Algorithm for the Digital Forgery Catching Technique for Image Processing Application Manish Jain 1, Vinod Rampure 2 ¹Department of Computer Science and Engineering, Modern Institute of Technology and research centre, India ² Department of Computer Science and Engineering, Modern Institute of Technology and research centre, India Abstract In the recent years, forgeries have a challenge in every digital aspect. In this paper, we propose an algorithm for digital forgery for both joint photographic experts group (JPEG) and graphics interchange format (GIF). The aim is to capture the forged using original. Predominately existing work focus only JPEG. Our target is to propose a generalised algorithm of forgery that separates the original and forged cue to contribution in digital processing. This work surpasses all the literature work as a point of forgery in both types s. At the end, the experimental setup results ensure that validate of the design approach. The robustness of algorithm of digital forgery is tested on MATLAB R2015a (64bit) tool. In particular, the pseudo-code of the forgery combining both JPEG and GIF is presented. Keywords Forensics, Forged, Euclidean, Compression, Scaling, JPEG, GIF. I. INTRODUCTION The digital is 2D in x and y spatial plane. The intensity of the at any point is determined by spatial coordinate [1]. In this regard, capture, modify, compression, and generation operations are performed by converting the into digital numbers such as 0s and 1s known as bits. The digital is composed of a finite number of pixels. The size of the standard, 1024x1024 pixel and 256 type colours are required 3MB of space in the RAM memory [2, 3]. Moreover, colours type is required more size in RAM memory. In the digital world, cameras are utilised for video and capture. In contrast to bright light, the camera's videos and s have saturation level is high. Consequently, when the dark light is saturation level is low. The saturation level lies between high and low when the light level is maintained [4]. The processing technique is highly used in such areas as biomedical, satellite, communication, electronics etc. In all these areas features like compression, enhancement, and compression are an open area of research, but all these features challenging task is forgeries phenomena happen [5, 6]. In a more formal way, forgeries can be categories into two parts such as analogue and digital [7]. In the type of analogue the continuous signal treated, whereas digital type has discontinuous signal treated. In fact, the digital is popular nowadays in terms of quality of the [8, 9]. The digital forgeries technique is a most used area of current research [10]. The digital type forgery is most active research field with many benefits and threats with the consideration of complexity in the objective. The categorization of forgeries is shown in Figure.1. The remainder of the article is organised as follows: Section 1 describes the fundamental of digital forgery. The contribution related to this work are given in Section 2. Section 3 presents the existing work on digital forgery with their pros and cons. The proposed algorithm five steps are presented and illustrated in Section 4. In this section individual element pseudo code is also presented. Section 5 shows the experimental setup results in terms of original and forged as well as simulation platform setting. Last, the conclusion and future direction are outlined in Section 6. www.ijaetmas.com Page 50
Analog forgeries Audio Forgeries Image Digital forgeries Video Text Animation Fig. 1 Categorization of forgeries. II. CONTRIBUTION In this paper, we proposed an algorithm for forged capture based on our flow chart. Symbolic forms based flow chart related to the digitally forged is shown in Figure. 2. Our algorithm works only on few steps to capture the forgery. Our proposed algorithm are designed to handle both JPEG and GIF s the steps such as: (i) Subtraction (ii) Labelling (iii) Extraction (iv) Extra point marked as black. Each step has introduced as pseudocode for automatic verification to our target achieve. The pseudo code steps of this forgery are processed in MATLAB R2015a (64bit) tool. Where the outputs are evaluated using original and forged. III. EXISTING WORK ON DIGITAL IMAGE FORGERIES Existing research background on digital forgeries is showed in this section which is essential to the readability of the proposed work. The challenging task on forgeries is how to check that intellectual quality in the digital from an authentic point of view [11, 12, 13]. In most of the case, intellectual s assets are original demanded i.e unforged. In digital world uncertainties about ethical, and validity issues in digital forgeries are challenging task. The need for prominent algorithm to identify the unforged digital is more vital than ever [14, 15, 16]. Less amount of article reported in existing literature on forgery capture with the standardised algorithm. The open research is how to recognise the fraud on digital assets. In this way, visual clarity is not altered for identifying the forged elements in the. In 2010, E. Ardizzone et al introduced forgeries capture using SIFT point matching [15]. This forgeries capture algorithm is based on JPEG but is not www.ijaetmas.com Page 51
for GIF. Jessica Fridrich et al introduce the digital forgery by copy move attack [14]. This methodology is based on exact match and robust match. Hwei-Jen lin et al. presented fast copy-move forgery detection [5]. Sevinc Bayram et al presented a survey regarding copy-move forgery detection technique as well as matching duplication blocks and performance results table in the forged [12]. In this method, some duplication information is captured by sorting algorithm with the vector store. None of the existing forgery capture methodology can have both the JPEG and GIF at the same time. Many of the article target only the JPEG forgery without any contribution on the GIF. All these lacks of existing work motivate us to introduce a methodology for both the JPEG and GIF forgery. Also, the primary interest in this article to propose an algorithm for forgery which has minimum steps to capture the forged. IV. THE PROPOSED FORGED IMAGE CAPTURE ALGORITHM Figure. 2 shows the forgery code, wherein there is a two required first: original and second: forged, i.e., two s are first converted into red-greenblue (RGB) to gray. However, the subtraction operation is performed for the extra and missing elements in the objective of unforged. If there are no subtraction operation loop is reset and again process the subtraction operation. After the labelling of the connected components operation is performed. Next step is the extract the component that large size; consequently, the extra part marked black. Furthermore, the following pseudo-code are proposed to describe the algorithm of the forged capture. Original Image Forged Image RGB to gray Original JPEG Subtraction No Forged JPEG YES Labelling Missing part Extracting Original GIF Forged GIF Extra part is marked as black Missing part Original Forged Output Fig. 2 Flow chart for forged. www.ijaetmas.com Page 52
* Pseudo-code for reading the original Img=imread( a1.jpg ); //Reading the original If(ndims(img)==3); img=rgb2gray(img); end * Pseudo-code for reading the forged I=imread( a.jpg); %reading the forged If(ndims(I)==3) I=rgb2gray(I); end * Pseudo-code for subtraction D=abs(double(I)-double(img)); * Pseudo-code for labelling the connected components [L,n]=bwlabel(D,8); Output=img; For j=1:n * Pseudo-code for extracting the components [row, col]=find(l==j); * Pseudo-code for extracting the component that has large size If(numel(row)>sz) * Pseudo-code for finding the starting position (x,y) for the component y=min(col); x=min(row); * Pseudo-code for extra part mark black Output(x:x+blk,y:y+blk)=0; * Pseudo-code for plotting subplot(2,2,1); imshow(img); subplot(2,2,2); imshow(i); subplot(2,2,3); imshow(output); www.ijaetmas.com Page 53
Vertical: 259 Vertical: 259 Vertical: 259 Vertical: 259 International Journal Of Advancement In Engineering Technology, Management and V. EXPERIMENTAL SETUP RESULTS The simulation of the proposed pseudo-code is done on MATLAB R2015a(64bit) tool with editor window software on computer, which has the Intel(R) Core(TM) i5 CPU, AMD graphics, the clock speed of 2.10GHz, 64bit Window8 operating system, and 4GB RAM memory. The dimensions are chosen as a 1024x1024 pixel for both JPEG and GIF s. The challenging task is to apply for both JPEG and GIF s but predominately this work achieves this target. Further, the simulation time is few second to run this pseudocode, which signifies that faster computation time to run this objective. The detail description of JPEG with respect to pixels, size, resolution, print size are shown in Table 1. The parameters of pixels, size, resolution, and print size of GIF are shown in Table 2. As shown on both the Table 1 and 2 the forgery they require only 39.5KB size whereas the processing time in few second. Figures 3 and 4 shows the specification of JPG and GIF s. 72 pixel/inch 72 pixel/inch Original Forged Figure. 3 Specification of JPG. 72 pixel/inch 72 pixel/inch Original Forged Figure. 4 Specification of GIF. Table 1. Selected original and forged JPEG description www.ijaetmas.com Page 54
Image Pixels Size Resolution Print size Original 39.5KB=40,538 72 pixel/inch Width: 9.14 (Fig. 3) Vertical: 194 bytes Forged Vertical: 194 39.1KB=40,109 bytes 72 pixel/inch Width: 9.14 Table 2. Selected original and forged GIF description Image Pixels Size Resolution Print size Original 39.5KB=40,109 72 pixel/inch Width: 9.14 (Fig. 4) Vertical: 194 bytes Forged Vertical: 194 39.1KB=40,538 bytes 72 pixel/inch Width: 9.14 Image type description 40,600 40,500 40,400 40,300 40,200 40,100 40,000 39,900 39,800 1 2 JPEG GIF JPEG GIF Figure. 5 Sample description. VI. CONCLUSIONS Image forgery of the digital is challenging task in recent research. Based on the algorithm the forgery detection is presented in this paper. As an algorithm, the authenticity of the is performed on MATLAB R2015a (64bit) simulation tool. Thus the problem of digital forgery is finally tracked in this work. Initially, in this work, we have shown a pseudo-code of various elements in original digital forgery. After the search approach by our proposed algorithm, we capture both the original and forged. Therefore the robust technique is on time-based, which is few second. www.ijaetmas.com Page 55
CONFLICT OF INTEREST The authors declare that they have no conflict of interest. REFERENCES [1] Edoardo Ardizzone, Alessandro Bruno, and Giuseppe Mazzola. "Detecting multiple copies in tampered s." 17th IEEE International Conference on Image Processing (ICIP), pp. 2117-2120. 2010. [2] Husrev T Sencar, and Nasir Memon. "Overview of state-of-the-art in digital forensics." Algorithms, Architectures and Information Systems Security, 3, pp. 325-348, 2008. [3] Guohui Li, Qiong Wu, Dan Tu, and Shaojie Sun. "A sorted neighborhood approach for detecting duplicated regions in forgeries based on DWT and SVD." IEEE International Conference on Multimedia and Expo, pp. 1750-1753, 2007. [4] Hailing Huang, Weiqiang Guo, and Yu Zhang. "Detection of copy-move forgery in digital s using SIFT algorithm." IEEE Pacific-Asia Workshop on Computational Intelligence and Industrial Application, vol. 2, pp. 272-276, 2008. [5] Hwei-Jen Lin, Chun-Wei Wang, and Yang-Ta Kao. "Fast copy-move forgery detection." WSEAS Transactions on Signal Processing, Vol. 5, no. 5, pp. 188-197, 2009. [6] Kuo-ming Hung, Ching-tang Hsieh, and Kuan-ting Yeh. "Multi-Purpose Watermarking Schemes for Color Halftone Image Based on Wavelet and Zernike Transform." In WSEAS Transaction on Computer. 2007. [7] Phen Lan Lin, Chung-Kai Hsieh, and Po-Whei Huang. "A hierarchical digital watermarking method for tamper detection and recovery." Pattern recognition, Vol. 38, no. 12, pp. 2519-2529, 2005. [8] Alin C Popescu, and Hany Farid. "Exposing digital forgeries by detecting traces of resampling." IEEE Transactions on signal processing, Vol. 53, no. 2, pp. 758-767, 2005. [9] Wei Zhou, and Chandra Kambhamettu. "Estimation of illuminant direction and intensity of multiple light sources." In European conference on computer vision Springer Berlin Heidelberg, pp. 206-220, 2002. [10] Alin C Popescu, and Hany Farid. "Exposing digital forgeries in color filter array interpolated s." IEEE Transactions on Signal Processing, Vol. 53, no. 10, pp. 3948-3959, 2005. [11] Guohui Li, Qiong Wu, Dan Tu, and Shaojie Sun. "A sorted neighborhood approach for detecting duplicated regions in forgeries based on DWT and SVD." IEEE International Conference on Multimedia and Expo, pp. 1750-1753, 2007. [12] Sevinc Bayram, Husrev Taha Sencar, and Nasir Memon. "A survey of copy-move forgery detection techniques." IEEE Western New York Image Processing Workshop, pp. 538-542, 2008. [13] Guohui Li, Qiong Wu, Dan Tu, and Shaojie Sun. "A sorted neighborhood approach for detecting duplicated regions in forgeries based on DWT and SVD." IEEE International Conference on Multimedia and Expo, pp. 1750-1753, 2007. [14] A Fridrich, Jessica, B. David Soukal, and A. Jan Lukáš. "Detection of copy-move forgery in digital s." In in Proceedings of Digital Forensic Research Workshop. 2003. [15] Edoardo Ardizzone, Alessandro Bruno, and Giuseppe Mazzola. "Detecting multiple copies in tampered s." 17th IEEE International Conference on Image Processing (ICIP), pp. 2117-2120, 2010. [16] Lelong, Pierre, Govert Dalm, and Jan Klijn. "Image processing method and device for constructing an from adjacent s." U.S. Patent No. 5,444,478. 22 Aug. 1995. www.ijaetmas.com Page 56