Digital Image Forgery Detection using Artificial Neural Network and Auto Regressive Coefficients

Gopi, E. S.; Lakshmanan, Nitya; Gokul, T.; KumaraGanesh, S.; Shah, Prerak.r.

doi:10.1109/ccece.2006.277398

Cited by 30 publications

(14 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Auto regressive coefficient as element vector and artificial neural network (ANN) classifier method is developed by the Gopi et al [5] to detect image tampering. In it, 300 attributes vectors were used (form different images) to train an ANN.…”

Section: Related Workmentioning

confidence: 99%

Digital Image Forgery Detection using Correlation Coeficients

Saini¹,

Singh²,

Sethy³

et al. 2015

IJCA

View full text Add to dashboard Cite

In digital era, it has become easy to modify any image. Due to this the trust and validation of it is going to lose. Now it has become major problem of digital world to regain the lost trust. The background behind the modification and any changes in an image is easy availability of software tools on internet. Images can be transformed from one image format to another and any part of image can be altered pixel by pixel. Before the digital age, it was literally easy to detect the altered photographs. But now with the advent in the commercial software like various image photo editing software like Adobe Photoshop, XnView; ProShow Gold etc. make image forgery simple, the tampering of the photographs have become very easy, can be carried out without any noticeable signs of tampering and it is becoming harder to expose and mark the authentic ones. With the increased dependency over the digital images for exchanging the information, the need to keep their authenticity increases and digital images also use as authenticated facts for an offence. If it will not contain the authenticity then a problem will arise. An image forgery is made either by summing some templates, or hiding some kind of information in an image, in which the consistency is lost. This paper identifies the key methods for detecting forgery in the digital images. To identify and detect the forged areas, the image is divided into overlapped patches of some fixed size. In our paper we will discuss the correlation method, that how it find outs the forged part in an image. Firstly, the digital image tampering process is discussed. After that, it shows that different algorithms have different approaches to detect the forgery.

show abstract

Section: Related Workmentioning

confidence: 99%

Digital Image Forgery Detection using Correlation Coeficients

Saini¹,

Singh²,

Sethy³

et al. 2015

IJCA

View full text Add to dashboard Cite

show abstract

“…Use of machine learning techniques for image forgery detection is relatively new. E.S.Gopi et al used Auto Regressive coefficients as feature vectors and ANN for training the system [11].HMM and SVM were used majorly for speech recognition, signature verification, license plate detection and classification etc. E.Justino et al used SVM and HMM classifiers for off-line signature verification [12].…”

Section: Related Workmentioning

confidence: 99%

Image Forgery Authentication and Classification using Hybridization of HMM and SVM Classifier

Hashmi¹,

Keskar²

2015

IJSIA

View full text Add to dashboard Cite

show abstract

“…Gopi et al, [39] developed a model that used auto regressive coefficients as feature vector and artificial neural network (ANN) classifier to detect image tampering. 300 feature vectors from different images are used to train an ANN and the ANN is tested with another 300 feature vectors.…”

Section: Copy-move Forgery Detectionmentioning

confidence: 99%

“…The sole purpose of classifier is to classify an image either as original or forged. Various classifiers have been used such as neural networks [25],SVM [26,27,28] and LDA [29].Finally some forgeries like copy move and splicing may require post processing which involve operations like localization of duplicate regions [30,31,32,33].…”

Section: Figure 3 Framework For Image Forgery Detectionmentioning

confidence: 99%