This paper considers the use of original content retrieval profiles for a peta-byte class digital video archive system that is currently installed in the Yamagata Video Archive Research Center (YRC). In this system, a large-scale automatic tape feed robot equipped with 7,200 volumes of videotapes is implemented. In addition, this paper also discusses how the MPEG-7 can be used for constructing the original content profile and for implementing the content retrieval application software. A series of verification tests such as a comparison between a conventional flat-type, text-based retrieval system and a scalable, high-level MPEG-7-based description retrieval system were conducted at our lab. And the results showed the effectiveness of our proposed original content profile and the significant improvement in overall retrieval efficiency.
展开▼
