image
Image Splicing:
This technique is more aggressive than image retouching. Image Splicing is a technique that involves a
composite of two or more images which are combined to create a fake image. Fig. 1.2 shows a base image. Fig.
1.3 shows shark inside sea. From Fig. 1.3 region occupied by shark is copied and it is pasted below the
helicopter in the base image. This copy-paste operation from one image into another image forms a spliced
image as shown in fig. 1.4.

Fig. 1.2 Base Image Fig. 1.3 Shark Image Fig1.4Base image with shark

pg. 18
www.ijtrs.com
www.ijtrs.org
Paper Id: IJTRS-V1-I4-011 Volume 1 Issue 4, July 2016
 ISSN No.: 2454- 2024 (online)

International Journal of Technical Research & Science

Copy-move attack is more or less similar to Image Splicing in view of the fact that both techniques modify
certain image region (of a base image), with another image. However, instead of having an external image as the
source, copy-move attack uses portion of the original base image as its source. In other words, the source and
the destination of the modified image originated from the same image. In a copy-move attack, parts of the
original image is copied, moved to a desired location, and pasted. This is usually done in order to conceal
certain details or to duplicate certain aspects of an image. Blurring is usually applied along the border of the
modified region to reduce the effect of irregularities between the original and pasted region. Fig. 1.5 shows
original image of a garden view. In Fig. 1.6 a region occupied by a deer is copied and pasted on the grass at
front side in the same view.

Fig. 1.5 Original Image Fig. 1.6 Forged image

Morphing: It is a special effect in motion pictures and animations that changes one image or shape into another
through a seamless transition. Most often it is used to depict one person turning into another through
technological means or as part of a fantasy or surreal sequence.

Fig. 1.7 (a) Original Fig. 1.7 (b) Forged Fig. 1.7 (c) Original

2. DIGITAL IMAGE FORENSICS

Digital Image Forensics is a quite recent discipline; nonetheless, it is tightly connected with a number of
different research fields. DIF inherits its goals and attitude from classical (analog) forensic science and from the
more recent field of computer forensics. Forensic disciplines in general aim at exposing evidence of crimes; to
do so, they have to deal with the burglars’ ability in either hiding or possibly counterfeiting their traces. In
digital imaging both the acquisition process and the tampering techniques are likely to leave subtle traces. The
task of forensics experts is to expose these traces by exploiting existing knowledge on digital imaging
mechanisms, being aided by consolidated results in multimedia security research.
2.1 Applications of Digital Image Forensics
Digital forensics is commonly used in both criminal law and private investigation, Forensic analysis the images
on online social networks, Used for detecting tampered or forged image, Image forgery detection system is
needed in many fields for protecting copyright and preventing forgery or alteration of images. It is
applied in areas such as journalism, scientific publications, digital forensic science, multimedia security,
surveillance systems etc.
The copy-move forgery is one of the difficult forgeries to detect in image processing. It is common image
tampering technique used now a day. In this some part of the image needs to be covered to add or remove
information of an image.

pg. 19
www.ijtrs.com
www.ijtrs.org
Paper Id: IJTRS-V1-I4-011 Volume 1 Issue 4, July 2016
 ISSN No.: 2454- 2024 (online)

International Journal of Technical Research & Science

2.2 Approaches to detect Digital Image Forgery
There are two approaches for detecting digital image forgery. One is active approach and the other is passive
approach.
2.3 Active Approach
An active detection method which consists of adding image details in order to describe digital tampering such as
name, date, signature, etc. It requires a special hardware implementation to mark the authentication of the
digital image.
2.4 Passive Approach

Passive method detects the duplicated objects in forged images without need of original image
watermark and depends on traces left on the image by different processing steps during image
manipulation. Passive approach also determines the amount and the location of forgery in the image.

Digital Image Forgery Detection

Active Approach Passive Approach

Digital
Image Image
Watermarking Sign/Hash
Statistics Contents
code

Fig. 2.1 Image Forgery Detection Approaches

3. LITERATURE SURVEY
I Amerini, et al. (2014) proposed a system which could evaluate the effectiveness of the attacking methods
from the side of perceptual image quality; a new version of a SIFT key point removal method based on a
perceptual metric. The author explained the criterion for the choice of the quality metric q(·) and then a
comparison between the proposed system and Counter-forensics of SIFT-based copy-move detection by means
of key point classification , both in terms of key point removal and in terms of final perceptual quality, is
presented. Author has demonstrated that the proposed method obtains the lowest possible impact on visual
quality with respect to the methods presented so far still achieving to remove a relevant number of key points.
Tu K. Huynh, et al.( 2015) presented a survey on Image Forgery Detection (IFD) techniques applied for
both Copy-Move and spliced images. The author has classified the algorithms on the basis of processing input
images with or without transformation before extracting the image features for the copy- move images. For
the spliced images, groups of detection techniques are based on image features or camera features.
Reducing the complexity, increasing the detection rates, researching faster algorithms or building the large
database to test is concluded.
Sushama Kishor Bhandare, et al.(2015) presented a review of the forensic methods for detecting globally and
locally applied contrast enhancement, cut-and-paste forgery, histogram equalization, and noise in the digital
image .Author has concluded that the techniques that are robust against the post processing operations and
anti forensic techniques need to be developed.
Nandini Singhal, et al. (2015) presented a review of techniques for pixel based forgery detection. Two
techniques presented in the paper are copy-move or cloning and fast-copy move detection. In copy-move or
cloning technique a part of the image is copied and pasted into another part of the image which has
limitation of only shifting of copied regions. In order to overcome this limitation another technique fast-
pg. 20
www.ijtrs.com
www.ijtrs.org
Paper Id: IJTRS-V1-I4-011 Volume 1 Issue 4, July 2016
 ISSN No.: 2454- 2024 (online)

International Journal of Technical Research & Science

copy move detection having complex but precise results is discussed. But its main disadvantage is that it is not
able to detect for very small region. Author has discussed a special type of forgery detection which can
detect the duplicated regions accurately and quickly.
Harpreet Kaur, et al. (2015) presented different techniques to detect copy move forgery using block based
method. Author has presented limitations of different techniques used for passive method to detect copy
move forgery. Author has concluded that the comparative work can be extended by proposing a novel technique
with which the existing limitations can be overcome.
Mohammad Farukh Hashmi, et al. (2015) presented an approach for image forgery authentication. Author has
stated that a non morphed and non forged image shows homogeneity in non spectral domain. This
homogeneity is lost when any forgery or morphing is applied on the images. Author proposed a system by
applying a set of transform on the images. DCT statistics, LBP features with curvelet statistics and Gabor
transform of the images has been combined to represent an image in the transformed domain. CASIA
image dataset with seven thousand authentic and same numbers of tempered images is used to verify the
technique. Dataset is divided into two equal halves to perform training and testing. Transformed images are used
to train Hidden Markov model as HMM can extract probabilistic state information from a large statistical model.
Test images are tested in transformed domain by HMM with log likelihood estimator. In case HMM
returns an indeterminist result or multiple subset of result, the transformed test image was tested with two class
SVM classifier with RBF kernel. The system shows the accuracy of over 89% for 500 test instances. Sensitivity
and Specificity were found to be 90% and 88% respectively.
Snigdha K. Mankar, et al. (2015) studied various techniques like the SVM classifier, Pixel-based and
partition-based to detect forgery of images. Author has concluded that multimedia authentication techniques
have emerged to verify content integrity and prevent forgery of images.
Jessica Fridrich, et al proposed a system to detect malicious manipulation with digital images. The proposed
system, implemented in C, may successfully detect the forged part even when the copied area is enhanced/
retouched to merge it with the background and when the forged image is saved in a lossy format, such as JPEG.
The performance of the proposed method is demonstrated on several forged images.

4. COPY-MOVE FORGERY DETECTION TECHNIQUES

A number of techniques are available to detect copy-move forgery which can be classified into two main
categories such as block-based and key point-based methods.
4.1 Blocks-based methods
In block-based methods, the image will be divided into overlapping blocks of specified size and a feature vector
will be computed for these blocks. Similar feature vectors are then matched to find the forged regions. For e.g.
DCT, DWT, PCA, KPCA etc.
4.2 Features matching-based methods
In this category, feature vectors are computed for regions with high entropy. There is no subdivision into blocks.
The feature vectors are matched to find the copied blocks. The well-known key-point detectors are Harris, SIFT,
SURF and FAST. For instance, the Scale Invariant Features Transform (SIFT), which is invariant to
illumination, scaling, rotation and JPEG compression.

Fig. 4.1 Example of copy-move attack on images

pg. 21
www.ijtrs.com
www.ijtrs.org
Paper Id: IJTRS-V1-I4-011 Volume 1 Issue 4, July 2016
 ISSN No.: 2454- 2024 (online)

International Journal of Technical Research & Science

General process of image forgery detection includes the following steps:

Input digital image

Dividing into
overlapping blocks

PCA,DCT,DWT,
Feature extraction FMT,SVD,SIFT,S
URF

Lexicographically
sorting

Locate forged
region

Detection result
Fig. 4.2 Block diagram of copy-move image forgery detection system

PCA: Principal Component analysis, DCT: Discrete Cosine transform, DWT: Discrete Wavelet transform,
SVD: Singular Value decomposition, SIFT: Scale Invariant Feature transform, SURF: Speeded up Robust
features.

Principle Component Analysis (PCA) is a candidate to extract the image features [9]. In the Dartmouth
Computer Science Technical Report of 2004, Alin C Popescu and Hany Farid used PCA to automatically detect
duplicated regions in a digital image. The technique works by first applying a principal component analysis to
small fixed-size image blocks to yield a reduced dimension representation. The representation is robust to minor
variations in the image due to additive noise or lossy compression. Duplicated regions are then detected by
lexicographically sorting all of the image blocks. The efficiency of the proposed technique on credible forgeries,
and its robustness and sensitivity to additive noise and lossy JPEG compression has been shown by the author.
The procedure to produce each feature vector is called principle component analysis in which values are
obtained by using the theorems of covariance matrix, eigenvectors and linear basis for each image block with
the initial conditions of zero-mean. Then a matrix S of block vectors quantized according to number of
quantization bins to reduce the mirror variations created. These quantization coefficients are then sorted
lexicographically and the duplicated regions has been detected by considering the offset of all pairs whose
distances in S less than a specific threshold. To obtain the efficient results, a duplication map is defined by
producing a zero image of the same size as original and assigning all pixels in a duplicated region to a
unique grayscale value. With the dimension of the PCA reduced representation and total number of image
pixels are Nt and N respectively, the algorithm has complexity of O(NtNLogN).
4.3. DCT, DWT
Discrete Cosine Transform (DCT) and Discrete Wavelet Transform (DWT) are popular techniques to
transform an input image to the frequency domain before extracting the image features.
Jessica Fridrich, et.al [8] used quantized Discrete Cosine Transform (DCT).The feature vectors are vectors of
quantized DCT coefficients. The quality factor in JPEG compression determines the quantization step for
DCT transform coefficients which is called the user-specified parameter Q. The mutual positions are
considered in case of too many matching blocks having the same shift vector to define a specific block
pair. For the color images, the algorithm requires a color to grayscale conversion. It takes MNlog2(MN)
steps in exact match.
Discrete Wavelet Transform (DWT) is always the potential candidate for research on CMFD. In fact,
many proposed algorithms to detect Copy-Move regions using DWT coefficients. Nandini Singhal, et. al [4]
proposed a system using DWT to detect pixel-based forgery. A forged image is taken as an input image. DWT is
applied to the input image to yield LL1 sub-band. The LL1 sub-band is divided into sub-images. Then phase
correlation is calculated. The offset between the copy-move regions is also calculated. The copy-move region is

pg. 22
www.ijtrs.com
www.ijtrs.org
Paper Id: IJTRS-V1-I4-011 Volume 1 Issue 4, July 2016
 ISSN No.: 2454- 2024 (online)

International Journal of Technical Research & Science

found out by pixel matching. Then MMO (Mathematical Morphological Operations) is applied to detect the
result.
4.4 SIFT, SURF
Scale Invariant Features Transform (SIFT)presents a method for extracting distinctive invariant features from
images that can be used to perform reliable matching between different views of an object or scene in given
image. The features are invariant to different scale and rotation, and provide robust matching across a large
range of affine transformation, distortion, change in 3D viewpoint, addition of noise, and change in illumination.
The features are highly distinctive and a single feature can be correctly matched with high probability against a
large database of features of image and video. SURF is a scale and rotation invariant interest point detector and
descriptor. It can be computed and compared much faster than other image features like SIFT and HOG.
When some object is copy-moved with the help of geometrical and illumination transform, it becomes difficult
to detect that object. Speed up Robust Feature (SURF) and Scale Invariant Feature Transform (SIFT) are
invariant with respect to geometrical and illumination transform. Ramesh Chand Pandey, et. Al (2014) [10]
proposed a method to detect copy move forgery in an image based on passive forensic scheme. The proposed
method used SURF and SIFT, which make it very fast and robust in detecting copy-moved regions. To achieve
very fast speed in copy-move forgery detection, SURF image features are used to find the image key-points
(interest points) and extract their64 dimensional descriptor which is used for rapid matching. The purposed
system reads an image. Image key-points / interest points are detected via SURF key-point detector. The
descriptors for the key-points are computed using SURF key-point descriptor. The best ten matches are
identified for every key-point. g2NN matching (g2NN-generalized 2 nearest neighbor) is applied. The dynamic
thresholding step is performed. The matched key-point are joined using a line to represent the copied region.

CONCLUSION
The paper surveys the different types of digital image forgery, approaches to detect digital forgery. Specifically
pixel-based forgery detection techniques are discussed. All the methods and approaches discussed in this paper
are able to detect forgery. But some algorithms are not effective in terms of detecting actual forged region. On
the other hand some algorithms have a very high time complexity. So, there is a need to develop efficient and
accurate image forgery detection algorithm, either by combining the existing techniques or by developing new
techniques.

REFERENCES
[1] I Amerini, F. Battisti, R. Caldelli, M. Carli, A. Costanzo, “Exploiting Perceptual Quality Issues In
Countering SIFT-Based Forensic Methods”, IEEE International Conference on Acoustic, Speech and Signal
Processing (ICASSP), p. 2664-2668, 2014.
[2] Tu K. Huynh, Thuong Le-Tien, KhoaV.Huynh, SyC.Nguyen, “A Survey on Image Forgery Detection
Techniques”,The 2015 IEEE RIVF International Conference on Computing & Communication
Technologies Research, Innovation, and Vision for Future (RIVF), p. 71-76, 2015.
[3] Sushama Kishor Bhandare, Nitin Krishnarao Bhil, “Digital Image Forensic”, International Journal of
Advance Research in Computer Science and Software Engineering, p. 839-842, 2015.
[4] Nandini Singhal, Savita Gandhani, “Analysis of Copy-move Forgery Image Forensics: A Review”,
International Journal of Signal Processing, Image Processing and Pattern Recognition Vol.8, No.7, pp.265-
272, 2015.
[5] HarpreetKaur, KamaljitKaur, “A Brief Survey of Different Techniques for Detecting Copy-Move Forgery”,
International Journal of Advanced Research in Computer Science and Software Engineering Volume 5,
Issue 4, p. 875-882, 2015
[6] Mohammad FarukhHashmi, Avinash G. Keskar, “Image Forgery Authentication and Classification using
Hybridization of HMM and SVM Classifier”, International Journal of Security and Its Applications Vol. 9,
No. 4, pp. 125-140, 2015.
[7] Snigdha K. Mankar, Prof. Dr. Ajay A. Gurjar, “Image Forgery Types and Their Detection: A Review”,
International Journal of Advanced Research in Computer Science and Software Engineering, Volume 5,
Issue 4, April 2015.
[8] Jessica Fridrich, David Soukal, and Jan Lukáš, “Detection of Copy Move Forgery in Digital Images”,
Digital Forensic Research Workshop, Cleveland, Ohio, USA, 2003
[9] Alin C Popescu and HanyFarid, “Exposing Digital Forgeries by Detecting Duplicated Image
Regions”, Dartmouth Computer Science Technical Report TR2004-515, USA, August 2004.

pg. 23
www.ijtrs.com
www.ijtrs.org
Paper Id: IJTRS-V1-I4-011 Volume 1 Issue 4, July 2016
 ISSN No.: 2454- 2024 (online)

International Journal of Technical Research & Science

[10] Ramesh Chand Pandey, Sanjay Kumar Singh, K. K. Shukla and RishabhAgrawal, “Fast and Robust Passive
Copy-Move Forgery Detection Using SURF and SIFT Image Features”.
[11] MohdDilshad Ansari, S. P. Ghrera and VipinTyagi, ”Pixel-Based Image Forgery Detection: A Review”,
IETE Journal of Education, 55:1, 40-46, 2014.
[12] Andrea Costanzo, Irene Amerini, Roberto Caldelli and Mauro Barni, “Forensic Analysis of SIFT Keypoint
Removal and Injection”, IEEE Transactions on Information Forensics and Security, VOL. 9, NO. 9,
september 2014

pg. 24
www.ijtrs.com
www.ijtrs.org
Paper Id: IJTRS-V1-I4-011 Volume 1 Issue 4, July 2016