Detecting Image Forgery Using Perspective Constraints

Yao, Heng; Wang, Shuozhong; Zhao, Yan; Zhang, Xinpeng

doi:10.1109/lsp.2011.2182191

Cited by 48 publications

(24 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[15][16] have used geometry based methods and considered the relative sizes of the objects in the image with respect to each other so as to detect the forged ones from the clean ones. The authors of the two papers have also claimed that their algorithm produces good results even for low resolution images.…”

Section: Related Workmentioning

confidence: 99%

“…[15] the ratio of the heights of two objects in the same image is considered to detect the forgery, the authors have proposed a perspective constraint method through which this ratio is obtained using the vanishing line (a geometry term). This helps to calculate the ratio without using the parameters of camera.…”

Section: Related Workmentioning

confidence: 99%

“…[16] have proposed a similar model to that of the Ref. [15] but have stated that it works better for images with tilt and roll. In their proposed method, the user has to do some of the important things manually, like selecting the objects and the vanishing lines; which may sometimes not be as efficient as an automatic procedure.…”

Section: Related Workmentioning

confidence: 99%

“…[15][16] have considered the impact of their algorithm on low resolution images. Image resolution is an important factor as; in real life we generally do not have images which are of the same size or resolution.…”

Section: Rq2mentioning

confidence: 99%

See 3 more Smart Citations

Splicing forgery detection and the impact of image resolution

Devagiri

Cheddad

2017

2017 9th International Conference on Electronics, Computers and Artificial Intelligence (ECAI)

View full text Add to dashboard Cite

Conte xt. There has been a rise in the usage of digital images these days. Digital images are being used in many areas like in medicine, wars, etc. As the images are being used to make many important decisions, it is necessary to know if the images used are clean or forged. In this thesis we have considered the area of splicing forgery. In this thesis we are also considering and analyzing the impact of lowresolution images on the considered algorithms. Obje ctives. Through this thesis we try to improve the detection rate of splicing forgery detection. We also examine how the examined splicing forgery detection algorithm works on low-resolution images and considered classification algorithms (classifiers). Methods. The research methods used in this research are Implementation and Experimentation. Implementation was used to answer the first research question i.e., to improve the detection rate in splicing forgery. Experimentation was used to answer the second research question. The results of the experiment were analyzed using statistical analysis to find out how the examined algorithm works on different image resolutions and on the considered classifiers. Resul ts. One tailed Wilcoxon signed rank test was conducted to compare which algorithm performs better, the T+ value obtained was less than To so the null hypothesis was rejected and the alternative hypothesis which states that Algorithm 2 (our enhanced version of the algorithm) performs better than Algorithm 1 (original algorithm), is accepted. Experiments were conducted and the accuracy of the algorithms in different cases were noted, ROC curves were plotted to obtain the AUC parameter. The accuracy, AUC parameters were used to determine the performance of the algorithms. Conclus ions. After the results were analyzed using statistical analysis, we came to the conclusion that Algorithm 2 performs better than Algorithm 1 in detecting the forged images. It was also observed that Algorithm 1 improves its performance on low resolution images when trained on original images and tested on images of different resolutions but, in the case of Algorithm 2, it performance is improved when trained and tested on images of same resolution. There was not much variance in the performance of both of the algorithms on images of different resolution. Coming to the classifiers, Algorithm 1 improves its performance on linear SVM whereas Algorithm 2 improves its performance when using the simple tree classifier.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Rq2mentioning

confidence: 99%

See 2 more Smart Citations

Splicing forgery detection and the impact of image resolution

Devagiri

Cheddad

2017

2017 9th International Conference on Electronics, Computers and Artificial Intelligence (ECAI)

View full text Add to dashboard Cite

show abstract

“…Fridrich [4] estimates the parameters of the noise model from the original camera, and then calculates the correlations between the estimated camera noise and the extracted image noise to authenticate an image. Other effective features to expose forgery include inconsistencies in the illumination [5], light direction, anomalous height ratio of two objects in an image [6], etc.…”

Section: Introductionmentioning

confidence: 99%

Expose Spliced Photographic Basing on Boundary and Noise Features

Hou

Cheng

2015

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYThe paper proposes an algorithm to expose spliced photographs. Firstly, a graph-based segmentation, which defines a predictor to measure boundary evidence between two neighbor regions, is used to make greedy decision. Then the algorithm gets prediction error image using non-negative linear least-square prediction. For each pair of segmented neighbor regions, the proposed algorithm gathers their statistic features and calculates features of gray level co-occurrence matrix. K-means clustering is applied to create a dictionary, and the vector quantization histogram is taken as the result vector with fixed length. For a tampered image, its noise satisfies Gaussian distribution with zero mean. The proposed method checks the similarity between noise distribution and a zero-mean Gaussian distribution, and follows with the local flatness and texture measurement. Finally, all features are fed to a support vector machine classifier. The algorithm has low computational cost. Experiments show its effectiveness in exposing forgery.

show abstract