U.S.-Japan Joint Seminar: International Digital Library Annotation and Resource Discovery of Geographical Image Data

该奖项支持美国科学家参加将于2002年10月9日至11日在日本Karuizawa举行的关于地理图像数据的注释和资源发现的美日研讨会。联合组织者是布法罗纽约州立大学(SUNY)的张爱东教授和日本东京国家信息学研究所(NII)的弗雷德里克·安德烈斯(Frederic Andres)博士。本次会议将讨论当前美日项目关于大规模地理图像数据集的元数据建模、资源发现和查询的研究成果。研讨会旨在展示当前的研究成果，澄清各种问题，并确定未来联合工作的有希望的领域。具体内容包括：1)提取地理图像的语义特征，2)检测任意形状的地理图像聚类，3)制定集成系统的元数据以将查询定向到相关数据库，4)建立基于元数据的资源选择方法的理论基础，5)支持基于内容的地理图像检索的查询处理。这些方法为在地理图像数据库上创建支持分布式地理图像检索的元级系统奠定了基础。通过这个项目，收集了大量的地理数据(JIOMEDIA)，并根据语义对其进行了分类。用户可以出于研究和教育目的访问该资产。正在开发一套软件工具，将用于建立一个管理大规模地理数据资源的元服务器环境。研讨会通过许多博士后和研究生的参与促进了国际人力资源的发展，其中一些人发表了论文。通过思想和技术的交流，这个项目将扩大我们的基础知识基础，促进国际理解与合作。研究人员计划在他们的网页上公布研讨会的会议记录。

0137140ZhangThis award supports the participation of American scientists in a U.S.-Japan seminar on annotation and resource discovery of geographical image data to be held in Karuizawa, Japan from October 9-11, 2002. The co-organizers are Professor Aidong Zhang of the State University of New York (SUNY) at Buffalo and Dr. Frederic Andres of the National Institute of Informatics (NII) in Tokyo, Japan. This meeting will discuss research results of a current U.S.-Japan project on metadata modeling, resource discovery, and querying of large-scale geographical image datasets. The seminar aims to demonstrate the current research achievement, clarify a variety of questions, and identify promising areas for future joint work. The specific topics include: 1) extracting semantic features of geographic images, 2) detecting clusters of arbitrary shape of geographic images, 3) formulating the metadata for the integrated system to direct a query to relevant databases, 4) establishing a theoretical foundation of resource selection approaches based on the metadata, and 5) query processing to support content-based retrieval of geographic images.The researcher's current project investigates novel approaches to supporting effective and efficient access to various geographic image databases over the Internet. These approaches establish a foundation for creating a meta-level system on top of geographic image databases to support distributed geographic image retrieval. Through this project a huge volume of geographic data (JIOMEDIA) is being collected and classified based on the semantics. This asset will be accessible by users for research and educational purposes. A set of software tools is being developed and will be made available for use of establishing a metaserver environment which manages large-scale geographic data resources. The seminar advances international human resources through the participation of many postdocs and graduate students with some of them presenting papers. Through the exchange of ideas and technology, this project will broaden our base of basic knowledge and promote international understanding and cooperation. The researchers plan to publish proceedings of the seminar on their Web pages.