Developing a holistic model for digital library evaluation
Journal of the American Society for Information Science and Technology
| Hi-index | 0.00 |
In this paper, we suggest the Unified Video Retrieval System (UVRS) which provides the content-based query integrating feature-based queries and annotation-based queries of indefinite formed and great capacious video data, similarity query. And it supports the approximate query results by using the query reformulation in case that result of query does not exist. The UVRS divides a set of video into video documents, sequences, scenes and objects, and suggests the Three layered Object-oriented Metadata Model (TOMM) to model metadata. The TOMM is composed of a raw-data layer for physical video stream, a metadata layer to support the annotation-based retrieval, feature-based retrieval, and similarity retrieval and a semantic layer to reform the query. Based on this model, we suggest the video query language which make the annotation-based queries, feature-based queries based on color, spatial, temporal and spatio-temporal correlation and similar queries possible and Video Query Processor (VQP) to process the query. Specially, in case similarity queries on given scene or object, we present the formula expressing degree of similarity based on color, spatial, temporal order.If there is no query result, then it will be carry out query reformulation process which finds the possible attributes to relax the query and automatically reforms the query by using the knowledge from the semantic layer which is made on the basis of a Attribute Association Tree (AAT). We illustrate performance evaluation of similarity using recall and precision. The Suggested system is implemented by using the Visual C++, ActiveX and ORACLE. The UVRS can automatically input features of an object, color and spatial location, without image processing.