SEI(BIO)+II: Metadata-rich Image Databases for Biodiversity Applications

SEI(BIO) II：用于生物多样性应用的元数据丰富的图像数据库

• 批准号: 0430769
• 负责人: Robert Morris
• 金额: --
• 依托单位: University of Massachusetts Boston
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2004
• 资助国家: 美国
• 起止时间: 2004-10-01 至 2007-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0430769&HistoricalAwards=false
• 关键词: SEI; BIO; +II; Metadata; rich

By exploiting the new JPEG2000 standard for image data, the project builds a flexible novel infrastructure by which images can both act as a mini-database about their content, and at the same time be part of a broader image database about their general subject. While the focus of this project is on biodiversity image libraries, the technology does not depend on the subject of the images. Because JPEG2000 supports complex structure of embedded image metadata and multiple images in the same JPEG file, highly structured information about the image content and origin conveniently remains with the image file itself and is available to anyone who acquires the image. Such a person or software agent can add more annotation to the image and circulate it further on the Internet. Roaming metadata encriched images (RMEIs) are used to automate some of the tasks involved in managing the curation of large image libraries.The project addresses image curation needs, and enhances the functionality, of the USDA PLANTS National Database, an important national plant information resource. By supporting the rapid collation of distributed annotation of plant images by both experts and the public, the project provides a facility to enhance early detection and response to invasive plants. A series of continuing workshops for a group of school teachers who are already accomplished nature photographers will produce models for image-centric biodiversity K12 instruction.

通过利用新的图像数据JPEG2000标准，该项目建立了一个灵活的新型基础设施，图像既可以作为关于其内容的小型数据库，同时又可以作为关于其一般主题的更广泛的图像数据库的一部分。虽然该项目的重点是生物多样性图像库，但该技术并不依赖于图像的主题。由于JPEG 2000支持嵌入式图像元数据的复杂结构和同一JPEG文件中的多个图像，因此有关图像内容和来源的高度结构化信息可以方便地保留在图像文件本身中，并可供任何获取图像的人使用。这样的人或软件代理可以为图像添加更多注释，并在互联网上进一步传播。 漫游元数据丰富的图像（RMEI）用于自动化管理大型图像库的策展所涉及的一些任务。该项目解决了图像策展需求，并增强了美国农业部植物国家数据库（一个重要的国家植物信息资源）的功能。通过支持专家和公众对植物图像的分布式注释进行快速整理，该项目提供了一个设施，以加强对入侵植物的早期检测和应对。为一群已经完成自然摄影师的学校教师举办的一系列持续研讨会将为以图像为中心的生物多样性K12教学制作模型。