Professional Documents
Culture Documents
A Novel Approach For Forgery Detection of Images
A Novel Approach For Forgery Detection of Images
A Novel Approach For Forgery Detection of Images
Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com Volume 2, Issue 8, August 2013 ISSN 2319 - 4847
Abstrac
Each image has its own unique characteristics or properties. When we edit an image we are changing these characteristics and by examining the changes in the characteristics we can determine whether the image is edited or not. Image editing can be of two types, usual editing and copy move forgery. In copy-move forgery a part of original digital image is copied and pasted to another part in the same original image to make it, forged one. Because the copied part comes from the same image, its important properties will be compatible with the rest of the image and thus will be more difficult to distinguish and detect these parts.
Keywords- Copy-Move forgery; digital signature; DWT; phase correlation 1. INTRODUCTION In the past few years, there has been a growing interest in the development of detection of editing in images. For editing our photos digitally, there are numbers of different photo editing software and tools available. Image editing is a technique to improve look and feel of photographs and you can compose two or more different photographs or graphics to make something more appealing, interesting and unique concept. Each image has its own unique characteristics or properties. When we edit an image we are changing these characteristics and by examining the changes in the characteristics we can determine whether the image is edited or not. But the difficult part in image editing is the copy-move forgery. Forgery is the process of making, adapting, or imitating objects, statistics, or documents with the intent to deceive. As result of powerful image processing tools, digital image forgeries have already become a serious social problem. It easy to manipulate digital images and to create forgeries that is difficult to distinguish from authentic photographs. In earlier days we used watermarking and digital signature methods for image manipulation and forging. But both these methods requiring the preprocessing of the data such as embedding watermark in the images. This makes them relatively difficult to apply on images.
Figure 1 Tampered image Copy-move forgery is a type of image forgery, in which a part of original digital image is copied and pasted to another part in the same original image to make it, forged one. This can be done for two purposes either to conceal an important object or sometimes to show more than one object.
Figure 2 Forged test image Jeep (above) and its original version (below).
Page 55
Figure 3 Flow diagram of editing detection Secondly we can compare each block of the image with every other block in that particular image based on the correlation value. For that particular image we can find the maximum of the correlation value. Similarly we do the correlation analysis for a set of test images and find the maximum value of correlation for each of these images and from these maximum values we take the higher value as the correlation threshold. 2.1 Usual Edition Detection If we do any usual editing like contrast manipulation, brightness enhancement etc., we are changing the characteristics of the image by providing some external characteristics. Here we are actually adding some gradient values to the pixel values of the image. To detect the usual edition we go for the block wise comparison based on the intensity values. If any external characteristics are added to the pixel values of the images, then the range of pixel values will not be within the range shown by the rsmp code ie, the value will exceed the range b&c. 2.2 Copy Move Forgery Detection In the case of copy-move forgery because the copied part comes from the same image, its important properties will be compatible with the rest of the image so when we do the pixel intensity value comparison, its range will be within the rsmp code range and we cannot detect. So we go for the correlation analysis. Since here the same part are copy pasted, the correlation value should be above the correlation threshold(a).Thus we can detect the copy pasted part. 2.3 Discrete Wavelet Transform The basic idea of using Discrete Wavelet Transform is to reduce the size of the image at each level. An example image along with its wavelet transform applied up to level 3 is shown in Fig. 4.
Page 56
Figure 4 An image and its Wavelet Transform At each level the image is decomposed into four sub images. The sub images are labeled LL, LH, HL and HH. LL corresponds to the coarse level coefficients or the approximation image.
Figure 5 Image pyramid If the number of levels used for decomposition is L, then the matching is performed on the LL image at level L denoted by LLL. Fig.5 shows the image pyramid. 2.4 Phase Correlation This is a method of image registration. The ratio R between two images img1 and img2 is calculated as follows: R= [F (img1) conj (F (img2))]/ [||F (img1) conj (F (img2)) ||] The inverse Fourier transform of R is the phase correlation . 2.5 RSMP code Second order derivative filter is used to find out the rsmp code.Then we do the radon transform of the result of derivative and find its auto covariance . To compute the radon transformation, pixels are divided into four sub pixels and each sub pixel projected separately.The radon transformation is computed at angles from 0 to 179, in 1 increments. Find the maximum value of the magnitudes of the FFT of obtained sequences. In the case of rotation, peaks appearing in the spectrum can help us to determine the angle of rotation transformation. In scaling, there is a direct relation between the normalized position of peaks fn, and the scaling factor, N:
Page 57
Figure 6 Forgery detection result (a) original image (b) tampered image(c) detection result on LLL level image (d) detection result on LLL-1 level image (e) detection result on LLL-2 level image (f) detection result on tampered image.
The detected results over tampered image for all DWT levels are shown in Figure 7. To see how these methods perform under the modifications, we have used US currency note image to illustrate detection as shown in Fig.7
Figure 7 Forgery detection result (a) original image (b) tampered image(c) detection result with 15% normal noise(d) detection result with 25% normal noise(e) detection result with 35% normal noise (f) detection result with 45% normal noise.
The Fig.8 shows the performance of the algorithm results for the image having more than one duplicated regions.
Figure 8 Forgery detection result (a) original image (b) tampered image having more than one duplicated result.
4. CONCLUSION
When we edit an image we are changing the unique characteristics of the image and by examining the changes in these characteristics we can determine whether the image is edited or not. In copy move forgery detection part of original digital image is and pasted to another part in the same original image to make it, forged one. In addition to these copy paste operation the pixel values can also be changed due to stenographic operation. Here we developed a method based on DWT for detecting editing operations performed on the image. The DWT based approach drastically reduces the time needed for the detection process and increases accuracy of detection process.
Acknowledgement
The authors thank the Management and the Principal of Christ Knowledge City, Mannoor, Kerala and ILM College of Engineering & Technology, Perumbavoor, Kerala for providing excellent computing facilities and encouragement.
References
[1.] Saiqa Khan, Arun Kulkarni, Reduced Time Complexity for Detection of Copy-Move Forgery Using Discrete Wavelet TransformInternational Journal of Computer Science Applications, Vol. 6, No. 7, pp: 31-36, Sep 2010. [2.] G.Li, Q.Wu, D.Tu, and Shaojie Sun, A sorted neighborhood approach for detecting duplicated regions in image forgeries based on DWT and SVD, IEEE International Conference on Multimedia & Expo, 2007 [3.] Saiqa Khan, Arun Kulkarni,Robust Method for Detectoim of Copy-Move Frogery in Digital Images International Journal of Computer Science and Engineering, 2010 IEEE [4.] Myna.A.N. , M.G.Venkateshmurthy , C.G.Patil Detection ofRegion Duplication Forgery In Digital Images Using Wavelets and Log-polar Mapping, in Proc. of International Conference on Computational Intelligence and Multimedia Applications, Volume 3,13-15,pp.371377,July2-6,200.
Page 58
Lija Thomas received the B-Tech degree in Electronics and Communication Engineering from Mahatma Gandhi University, Kerala, India in 2010. She received M.Tech degree in Communication Engineering from Mahatma Gandhi University, Kerala, India in 2012. She is currently working as an Assistant Professor in ILM College of Engineering & Technology, Perumbavoor, kerala, India. Her areas of interests are Electronics Circuits, Communication Engineering and digital Image Processing.
Page 59