[attachment=10680]
ABSTRACT
Recent development in multimedia processing and network technologies has facilitated the distribution and sharing multimedia through networks. To increase the security demands of multimedia contents, traditional image content protection schemes use extrinsic approaches, such as watermarking or fingerprinting. However, under many circumstances, extrinsic content protection is not possible. To solve these problems forensic tools via intrinsic fingerprints are developed. Source coding is a common step of natural image acquisition. To focus the digital image source coder forensics via intrinsic fingerprints, the unique intrinsic fingerprint of many popular image source encoders are taken as the evidence for security. Based on the intrinsic fingerprint of image source coder, forensic detector identifies which source encoder is applied, what the coding parameters are, along with confidence measures of the result. Hence the image transmitted from sender to receiver is secured
Existing System
COPYRIGHT infringements and data piracy have recently become serious concerns for the ever growing online video repositories. Videos on commercial sites e.g., http://youtube.com and http://metacafe.com, are mainly textually tagged. These tags are of little help in monitoring the content and preventing copyright infringements. Approaches based on content-based copy detection (CBCD) and watermarking have been used to detect such infringements. The watermarking approach tests for the presence of a certain watermark in a video to decide if it is copyrighted. The other approach, CBCD, finds the duplicate by comparing fingerprint of the query video with the fingerprints of the copyrighted videos. A fingerprint is a compact signature of a video which is robust to the modifications of the individual frames and discriminative enough to distinguish between videos. The noise robustness of the watermarking schemes is not ensured in general, whereas the features used for fingerprinting generally ensure that the best match in the signature space remains mostly unchanged even after various noise attacks. Hence, the fingerprinting approach has been more successful.
PROPOSED SYSTEMS
In contrast to the fast sequential search scheme applying temporal pruning to accelerate the search process which assumes query and target subsequence are strictly of the same ordering and length, this approach adopts spatial pruning to avoid seeking over the entire database sequence of feature vectors for exhaustive comparison.
This approach does not involve the presegmentation of video required by the proposals based on shot boundary detection shot resolution, which could be a few seconds in duration, is usually too coarse to accurately locate a subsequence boundary. Meanwhile, this approach based on frame sub sampling is capable of identifying video content containing ambiguous shot boundaries
This robust model for evaluating video similarity is not only based on the percentage of similar frames which in essence ignores the temporal characteristic of videos. The advantage of this method is that, it is not only based on average distance of frame pairs to capture visual content, but also well considers temporal order and frame alignment.
Particularly, the tolerance to permutation of elements makes it distinguished from the widely used Edit distance, since it can account for the cross mapping relationship in the presence of content reordering. Given a query in the form of “ABCD,” this method can correctly rank a video “ACBD” higher than “AADD,” while Edit distance scores them equally with respect to the query. Therefore, this method can better resemble the “similar” notion of human perception. Good performance has also been demonstrated in the experiments. Further search videos with changes from query due to content editing, a number of algorithms have been proposed to evaluate video similarity.
Module Description:
1. Extraction
2. Mapping
3. Refine Search
Extraction:-
In this module it will extract both the main and clip video in frames according to their video size and store in the selected location in the form of Image.
Mapping:-
In this module it will map first image of both clip and main video and check weather they are equal or not.
Refine Search:-
In this module it will take all the images of both clip and main video and filter that images and then compare that images.
Requirements :-
Software Requirements:
IDE : Visual Studio.NET 2008
Language : C#.NET
Framework : .NET Framework
Hardware Requirements:
Processor : Pentium IV 2.4 GHz
Ram : 1GB
Hard disk : 160 GB