워터마킹을 이용한 동영상 데이터의 저작권 보호와 인증

Abstract

Due to the rapid development of computer and IT technologies, the problems of IPR (Intellectual Property Rights) are highly increasing as multimedia contents such as image, audio, and video are illegally copied and distributed. The illegal copy and distribution of multimedia contents are a serious problem as they discourage author's desire of creative writing and bring tremendous economic loss. Therefore, the copyright protection of multimedia data requires authentication techniques that can verify content authentication, ownership authentication, and illegal copy. Such technologies to protect multimedia data as cryptography method and access control technology are highly developed, but once the multimedia contents whose code is cracked, they can be illegally copied and distributed. In addition, the multimedia contents can be illegally used even by those users who are allowed to access. Therefore, it is difficult to protect copyright and ownership effectively using the existing technologies in the rapidly-developing content market and electronic commerce. Therefore, in recent years, digital watermarking gains attention because it includes ownership information into the multimedia content itself to prevent illegal copy or distribution and to protect copyright effectively. The digital watermarking method applies its classic watermarking concept to multimedia data to protect copyright. Recently, more powerful watermark technologies are studied using many technical methods, while avoiding from the method of simply inserting watermark. Most studies on the existing watermarking technology are targeted on still image, but transparency and stiffness should be offered by inserting invisibly coded digital watermark into the image. For this purpose, elaborate recognition information should be used in the processing of watermarking. This study examines digital watermarking method and cut detection algorithm using Mean Square Error (MSE) to protect the copyright of video data or multimedia content. Cut detection algorithm extracts key frame that is gradually changed, by comparing and analyzing video data frames. The extracted frames use image application watermarking to the conversion of wavelet that is frequency based watermark, with the perspective of human oriented visual system. This study suggests a system that allows users to protect ownership (copyright) and make authentication while inserting and extracting copyright information into the key frame, without marring the multimedia content or video data.