See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/342488952

A Survey on Image Forgery Detection Using Different Forensic Approaches

Article in Advances in Science Technology and Engineering Systems Journal · January 2020
DOI: 10.25046/aj050347

CITATIONS READS
39 4,673

3 authors, including:

Akram Hatem Saber Mohd Ayyub Khan

Aligarh Muslim University Aligarh Muslim University
7 PUBLICATIONS 83 CITATIONS 16 PUBLICATIONS 91 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Akram Hatem Saber on 03 March 2021.

The user has requested enhancement of the downloaded file.

 Advances in Science, Technology and Engineering Systems Journal Vol. 5, No. 3, 361-370 (2020)
ASTESJ
www.astesj.com
ISSN: 2415-6698

A Survey on Image Forgery Detection Using Different Forensic Approaches

Akram Hatem Saber1,*, Mohd Ayyub Khan1, Basim Galeb Mejbel2
1
Department of Electronics Engineering, Aligarh Muslim University, 202002, India
2
Department of Computer Technician Engineering, AL-Esraa University, 10069, Iraq
ARTICLE INFO ABSTRACT
Article history: Recently, digital image forgery detection is an emergent and important area of image
Received: 18 February, 2020 processing. Digital image plays a vital role in providing evidence for any unusual incident.
Accepted: 13 May, 2020 However, the image forgery my hide evidence and prevents the detection of such criminal
Online: 11 June, 2020 cases due to advancement in image processing and availability of sophisticated software
tamper of an image can be easily performed. In this paper, we provide a comprehensive
Keywords:
review of the work done on various image forgeries and forensic technology. Many
Forgery Detection
techniques have been proposed to detect image forgery in the literature such as digital
Active Forensic Approaches
watermarking, digital signature, copy-move, image retouching, and splicing. The
Passive Forensic Approaches
investigation done in this paper may help the researcher to understand the advantage and
Tampering Identification
handles of the available image forensic technology to develop more efficient algorithms of
Copy-Move
image forgery detection. Moreover, the comparative study surveys the existing forgery
Digital Watermarking
detection mechanisms include deep learning and convolution neural networks concerning
Digital Signature
it is on benefits and demerits.
Image Cloning
Image Splicing

1. Introduction of active tampering detection and security structure is

embedded into the image but most of the image processing
Digital images are the major information source in recent days, tools are not contained any watermarking or signature
due to its availability and sophistication [1]. Also, it is widely module[4].
used in different fields, detection of digital image forgery is In recent days different methods are developed for made image
utilized in numerous applications that are linked to media, reliable and secure that is analogous to watermarking like
publication, law, military, medical image science applications, message authentication code, image checksum, image hash,
satellite image, and world wide web publications. Because it is and image shielding. Passive image forensics is a challenging
very easy to manipulate and edit [2]. For this reason, different task in image processing techniques[5]. It is not a particular
types of cameras and the user-friendly software are used to method for all cases but different methods each can detect the
create and edit the digital images [3]. Digital images are special forgery. The stream of passive tempering detection is to
frequently used to support the important decision for many deal with analyzing raw image based on different statistics and
situations. Moreover, the digital images are a popular source of semantics of an image content to localize tampering of
information and the reliability of digital image and it becomes image[6].
an important issue. There are several types of image forgery that include image
For image forensics, the techniques are classified into two, such retouching, image splicing, copy and move attack. Image
as the active approach and passive approach. In the case of retouching is considered a minimum harmful type of digital
active approach: in this method, the digital image entails the image forgery. An original image does not significantly
various types of preprocessing like watermark embedded or change, but they reduced some features of the original image.
signature are added in the original image. Digital watermarking This technique is used to edit the image for a popular magazine.
and signature are two different active protection techniques. If This type of image forgery is located in all magazine covers
the image has tampered, special information is not extracted and also it used to improve the specific features of an image[7].
from the obtained image. Watermarking is one of the methods On the other hand, Image splicing or photo montage refers to
*
make a forgery image and it is more aggressive than image
Akram Hatem Saber, Aligarh, +919515584268, [email protected]
www.astesj.com 361
https://dx.doi.org/10.25046/aj050347
 A.H. Saber et al. / Advances in Science, Technology and Engineering Systems Journal Vol. 5, No. 3, 361-370 (2020)
retouching. Image splicing is an easy process and it pastes the 2. Digital Image Forgery Detection Methods
regions from isolated sources. This method is referred to as
paste-up formed by sticking together the image by using digital Typically, the methodologies used for forgery detection are
tools like photoshop. This technique is a group of two or more classified into two types such as active forensics and passive
images that are combined to generate fake images[8]. forensics, in which digital watermarking and digital signature
However, copy and move attack is also one of the popular and are the types of active techniques. Then, the splicing, image
difficult images tampering technique. It required the cover part retouching, image cloning, and copy-move techniques are the
of a similar image to add or remove the information. Copy and categories of the passive technique [17]. The description of
move attack, the aim is to hide some information in the original these techniques are investigated in the following sub-sections.
image. The detection of the forged image from the original one
is very hard. The naked eye is not able to identify the tampered
region from a forged image. The image tampering is a general
manipulation of digital images. Traditional block-based
forgery detection methodologies are categorized as the input
images into overlapping and regular image blocks and also
tampered regions are identified by matching blocks of pixel or
transform coefficient[9].
Normally, the image forgery detection is performed by using
the following techniques: JPEG quantization tables, Chromatic
Aberration, Lighting, Camera Response Function (CRF), Bi-
coherence and higher-order statistics, and Robust matching.
The digital cameras encode the images based on JPEG
compression [10], which configures the devices at various
compression level. Then, the sign of image tampering is Figure 1. Categories of image forgeries [17]
evaluated by analyzing the inconsistency of lateral chromatic
aberration [11]. In which, the average angular between the local 2.1. Digital Watermarking
and global parameters is computed for every pixel in the image. In this type of image forgery, a digital watermark is added on
If the average value exceeds the threshold, it is stated that the the photo, which is more or less visible. Then, the appended
deviation is unpredictable in the image due to the forgery of the information is more or less transparent, so it is very difficult to
image. Then, for each object in the image, the notice the watermark. Ferrara, et al. [18] suggested a new
inconsistencies and the illuminating light source is detected to forensic tool for analyzing the original image and forged
identify the forgery [12]. Typically, different measurements regions based on the interpolation process. The image splicing
such as infinite, local and multiple are considered for can be detected by the use of the conditional Co-occurrence
determining the error rate. Then, the CRF is mainly used to Probability Matrix (CCPM) [19], which uses the third-order
expose the image splicing instituted on the geometry invariant statistical features during the forgery detection. Normally, the
of the image. In which, the suspected boundary is identified watermarking schemes are categorized as reversible and
within each region of the image, and it is validated for irreversible. In which, the image irreversible distortions are
identifying the inconsistencies [13]. The bi-coherence features avoided based on the original features of the image by using
[14] are widely used for detecting the splicing on images that the reversible watermarking techniques. The watermarking can
estimate the mean of magnitude and phase entropy for be mainly used to indicate the source or authorized consumer
augmenting the images. Moreover, it extracts the features for of the image. It is a pattern of bits that is inserted into a digital
the authentic counterpart and incorporates it to capture the media for identifying the creator [20]. The watermarking
characteristics of various object interfaces. Finally, the exact techniques are semi-fragile, fragile, and content based, which
replicas are identified by matching the features concerning the are mainly used for image authentication application.
block size, which is done by the use of robust matching [15]. Li, et al. [21] implemented a new method for detecting the copy
But, it requires the human intervention for interpreting the move forgery, where the Local Binary Pattern (LBP) was
output of replicas detection [16]. Generally, the region utilized to extract the circular blocks. The stages involved in
duplication is performed on the image based on the geometrical this system are preprocessing, feature extraction, feature
and illumination adjustments. It is a very simple operation in matching, and post processing. Here, it is stated that when the
which a continuous portion of pixels is copied and pasted on region is rotated at different angles, it is highly difficult to
some other location in the image. This paper is fully focused detect the forgeries. Hussain, et al. [22] suggested a multi-
on the detailed investigation of the image forgery detection resolution Weber Local Descriptors (WLD) for detecting the
mechanisms. The remaining sectors present in the study are image forgeries based on the features obtained from the
arranged as follows: Section II investigates some of the image chrominance components. Here, the WLD histogram
forgery detection mechanisms used in digital image processing. components are calculated and the Support Vector Machine
Section III surveys the forensic approaches and its working (SVM) classifier is utilized to detect the forgery. In this paper,
procedure for image forgery detection. Section IV presents a two different types of forgeries such as splice and copy-move
detailed investigation of the existing methodologies used for are detected by using the multi-resolution WLD approach.
image forgery detection with its advantages and disadvantages.
The overall conclusion of the paper is presented in Section V.
www.astesj.com 362
 A.H. Saber et al. / Advances in Science, Technology and Engineering Systems Journal Vol. 5, No. 3, 361-370 (2020)
forgery, it is very difficult to extract the exact shape of the
image. Typically, the image splicing method [25] is
categorized into two types such as boundary-based and region-
based. Alahmadi, et al [26] suggested a passive splicing
forgery detection mechanism for verifying the authenticity of
digital images. Here, the features are extracted from the
chromatic channel for capturing the tampering artifacts. Kakar,
et al [27] utilized a forgery detection approach for detecting the
splicing in the digital images. Here, the small inconsistencies
in the motion blur are detected by analyzing the special
characteristics of image gradients [28]. The stages involved in
this detection are image subdivision, motion blur estimation,
smoothing, blur computation, interpolation and segmentation
[29]. The authors of this paper [30] employed a machine
learning algorithm for detecting the image splices. The
illumination analysis is highly effective for the detection of
image splicing [31]. To increase the effect of photorealism, an
image splicing operation is performed with the operations of
color and brightness adjustment. In this paper [32] the radial
distortion from various portions of the image is estimated for
the detection of image splicing.

Figure 2. Digital watermarking [18]

2.2. Digital Signature

Normally, the authenticity of the digital messages is validated
based on the digital signature. Because, based on the valid
signature, the recipient can believe that the message is formed
by the recognized sender. Thus, the digital signature is widely
used in the fields of financial transactions, contract Figure 3 (a). Original image and (b). Duplicated image [32]
management software, and software distribution [20].
Normally, the digital signature embeds some secondary 2.4. Image Retouching
information, which is obtained from the image. In this method
Among the other image forgeries, image retouching is
[23], the distinct features are extracted from the image during
considered as the less harmful forgery technique, in which
the initial stage, based on these, the image authenticity is
some enhancement can be performed on the image. Also, it is
validated. Typically, the digital signature has the following
popular in photo editing applications and
properties:
magazines. Muhammad, et al [33] suggested an un-decimated
• Only the sender can sign the image and the receiver can dyadic wavelet transformation technique for detecting the
validate the signature copy-move forgery. Typically, more sophisticated tools are
• Unauthenticated users cannot able to forge the signature available for making this type of forgery by applying the soft
• It provides an integrity touch on the edges. So, it is very difficult to differentiate the
• Also, it achieves non-reputation color and texture of the stimulated part with the unoriginal part.
2.3. Splicing Method Moreover, it makes the forgery detection as highly
complicated, because of two or more identical objects in the
Image splicing is a kind of forgery detection method, in which same image. So, the authors of this paper utilized similarity
a single image is created based on the combination of two or measurements for detecting this forgery, in which the noisy
more images [24]. It is also termed as image composition, in inconsistency is analyzed between the copied and moved parts.
which various image manipulation operations are performed. Here, it is stated that the transformation methods such as FMT,
Typically, many inconsistencies may be created in the image Scale Invariant Feature Transform (SIFT), and Discrete
features due to the splicing operation. In this technique, the Wavelet Transform (DWT) can detect the forgery in a highly
composition between the two images is estimated and compressed image. Ghorbani, et al [34] recommended a
incorporated for creating a fake image. Based on the image Discrete Cosine Transform Quantization Coefficients
block content, the difference between the illumination and Decomposition (DCT-QCD) for detecting the copy-move
reference illuminate color is estimated. In this digital image forgery. The integrity and authenticity verification of digital

www.astesj.com 363
 A.H. Saber et al. / Advances in Science, Technology and Engineering Systems Journal Vol. 5, No. 3, 361-370 (2020)
images is a very difficult process, specifically the images used methodology can detect blur degradation, noise, and some
for news items, medical records, and court law. Because the other arbitrary changes in the duplicate image regions like
copy-move forgery may be created for those types of images. noise addition and gamma correction gamma is a non-linear
adjustment to individual pixel values. The steps involved in this
method are image tiling with overlapping, representation blur
moment invariants, transformation, similarity analysis, and
map creation for duplication region detection. Moreover, the
dimensionality of blocks was reduced by using the principle
component transformation. Muhammad, et al [41] employed a
Dyadic undecorated Wavelet Transformation (Dew) technique
for detecting blind copy-move image forgery detection. This
transformation technique aimed to extract the low frequency
and high-frequency components by estimating the similarity
between the blocks [42]. Moreover, the Euclidean distance is
computed between every pair of blocks in the image. Then, the
match is identified by computing the threshold value between
the sorted lists [43]. In the wavelet transformation, the
downsampling process is not involved, and the coefficients are
not shrunk between the scales. Lynch, et al [44] aimed to detect
(a) (b) the copy-move forgery by the use of expanding block
Figure 4 (a). Forged image and (b). Real image [33] algorithms. Also, it intended to identify the duplicated regions
2.5. Copy-Move Method in the image by estimating the size and shape [45]. In this paper
[46], it is stated that the copy-move forgery is performed for
Among the other forgery methods, the copy-move method an hiding the region of the image by wrapper it with a duplicate
extensively used type of image tampering, where the specific image. Still, recognizing the forged region is extremely
portion is copied and pasted on some other region [35]. The intricate due to the precise copy of another region [47]. This
main motive of this method is to hide a significant element or detection mechanism contains the stages of feature extraction,
highlight a precise object. Bayram, et al [36] implemented a comparison, and similarity estimation for taking copy decisions
proficient method for detecting the copy-move forgery. The [48]. As shown in figure below the procedure of copy move
authors stated that the block matching procedure is used to technique first step the input image preprocessed, second step
detect this type of forgery by separating the image into block division, third step feature extracted, last step the blocks
overlapping chunks. Also, it identifies the duplicated which carry same feature triggered and mapped as a forgery.
connected image blocks by finding the distance between the
neighbor blocks [37]. For taking the forgery decision, only the
duplicate blocks detection is not enough, because the natural
images have many similar blocks [38]. Moreover, the Fourier
Mellin Transform (FMT) is used to perform the operations like
scaling, translation, and rotation for image forgery detection
[39].

(a) (b)
Figure 5 (a). Original image and (b). Tampered image [36].
As shown in figure above copy move image forgery (a) original
image and (b) is tampered image Mahdian, et al [40] utilized
a detection method for identifying the copy-move forgery
based on the blur moment invariants. This detection Figure 6. Procedure of Copy Move technique [48]

www.astesj.com 364
 A.H. Saber et al. / Advances in Science, Technology and Engineering Systems Journal Vol. 5, No. 3, 361-370 (2020)
3. Forensic Approaches Murali, et al [52] investigated various image forgery detection
mechanisms for identifying the forged regions in the forged
In this section, some of the forensic approaches used for digital image. In this paper, it is stated that the copy-move and copy
image forgery detection are surveyed with its working create types are the two kinds of image forgeries, which are
stages. Omen, et al [49] utilized a fractal dimension and implemented at earlier stages. It is detected by using the JPEG
Support Vector Decomposition (SVD) techniques to detect and compression analysis and filtering algorithms. Here, the
isolate the duplicated regions in the image. In this scheme, the algorithms are evaluated based on the factors of image
image is classified into various groups based on its fractal formation, time complexity, multiple forgery detection, and
dimension, which is used to identify the variations. Then, the image transformation. Piva [53] provided a comprehensive
copied and pasted regions are identified by using an efficient overview of image forensics for determining whether the image
texture-based classification technique. Here, it is stated that the content is authenticated or not. The methods investigated in this
SVD is one of the widely used robust and reliable matrix paper were acquisition-based, coding-based methods, and
factorization methods, which offers algebraic and geometric editing based methods. Pan, et al [54] suggested a feature
invariant features for classification. Also, the SVD technique matching technique for identifying the duplicated regions in the
provides maximum energy packaging for exhibiting good digital image.
stability from distortion. It helps to locate the duplicated
regions by adding noise and avoiding the blurred 4. Comparative Study
edges. Chierchia, et al [50] implemented a Bayesian Markov
Random Field (MRF) technique to identify the image forgeries This section surveys the existing forgery detection mechanisms
based on the sensor pattern noise. Here, the observed statistics with respect to its own benefits and demerits. This study is
and prior knowledge were balanced by the use of a Bayesian mainly focused on the detection of image forgery by using
approach. Also, the reliability of forgery detection is improved various forensic approaches. The methods that have been
by using the global optimization algorithm. Bianchi and investigated in this analysis are digital signature verification,
Piva [51] developed a new forensic algorithm for digital image watermarking, cosine transformation,
discriminating against the original and forged regions in the authentication watermarking, SURF, wavelet transformation,
image. Here, the effects of cumulation between various DCT binary pattern extraction, deep learning, block matching, and
coefficients are extracted with the simplified map by using the blind image forgery detection.
unified statistical model.
Table 1. Comparative analysis of various image forensic approaches

S.No Paper Title Methods Used Tampering Detection Pros/Cons Publication

Type Year
1. Research issues and Digital Signature The validity of multiple Advantage: 2005
challenges for Verification Schemes digital signatures are 1. It provides the clear overview of
multiple digital [55] verified. various signature verification
signatures schemes with its specific limitations.
Disadvantage:
1. However, it failed to state an
efficient and robust signature
verification scheme
2. ROI based tamper Digital image It is used to detect the Advantages: 2010
detection and watermarking [56] locations of the 1. Good performance in terms of hiding
recovery for tampered portion inside capacity and visual quality
medical images the Region of Interest 2. High embedding capacity
using reversible (ROI). Disadvantages:
watermarking 1. Lack of reversibility
technique 2. Limited hiding capacity
Induced distortions inside the regions
3. A comparison Discrete Cosine It detected a copy-move Advantages: 2010
study on copy- Transformation (DCT image forgery. 1. Energy compaction property
cover image and Principle 2. Reduced time complexity
forgery detection. Component Analysis 3. Increased accuracy
[57] Disadvantages:
1. It required to locate the possible
inconsistency
2. Increased false positive rate
4. A chaotic system Authentication It locates the tampered Advantages: 2011
based fragile watermarking scheme regions for image 1. High security
watermarking [58]. authentication. 2. Superior tamper detection and
scheme for image localization
tamper detection Disadvantages:
1. Increased computational complexity
2. Required to improve the
performance
5. DWT-DCT (QCD) Discrete Wavelet It detected a copy-move Advantages: 2011
based copy-move Transformation image forgery in an 1. Better accuracy
image forgery (DWT) and Discrete accurate manner. 2. Reduced dimensionality of features
detection Disadvantages:
www.astesj.com 365
 A.H. Saber et al. / Advances in Science, Technology and Engineering Systems Journal Vol. 5, No. 3, 361-370 (2020)
Cosine Transformation 1. Heavy compression
(DCT) [34] 2. It required to remove the position of
pasted areas
3. Increased complexity
6. Detection of region Speeded Up Robust A copy move forgery is Advantages: 2011
duplication forgery Features (SURF) [59] detected with better 1. Better detection rate
in digital images detection performance. 2. It evaluated the image with different
using SURF angles

Disadvantages:
1. Required to reduce the false match
rate
2. Also, it failed to identify the small
copied regions.
7. Passive copy move Undecimated dyadic A copy move image Advantages: 2012
image forgery wavelet transformation forgery is detected 1. It estimated the methods based on
detection using [33] efficiently. three case studies
undecimated dyadic 2. Better performance results
wavelet transform Disadvantages:
1. Noise estimation is not robust
2. It is not translation invariant
8. A novel video Inter-frame forgery It detected the frame Advantages: 2013
inter-frame forgery model detection insertion and deletion 1. It provides the good performance by
model detection mechanism [60] forgery. efficiently identifying the frame
scheme based on insertion and deletion
optical flow Disadvantages:
consistency 1. Reduced precision
2. Increased false detection rate
9. Digital image Fragile watermark Authentication based Advantages: 2013
tamper detection detection technique tampering detection is 1. Robust watermark
techniques-a [61] performed. 2. It accurately pinpoint the forgeries
comprehensive Disadvantages:
study 1. It required a digital signature on the
images
2. Not highly efficient
10. Survey on blind Blind image forgery It detects the copy- Advantages: 2013
image forgery detection [62] move, splicing, and 1. It evaluated different number of
detection retouching image matches for forgery identification
forgeries. 2. It efficiently identified the duplicated
blocks
Disadvantages:
1. It required to analyze the quality of
image
2. Increased time consumption
11. Splicing image Local Binary Pattern Here, an image splicing Advantages: 2013
forgery detection (LBP) and Discrete forgery is detected 1. Better detection performance
based on DCT and Cosine Transformation accurately. 2. Increased accuracy
Local Binary (DCT) [26] Disadvantages:
Pattern 1. Increased complexity
2. Not highly efficient
12. A Forensic Method SIFT based feature Dissimilar types of Advantages: 2013
for Detecting extraction and image tampering are 1. Highly efficient
Image Forgery codebook generation concentrated in this 2. Better accuracy
Using Codebook [63] paper that includes Disadvantages:
enhancing, composting 1. Requires more time for detection
and copy move. 2. It distorts the content
3. Inconclusive results

13. Region Duplication Hybrid wavelet It detected a copy move Advantages 2014
Forgery Detection transformation image forgery and 1. Effective compression
using Hybrid technique [64] region duplication 2. It detected the duplicated regions
Wavelet forgery. with increased accuracy
Transforms Disadvantages
1. It failed to detect the duplicated
regions, when the copied region is
rotated or scaled
2. Not highly efficient
14. Digital image Passive image A copy move image Advantages: 2014
forgeries and authentication forgery is detected in an 1. Reduced computational complexity
passive image techniques [20] efficient way. 2. Increased robustness
authentication Disadvantages:
techniques: A 1. Sharp edge disturbances after
survey splicing
2. Not reliable feature extraction

www.astesj.com 366
 A.H. Saber et al. / Advances in Science, Technology and Engineering Systems Journal Vol. 5, No. 3, 361-370 (2020)
15. Image Forgery Discrete Wavelet Copy move image Advantages: 2016
Detection using Transformation forgery is detected with 1. Good efficiency
Speed up Robust (DWT) and dyadic better accuracy. 2. Reliable
Feature wavelet transformation
Transform, techniques [65] Disadvantages:
Wavelet 1. Not more suitable for noisy image
Transform, 2. Time complexity is high
Steerable Pyramid
Transform
and Local Binary
Pattern
16. An Evaluation of Pixel based image Image splicing, copy- Advantages: 2017
Digital Image forgery detection [19] move and image 1. Better accuracy
Forgery Detection resampling forgeries are 2. High reliability
Approaches detected. Disadvantages:
1. Will not work in the noisy image
2. Time consuming
17. A Review Paper on Brute force, block A generalized schema is Advantages: 2017
Digital Image based and key point developed for detecting 1. Reduced complexity
Forgery Detection based techniques [66] a copy move image 2. Quit robust
Techniques forgery. Disadvantages:
1. Not efficient for complicated
background and texture
2. Less accurate
18. Boosting Image Deep learning The copy move image Advantages: 2018
Forgery Detection mechanism [67] features are identified 1. Reduced false positive
using Resampling for detecting the 2. Highly efficient
Features forgery. Disadvantages:
and Copy-move 1. Not highly robust
Analysis 2. Less accurate
19. Accurate and Lateral Chromatic Image forgery is Advantages: 2018
Efficient Image Aberration (LCA) and detected by analyzing 1. Increased efficiency
Forgery Detection block matching the hypothesis testing 2. Reduced complexity
Using Lateral algorithm [68] problem. Disadvantages:
Chromatic 1. Increased estimation error
Aberration 2. Not suitable for noisy images
20. Recent Advances in Passive digital image It detected the image Advantages: 2018
Passive Digital forensic approaches forgeries based on the 1. Better generalization ability
Image Security [69] artifacts. 2. Minimized time consumption
Forensics: A Brief Disadvantages:
Review 1. Handling difficulty in most forgery
cases
2. Performance degradation
21 Image Splicing this approach three Spliced image forgery Advantages: 2019
Detection using classifiers detection using image as 1- Increase the accuracy
Deep Multiclass Model input for CNN and 2- Localization of spliced forged
Residual Network using SVM Learner, processed through image efficiently
K-NN and Naïve various layers Disadvantages:
Bayes are used to train 1- Not suitable for copy-move forgery
the classifier detection
model[70] 2- Required highly performance
system to implement the algorithms
22 Image splicing paper proposes Spliced image forgery Advantages: 2019
forgery detection detection method with detection with two parts 1- Decrease the computational
combining coarse two parts: (C2RNet) and diluted complexity.
to refined Coarse-to- refined adaptive Clustering. 2- Tremendous decrease in the time.
convolutional convolutional neural
neural network and network (C2RNet) and Disadvantages:
adaptive clustering diluted adaptive
Clustering, replace 1- Slightly Poorer in visual
patch-level CNN in performance.
C2RNet.[71] 2- Poorer in Recall than that of several
of comparison methods.

23 Image Forgery low computational- Daubechies wavelet Advantages: 2019

Detection: A Low cost and effective data- transform is utilized, 1- Reduce computational cost.
Computational- driven model as a representing YCrCb 2- Increase accuracy.
Cost and Modified deep patches inside the
Effective Data- learning-based model image, neural network Disadvantages:
Driven Model [72] used to classify forged 1- Not highly robust
patches. 2- Time complexity is high

www.astesj.com 367
 A.H. Saber et al. / Advances in Science, Technology and Engineering Systems Journal Vol. 5, No. 3, 361-370 (2020)
24 Morphological Mathematical operates on grayscale Advantages: 2020
Filter Detector for morphological filter images, propose a non- 1- Robust to image compression
Image detector (considered trivial extension of 2- Very high accuracy
Forensics Gaussian low pass and a deterministic approach
Applications Median filtering)[73] originally detecting Disadvantages
erosion and dilation of 1- Mathematical complexity
binary images 2- Time complexity
25 Constrained Image Newly methods used Splice forgery detection, Advantages: 2020
Splicing Detection AttentionDM for and detects whether one 1- Performance improved
and CISDL[74] image has forged 2- Computational improved
Localization With regions Disadvantages:
Attention-Aware pasted from the other 1- Equal error rate and detection rate
Encoder-Decoder reduced
and Atrous 2- Slightly slower than DMAC
Convolution
26 Deep Learning Deep convolution Image splice detection Advantages: 2020
Local Descriptor neural network CNN, a and localization scheme 1- Robustness against JPEG
for Image two branch CNN used compression
Splicing Detection with automatically 2- Highly detection accuracy
and Localization learn hierarchical [75]
Disadvantages:
1- Huge complexity while used 30
linear high pass filter
2- Future fusion is complex

5. Conclusion and Future Work Transactions on Information Forensics and Security, vol. 10, pp. 1705-
1716, 2015.
This paper surveyed various image forensics approaches for [8] I. Amerini, L. Ballan, R. Caldelli, A. Del Bimbo, and G. Serra,
"Geometric tampering estimation by means of a SIFT-based forensic
identifying the forgeries performed on the digital images. The analysis," in Acoustics Speech and Signal Processing (ICASSP), 2010
techniques investigated in this paper are digital signature, IEEE International Conference on, 2010, pp. 1702-1705.
digital watermarking, copy-move, image splicing, and image [9] W. Li and N. Yu, "Rotation robust detection of copy-move forgery," in
cloning. Most of the authors stated that image forgery detection Image Processing (ICIP), 2010 17th IEEE International Conference on,
2010, pp. 2113-2116.
is a highly complicated process due to the advent of various [10] M. C. Stamm, S. K. Tjoa, W. S. Lin, and K. R. Liu, "Undetectable image
manipulation and editing tools. The feature is also playing an tampering through JPEG compression anti-forensics," in Image
essential role in forgery detection because the features are Processing (ICIP), 2010 17th IEEE International Conference on, 2010,
highly sensitive to some forgery operations. Moreover, pp. 2109-2112.
[11] V. Christlein, C. Riess, and E. Angelopoulou, "On rotation invariance in
different image processing techniques such as preprocessing, copy-move forgery detection," in Information Forensics and Security
feature extraction, feature selection, and classification are (WIFS), 2010 IEEE International Workshop on, 2010, pp. 1-6.
highly useful for detecting the forgeries in an exact manner. The [12] E. Kee and H. Farid, "Exposing digital forgeries from 3-D lighting
passive methods are highly suitable for forgery detection environments," in Information Forensics and Security (WIFS), 2010 IEEE
International Workshop on, 2010, pp. 1-6.
compared to the active approaches. Because it analyzes the [13] M. Kobayashi, T. Okabe, and Y. Sato, "Detecting forgery from static-
pixel variations and estimates the geometrical illuminations in scene video based on inconsistency in noise level functions," IEEE
an efficient manner. Among the other passive methods, the Transactions on Information Forensics and Security, vol. 5, pp. 883-892,
copy-move and image splicing are widely used by many 2010.
[14] Z. He, W. Sun, W. Lu, and H. Lu, "Digital image splicing detection based
researchers due to its benefits of reduced complexity and on approximate run length," Pattern Recognition Letters, vol. 32, pp.
increased accuracy. 1591-1597, 2011.
[15] M. Jaberi, G. Bebis, M. Hussain, and G. Muhammad, "Accurate and
References robust localization of duplicated region in copy–move image forgery,"
Machine vision and applications, vol. 25, pp. 451-475, 2014.
[16] D. Cozzolino, D. Gragnaniello, and L. Verdoliva, "Image forgery
[1] B. Mahdian and S. Saic, "A bibliography on blind methods for identifying detection through residual-based local descriptors and block-matching,"
image forgery," Signal Processing: Image Communication, vol. 25, pp. in Image Processing (ICIP), 2014 IEEE International Conference on,
389-399, 2010. 2014, pp. 5297-5301.
[2] J. Li, X. Li, B. Yang, and X. Sun, "Segmentation-based image copy-move [17] G. Muhammad, "Multi-scale local texture descriptor for image forgery
forgery detection scheme," IEEE Transactions on Information Forensics detection," in Industrial Technology (ICIT), 2013 IEEE International
and Security, vol. 10, pp. 507-518, 2015. Conference on, 2013, pp. 1146-1151.
[3] H.-D. Yuan, "Blind forensics of median filtering in digital images," IEEE [18] P. Ferrara, T. Bianchi, A. De Rosa, and A. Piva, "Image forgery
Transactions on Information Forensics and Security, vol. 6, pp. 1335- localization via fine-grained analysis of CFA artifacts," IEEE
1345, 2011. Transactions on Information Forensics and Security, vol. 7, pp. 1566-
[4] C. Chen, J. Ni, and J. Huang, "Blind detection of median filtering in digital 1577, 2012.
images: A difference domain based approach," IEEE Transactions on [19] A. Kashyap, R. S. Parmar, M. Agrawal, and H. Gupta, "An Evaluation of
Image Processing, vol. 22, pp. 4699-4710, 2013. Digital Image Forgery Detection Approaches," arXiv preprint
[5] X. Lin, J.-H. Li, S.-L. Wang, A.-W.-C. Liew, F. Cheng, and X.-S. Huang, arXiv:1703.09968, 2017.
"Recent Advances in Passive Digital Image Security Forensics: A Brief [20] S. Mushtaq and A. H. Mir, "Digital image forgeries and passive image
Review," Engineering, 2018/02/17/ 2018. authentication techniques: A survey," International Journal of Advanced
[6] M. D. Ansari, S. P. Ghrera, and V. Tyagi, "Pixel-based image forgery Science and Technology, vol. 73, pp. 15-32, 2014.
detection: A review," IETE journal of education, vol. 55, pp. 40-46, 2014. [21] L. Li, S. Li, H. Zhu, S.-C. Chu, J. F. Roddick, and J.-S. Pan, "An efficient
[7] C.-M. Pun, X.-C. Yuan, and X.-L. Bi, "Image forgery detection using scheme for detecting copy-move forged images by local binary patterns,"
adaptive oversegmentation and feature point matching," IEEE
www.astesj.com 368
 A.H. Saber et al. / Advances in Science, Technology and Engineering Systems Journal Vol. 5, No. 3, 361-370 (2020)
Journal of Information Hiding and Multimedia Signal Processing, vol. 4, [42] K. Asghar, Z. Habib, and M. Hussain, "Copy-move and splicing image
pp. 46-56, 2013. forgery detection and localization techniques: a review," Australian
[22] M. Hussain, G. Muhammad, S. Q. Saleh, A. M. Mirza, and G. Bebis, Journal of Forensic Sciences, vol. 49, pp. 281-307, 2017.
"Image forgery detection using multi-resolution Weber local descriptors," [43] N. Chaitawittanun, "Detection of copy-move forgery by clustering
in 2013 IEEE EUROCON, , 2013, pp. 1570-1577. technique," International Proceedings of Computer Science &
[23] T. Mahmood, T. Nawaz, R. Ashraf, M. Shah, Z. Khan, A. Irtaza, et al., Information Technology, vol. 50, 2012.
"A survey on block based copy move image forgery detection [44] G. Lynch, F. Y. Shih, and H.-Y. M. Liao, "An efficient expanding block
techniques," in Emerging Technologies (ICET), 2015 International algorithm for image copy-move forgery detection," Information Sciences,
Conference on, 2015, pp. 1-6. vol. 239, pp. 253-265, 2013.
[24] D. Chauhan, D. Kasat, S. Jain, and V. Thakare, "Survey on Keypoint [45] L. Chen, W. Lu, and J. Ni, "An image region description method based
Based Copy-move Forgery Detection Methods on Image," Procedia on step sector statistics and its application in image copy-rotate/flip-move
Computer Science, vol. 85, pp. 206-212, 2016/01/01/ 2016. forgery detection," International Journal of Digital Crime and Forensics
[25] T. Huynh-Kha, T. Le-Tien, S. Ha-Viet-Uyen, K. Huynh-Van, and M. (IJDCF), vol. 4, pp. 49-62, 2012.
Luong, "A Robust Algorithm of Forgery Detection in Copy-Move and [46] A. Kaur and R. Sharma, "Optimization of copy-move forgery detection
Spliced Images," International Journal of advanced Computer SCience technique," Computer Engineering and Applications Journal, vol. 2,
and Applications, vol. 7, pp. 1-8, 2016. 2013.
[26] A. A. Alahmadi, M. Hussain, H. Aboalsamh, G. Muhammad, and G. [47] M. Sridevi, C. Mala, S. Sandeep, and N. Meghanathan, "Copy–move
Bebis, "Splicing image forgery detection based on DCT and Local Binary image forgery detection in a parallel environment," SIPM, FCST, ITCA,
Pattern," in Global Conference on Signal and Information Processing WSE, ACSIT, CS and IT, vol. 6, pp. 19-29, 2012.
(GlobalSIP), 2013 IEEE, 2013, pp. 253-256. [48] B. Liu, C.-M. Pun, and X.-C. Yuan, "Digital image forgery detection
[27] P. Kakar, N. Sudha, and W. Ser, "Exposing digital image forgeries by using JPEG features and local noise discrepancies," The Scientific World
detecting discrepancies in motion blur," IEEE Transactions on Journal, vol. 2014, 2014.
multimedia, vol. 13, pp. 443-452, 2011. [49] R. S. Oommen, M. Jayamohan, and S. Sruthy, "Using Fractal Dimension
[28] A. C. Gallagher and T. Chen, "Image authentication by detecting traces and Singular Values for Image Forgery Detection and Localization,"
of demosaicing," in Computer Vision and Pattern Recognition Procedia Technology, vol. 24, pp. 1452-1459, 2016/01/01/ 2016.
Workshops, 2008. CVPRW'08. IEEE Computer Society Conference on, [50] G. Chierchia, G. Poggi, C. Sansone, and L. Verdoliva, "A Bayesian-MRF
2008, pp. 1-8. approach for PRNU-based image forgery detection," IEEE Transactions
[29] J. G. Han, T. H. Park, Y. H. Moon, and I. K. Eom, "Efficient Markov on Information Forensics and Security, vol. 9, pp. 554-567, 2014.
feature extraction method for image splicing detection using [51] T. Bianchi and A. Piva, "Image forgery localization via block-grained
maximization and threshold expansion," Journal of Electronic Imaging, analysis of JPEG artifacts," IEEE Transactions on Information Forensics
vol. 25, p. 023031, 2016. and Security, vol. 7, pp. 1003-1017, 2012.
[30] T. J. De Carvalho, C. Riess, E. Angelopoulou, H. Pedrini, and A. de [52] S. Murali, G. B. Chittapur, and B. S. Anami, "Comparision and analysis
Rezende Rocha, "Exposing digital image forgeries by illumination color of photo image forgery detection techniques," arXiv preprint
classification," IEEE Transactions on Information Forensics and arXiv:1302.3119, 2013.
Security, vol. 8, pp. 1182-1194, 2013. [53] A. Piva, "An overview on image forensics," ISRN Signal Processing, vol.
[31] F. Hakimi and I. M. H. Zanjan, "Image-Splicing Forgery Detection Based 2013, 2013.
On Improved LBP and K-Nearest Neighbors Algorithm," International [54] X. Pan and S. Lyu, "Region duplication detection using image feature
Journal Of Electronics Information & Planning, 2015. matching," IEEE Transactions on Information Forensics and Security,
[32] H. R. Chennamma and L. Rangarajan, "Image splicing detection using vol. 5, pp. 857-867, 2010.
inherent lens radial distortion," arXiv preprint arXiv:1105.4712, 2011. [55] M.-S. Hwang and C.-C. Lee, "Research Issues and Challenges for
[33] G. Muhammad, M. Hussain, and G. Bebis, "Passive copy move image Multiple Digital Signatures," IJ Network Security, vol. 1, pp. 1-7, 2005.
forgery detection using undecimated dyadic wavelet transform," Digital [56] O. M. Al-Qershi and B. E. Khoo, "ROI-based tamper detection and
Investigation, vol. 9, pp. 49-57, 2012. recovery for medical images using reversible watermarking technique,"
[34] M. Ghorbani, M. Firouzmand, and A. Faraahi, "DWT-DCT (QCD) based in Information Theory and Information Security (ICITIS), 2010 IEEE
copy-move image forgery detection," in Systems, Signals and Image International Conference on, 2010, pp. 151-155.
Processing (IWSSIP), 2011 18th International Conference on, 2011, pp. [57] F. Y. Shih and Y. Yuan, "16 A Comparison Study on Copy–Cover Image
1-4. Forgery Detection," Multimedia Security: Watermarking, Steganography,
[35] Z. Mohamadian and A. A. Pouyan, "Detection of duplication forgery in and Forensics, p. 297, 2017.
digital images in uniform and non-uniform regions," in Computer [58] S. Rawat and B. Raman, "A chaotic system based fragile watermarking
Modelling and Simulation (UKSim), 2013 UKSim 15th International scheme for image tamper detection," AEU-International Journal of
Conference on, 2013, pp. 455-460. Electronics and Communications, vol. 65, pp. 840-847, 2011.
[36] S. Bayram, H. T. Sencar, and N. Memon, "An efficient and robust method [59] B. Shivakumar and L. Baboo, "Detection of region duplication forgery in
for detecting copy-move forgery," in Acoustics, Speech and Signal digital images using SURF," IJCSI International Journal of Computer
Processing, 2009. ICASSP 2009. IEEE International Conference on, Science Issues, vol. 8, 2011.
2009, pp. 1053-1056. [60] J. Chao, X. Jiang, and T. Sun, "A novel video inter-frame forgery model
[37] M. Jaberi, G. Bebis, M. Hussain, and G. Muhammad, "Improving the detection scheme based on optical flow consistency," in The International
detection and localization of duplicated regions in copy-move image Workshop on Digital Forensics and Watermarking 2012, 2013, pp. 267-
forgery," in Digital Signal Processing (DSP), 2013 18th International 281.
Conference on, 2013, pp. 1-6. [61] M. Mishra and F. Adhikary, "Digital image tamper detection techniques-
[38] N. B. A. Warif, A. W. A. Wahab, M. Y. I. Idris, R. Ramli, R. Salleh, S. a comprehensive study," arXiv preprint arXiv:1306.6737, 2013.
Shamshirband, et al., "Copy-move forgery detection: Survey, challenges [62] T. Qazi, K. Hayat, S. U. Khan, S. A. Madani, I. A. Khan, J. Kołodziej, et
and future directions," Journal of Network and Computer Applications, al., "Survey on blind image forgery detection," IET Image Processing,
vol. 75, pp. 259-278, 2016. vol. 7, pp. 660-670, 2013.
[39] J. ZHENG, W. HAO, and W. ZHU, "Detection of Copy-move Forgery [63] E. Agnes, S. D. Mahalakshmi, and D. K. Vijayalakshmi, "A Forensic
Based on Keypoints' Positional Relationship," JOURNAL OF Method for Detecting Image Forgery Using Codebook," International
INFORMATION &COMPUTATIONAL SCIENCE, vol. 9, pp. 4729-4735, Journal of Advanced Research in Computer Science and Software
2012. Engineering, vol. 3, 2013.
[40] B. Mahdian and S. Saic, "Detection of copy–move forgery using a method [64] T. K. Sarode and N. Vaswani, "Region Duplication Forgery Detection
based on blur moment invariants," Forensic science international, vol. using Hybrid Wavelet Transforms," International Journal of Computer
171, pp. 180-189, 2007. Applications, vol. 90, 2014.
[41] G. Muhammad, M. Hussain, K. Khawaji, and G. Bebis, "Blind copy move [65] R. Kaur and T. Sharma, "Image Forgery Detection using Speed up Robust
image forgery detection using dyadic undecimated wavelet transform," in Feature Transform, Wavelet Transform, Steerable Pyramid Transform
Digital Signal Processing (DSP), 2011 17th International Conference on, and Local Binary Pattern," International Journal of Modern Computer
2011, pp. 1-6. Science and Applications (IJMCSA), vol. 4, 2016.

www.astesj.com 369
 A.H. Saber et al. / Advances in Science, Technology and Engineering Systems Journal Vol. 5, No. 3, 361-370 (2020)
[66] N. K. Gill, R. Garg, and E. A. Doegar, "A review paper on digital image
forgery detection techniques," in Computing, Communication and
Networking Technologies (ICCCNT), 2017 8th International Conference
on, 2017, pp. 1-7.
[67] T. M. Mohammed, J. Bunk, L. Nataraj, J. H. Bappy, A. Flenner, B.
Manjunath, et al., "Boosting Image Forgery Detection using Resampling
Detection and Copy-move analysis," arXiv preprint arXiv:1802.03154,
2018.
[68] O. Mayer and M. C. Stamm, "Accurate and Efficient Image Forgery
Detection Using Lateral Chromatic Aberration," IEEE Transactions on
Information Forensics and Security, 2018.
[69] X. Lin, J.-H. Li, S.-L. Wang, F. Cheng, and X.-S. Huang, "Recent
Advances in Passive Digital Image Security Forensics: A Brief Review,"
Engineering, 2018.
[70] Ankit Kumar Jaiswal and Rajeev Srivastava, "Image Splicing Detection
using Deep Residual Network," 2nd International Conference on
Advanced Computing and Software Engineering (ICACSE-2019).
[71] Bin Xiao, Yang Wei, Xiuli Bi, Weisheng Li, andJianfeng Ma, "Image
splicing forgery detection combining coarse to refined convolutional
neural network and adaptive clustering " Elsevier Information Sciences,
2019.
[72] Thuong Le-Tien, Hanh Phan-Xuan, Thuy Nguyen-Chinh, and Thien Do-
Tieu, " Image Forgery Detection: A Low Computational-Cost and
Effective Data-Driven Model " International Journal of Machine
Learning and Computing, Vol. 9, No. 2, April 2019.
[73] Giulia Boato, Duc-Tien Dang-Nguyen, and Francesco G. B. Denatale, "
Morphological Filter Detector for Image Forensics Applications" IEEE
Access 2020.
[74] Yaqi Liu, and Xianfeng Zhao, "Constrained Image Splicing Detection and
Localization With Attention-Aware Encoder-Decoder and Atrous
Convolution" IEEE Access2020.
[75] Yuan Rao, Jiangqun Ni, and Huimin Zhao, "Deep Learning Local
Descriptor for Image Splicing Detection and Localization" IEEE
Access2020.

www.astesj.com 370

View publication stats