International Journal of Engineering Trends and Technology (IJETT) – Volume 9 Number 1 - Mar 2014 The Way to Preserve Copyrights for Video Data and Elimination of Untrusted User’s from the Network K. Gowthami1, Kavitha Esther Rajakumari2 M.E Computer Science and Engineering1, Faculty of Computing2,Sathyabama University1, Sathyabama University2 Chennai-600119, India ABSTRACT-Due to rapid development in the field of multimedia there is a scope for redundancy in video data. So copyright violation is encountered as a serious problem. There exist a large amount of data which is manipulated and ill legal copies have been published in many social network sites (e. g. YouTube). In order to prohibit this redundant video data many methods have been proposed. Primary methods are based on water marking, content based copy detection (CBCD) and SIFT. In water marking, if the original image is not water marked it cannot detect the copied images. In content based copy detection, signature limitation has become a problem. Scale Invariant Feature Transform (SIFT) encounters a problem of computational cost and is not ease for large applications. In this paper three methods are proposed viz. Auto Dual threshold method, SIFT Singular Value Decomposition (SVD) and Graph based method. By applying these methods redundant frames can be eliminated, feature extraction made easy with SVD and matching the original data with copied data. Keywords— Water Marking, CBCD, Auto Dual Threshold, SVD, SIFT, Graph Based Sequence method. I. INTRODUCTION In this proposed work, a segmentation and graphbased video sequence matching method for Content Based Video Copy Detection. In particularly, SIFT (Scale Invariant Feature Transform) descriptor for video content description is utilized. But, the matching based on SIFT descriptor is quite computational expensive to extract the robust key point descriptors and perform pair wise matching of a large number of local features frame by frame. So that, initially the dual-threshold method is utilized and it is segmenting the videos into segments. After that, utilized an SVD-based technique to match two video frames with the SIFT point set descriptors. Finally, to get the video sequence matching result after then propose a graph- based method. It can convert the video succession ISSN: 2231-5381 matching into the longest path in the frame matching-result with time limit. The proposed work shows that the segmentation and graph-based video sequence matching methods can detect video copies efficiently. Here the server will identify misbehaving user in the proposed system and block the user for particular time. II. RELATED ARTICLES Many content-based video copy detection methods have been proposed in addition, copy is a subset of near replica. There are many methods proposed on nearduplicate detection. Z.Huang, H.T.Shen, J.Shao, B.Cui (2010) developed a paper for Practical Online Near-Duplicate Subsequence Detection for continuous video streams. Here Water-Marking Technique is used. CBCD has a signature has limitation; example can be two distinct broadcasts. Hence, proposed to transform a video stream in to one-dimensional Video Distance Trajectory (VDT) monitoring continues changes of consecutive frames with respect to a reference point, which is further segmented and represented by a series of compact signatures called Linear Smoothing function. O.Kucuktunc, U.Gudukbay (2010) developed a paper for Fuzzy Color Histogram –Based video segmentation. Here a video is composed of several shots with abrupt or gradual changes. For frame similarity Discrete Cosine Transform is being used. For identifying PiP, small changes efficient detection algorithm used. Hampapur et al. compared distance measures and video sequence matching methods for video copy detection. They employed convolution for motion direction feature, L1 distance for ordinal intensity signature (OIS), and histogram intersection for color histogram feature. The consequences show that the method using OIS performs better. Yuan et al. collective OIS with color histogram feature as a tool for describing video sequence. http://www.ijettjournal.org Page 32 International Journal of Engineering Trends and Technology (IJETT) – Volume 9 Number 1 - Mar 2014 Particularly, they divided each video frames the length of the time sequence into several blocks and proposed average gray values for each block. Then, they associated gray values of these alienated blocks individually along the time direction before they use those sequence information to describe the video content.An early method for detection of video data, based on colour histogram intersection is proposed by Satoh. Yeh and Cheng use a method that divides the picture in to four regions, and mines a Markov Stationary Feature (MSF)-extended HSV colour histogram. Basharat et al. present a video-matching framework using spatio-temporal segmentation. A set of features (colour, texture, motion, and SIFT descriptors) is extracted from each segment, and the comparison between two videos is computed with a bipartite graph and Earth Mover’s Distance (EMD). Williams et al. propose a video copy detection method based on efficiently matching local spatiotemporal feature points with a disk-based indexing scheme. In general, extracting and matching points of interest are costly operations in terms of computation time. Bertini et al. present a clip-matching algorithm that use video fingerprint based on standard MPEG7descriptors. An effective combination of color layout descriptor (CLD), scalable color descriptor (SCD), and edge histogram descriptor (EHD) forms the finger print. Fingerprints are extracted from each clip, and they are compared using an edit distance. Sarkar et al. use CLD as video fingerprints and propose a non-metric distance measure to efficiently search for matching videos in high dimensional space. III. SUMMARY OF EXISTING SYSTEM In existing system SIFT descriptor is being used for video content description. In particularly, SIFT (Scale Invariant Feature Transform) descriptor for video content description is utilized to the reason of the Discriminative ability and good stability of local features. However, matching based on SIFT descriptor is computationally expensive for large number of points and the high dimension. Drawbacks in the existing system are server will not identify the misbehaving user and also user tries to login and same query is being published many times. ISSN: 2231-5381 Fig 1: Functional Architecture Server monitors each and every users query videos. It will continuously monitor the query videos of each and every user’s communication. When monitoring the server will identify the copied frames from the input queries i.e. the matching frame result is been verified from the database which is already been trained. It identifies the unauthorized user and block that identified user at particular time Here additional opportunity is provided to all users’ i.e. (user’s login limitation). Suppose if attempt of user is crossed the limit then that particular user will discard from the network. IV. PROPOSED SYSTEM For categorizing redundant video data effectively and efficiently, initially the dual-threshold method is exploited and it is slicing the videos into segments with the at ease of homogeneous and then the key frame is extract from every segment. The SIFT features are extracted from that segments of key frames. After that, exploited an SVDbased technique to match two video frames with the SIFT point set descriptors. Absolutely, to get the video sequence matching outcome propose a graph- based method. It can transfer the video sequence matching into discovering the best ever path in the frame matching-result with time constriction. If the query video is equivalences then servers identifies the illicit user and block that spotted user at certain time. And over again if user prolongs to upload video then user will be isolated from network. http://www.ijettjournal.org Page 33 International Journal of Engineering Trends and Technology (IJETT) – Volume 9 Number 1 - Mar 2014 V. IMPLEMENTATION DETAILS D. SIFT-SVD A. User’s Video Query Query videos are analyzed. Features are extracted from these videos and compared to those stored in the referential database. The matching results are then analyzed and the detection results are returned. B. Extraction and Compression of frames For analyzing the video frames are being extracted and Compression is the decrease in size of data in order to save space or transmission time. Graphic image compression can be either lossy or lossless. A lossless frame compression is done. The use SVD method to measure the similarity between two SIFT feature point sets, and emphasize the similarity of “frame-to-frame”. Specifically, the goal of the SVD method to reduce and correct the wrong match between the two points in two SIFT feature sets. Characteristics: 1. Transposition and substitute 2. Energy Concentricity. invariance. E. Graph Method for Video Similarity Checking The graph-based video sequence matching method for video copy detection. The method is obtainable as follows: C. Elimination of Redundant Frames 1. 2. 3. Fragment the video frames and extract features of the key frames. Go with the query video and target video. Produce the matching result graph according to the matching results. - Time direction consistency: Condition 1 indicates that if the query video is a copy deriving from the target video, then the video subsequence temporal order amid query video and target video must be consistent, which is reasonable in real application. Fig 2: Imitative Image The above image is the replica of original image. So this frame is the redundant frame and this should be identified using Auto Dual Threshold method and should be removed. -Time jump degree: If Condition 1 is satisfied, Condition 2 is used to limit the time span of two matching results between the query video and the target video. If the time span exceeds a assured threshold, it is considered that there does not exist certain correlation between the two matching results. F. Secure Accessing If the user’s video is matched with original video then user will be blocked for certain time limit and again if user continues to upload video then user will be removed from network. Fig 3: Novel Image An auto dual-threshold method to eliminate redundant video frames. This method cuts incessant video frames into video segments by eliminating temporal replica of the visual information of continuous video frames. This method has the subsequent two characteristics. First, two thresholds are used. Particularly one threshold is used for detecting sudden changes of visual information of frames and another for regular changes. Second, the values of two thresholds are deter-mined adaptively according to video content. The auto dual-threshold method to remove the redundant frames. ISSN: 2231-5381 VI. PERFORMANCE EVALUTION From the view of video signature, LSF and CONT-CONX use the global signature, and SVD-SIFTGRAPH and MRF use the local signature even though the methods based on global signature can detect copy videos to some level, its capability to identify the copy videos with complex transformations (such as PiP, Shift, crop) is limited. Thus, in usefulness, the methods based on local signature execute better than the ones based on global signature. On the other hand, with respect to competence http://www.ijettjournal.org Page 34 International Journal of Engineering Trends and Technology (IJETT) – Volume 9 Number 1 - Mar 2014 the methods based on global signature (LSF, CONTCONX) have more advantages over the ones based on local signature (SVD-SIFT-GRAPH, MRF). The following graph shows the comparison among the methods mentioned above in terms of effectiveness and efficiency. REFERENCES [1] Z. Huang, H.T. Shen, J. Shao, B. Cui, and X. Zhou, “Practical Online Near-Duplicate Subsequence Detection for Continuous Video Streams,” IEEE Trans. Multimedia, vol. 12, no. 5, pp. 386-397,Aug. 2010. . [2] O. Kucuktunc, U. Gudukbay, and O. Ulusoya, “Fuzzy Color Histogram-Based Video Segmentation,” Computer Vision and Image Understanding, vol. 114, no. 1, pp. 125-134, 2010. [3] A. Hampapur, K. Hyun, and R. Bolle, “Comparison of Sequence Matching Techniques for Video Copy Detection,” Proc. SPIE,Storage and Retrieval for Media Databases, vol. 4676, pp. 194-201, Jan.2002. Performance Evalution [4] X. Wu, C.-W. Ngo, A. Hauptmann, and H.-K. Tan, “Real-Time NearDuplicate Elimination for Web Video Search with Content and Context,” IEEE Trans. Multimedia, vol. 11, no. 2, pp. 196-207,Feb. 2009. s [5] H. Liu, H. Lu, and X. Xue, “SVD-SIFT for Web Near-Duplicate Image Detection,” Proc. IEEE Int’l Conf. Image Processing (ICIP ’10),pp. 1445-1448, 2010. 300 250 200 150 100 50 0 1 2 3 [6] H.T. Shen, J. Shao, Z. Huang, and X. Zhou, “Effective and Efficient Query Processing for Video Subsequence Identification,” IEEE Trans. Knowledge and Data Eng., vol. 21, no. 3, pp. 321-334, Mar.2009. 4 [7] M. Douze, H. Je´gou, and C. Schmid, “An Image-BasedApproach to Video Copy Detection with Spatio-TemporalPost-Filtering,” IEEE Trans. Multimedia, vol. 12, no. 4,pp. 257-266, June 2010. Fig 4: Comparison Among Methods From the graph it is shown SIFT-SVD can obtain a better trade-off between the effectiveness and the efficiency. 1-SIFT-SVD 2-LSF 3-CONT-CONX 4-MRF [8] X. Zhou, L. Chen, A. Bouguettaya, Y. Shu, X. Zhou, and J.A.Taylor, “Adaptive Subspace Symbolization for Content-Based Video Detection,” IEEE Trans. Knowledge and Data Eng., vol. 22,no. 10, pp. 1372-1387, Oct. 2010. [9] Final CBCD Evaluation Plan TRECVID 2008 (v1.3), http://wwwnlpir.nist.gov/projects/tv2008/Evaluation-cbcd-v1.3.htm,2008. [10] E. Delponte, F. Isgro` , F. Odone, and A. Verri, “SVD-Matching Using Sift Features,” Graphical Models, vol. 68, no. 5, pp. 415-431, 2006. [11] TREC Video Retrieval nlpir.nist.gov/projects/t01v/, 2006. Evaluation, http://www- [12]Final CBCD Evaluation Plan TRECVID 2010(V2),http://wwwnlpir.nist.gov/projects/tv2010/Evaluation-cbcdv1.3.htm#eval, 2010. VII. CONCLUSION This paper shows that three methods are proposed viz. Auto Dual threshold method, SIFT Singular Value Decomposition (SVD) and Graph based method. By applying these methods redundant frames can be eliminated, feature extraction made easy with SVD and matching the original data with copied data and SIFT-SVD can obtain a better trade-off between the effectiveness and the efficiency. Additionally. the server will identify misbehaving user in the proposed system and block the untrusted .user from the network. [13] A. Hampapur and R. Bolle, “Comparison of Distance Measures for Video Copy Detection,” Proc. IEEE Int’l Conf. Multimedia and Expo (ICME), pp. 188-192, 2001. ACKNOWLEDGMENT We would like to thank Dr.B.Bharathi, Head of the Department, Department of Computer Science and Engineering, Ms.Kavitha Esther Rajakumari for her encouragement and support. ISSN: 2231-5381 http://www.ijettjournal.org Page 35