MRI: Development of an Instrument for Acquisition, Management, and Analysis of Super-resolution Aerial Imagery

该项目开发了一个初步的示范原型，旨在朝着制定一项文书迈出第一步，该文书将有助于在减灾方面作出明智的决策，从而大大造福于整个社会。这项研究将有利于环境监测、交通、教育、农业、城市规划以及公共卫生和安全方面的应用。该项目将通过让更多的少数民族学生参与研究，接触当地学校、媒体和互联网，并让学生与工业合作者接触，进一步扩大教育、招聘、留住和指导的使命。TerraFly基于网络的数据传播工具在这个少数民族服务和国际化的机构之前的工作为研究人员和学生提供了向广泛受众传播大型项目的经验，利用全球媒体的报道。具体来说，该项目确定了所需的数据集，允许研究人员指定地理和时间选择以及要混合的多种数据类型，以及可以由用户开发或上传的分析功能。该仪器将建立在TerraFly之上，继承其数据存储，映射，高性能查询，API，空间连接，屋顶地理编码，数据钻取和时空可视化模块。最终的仪器将是一个综合系统，包括：一个服务器群;一个包含所提供和创建的数据集的庞大数据库;一队气球载航空平台，以流传输多光谱图像和环境测量数据;批量和临时数据分析和知识提取模块;以及一个TerraFly扩展程序，用于在数千个其他数据集的背景下显示和查询超分辨率航空图像。该仪器目前的示范原型将包括一个大型数据库，其中载有提供的和创建的样本数据集;从几个球载传感器获取的图像，以及一个TerraFly扩展程序，用于在其他几个数据集的范围内显示和查询超分辨率航空图像。虽然演示原型将只涉及该研究所空气空间中的几个气球传感器，但将详细设计一种设计，以防止在最终仪器中更广泛地部署无系留气球时出现任何空中交通和环境问题。演示原型将获取，加载，存储，分析，处理，创建和提供地理空间数据，包括：传统的栅格和矢量地理空间数据，样本地理投影视频流，NASA提供的高分辨率数据，由承诺的工业合作伙伴提供的超分辨率航空图像样本，以及样本分析的地理参考文档。在先前的资助下，TerraFly地理空间数据管理项目开创了许多技术，这些技术成为主要地图提供商后来部署的榜样。拟议的原型是一种仪器，将部署数量级更清晰的航空图像，这将使大量的新应用和分析成为可能。由于其非侵入式低空超分辨率传感和自动知识提取，该仪器将使12个小组的多学科研究成为可能。

This project, developing an initial demonstration prototype, aims to make a first step towards an Instrument that will enable intelligent decision-making in disaster mitigation, greatly benefiting society at large. The enabled research will benefit applications in environmental monitoring, transportation, education, agriculture, urban planning, and public health and safety. This project will contribute to further broaden the mission of education, recruitment, retention and mentoring by engaging more minority students in research, by reaching out to local schools, media, and the Internet, and by engaging the students with industrial collaborators. Prior work on the TerraFly web-based data dissemination tools in this minority-serving and internationally-geared institution affords the researchers and students experience in disseminating large projects to a broad audience, leveraging coverage received by worldwide media.Specifically, the project identifies the required datasets, allowing researchers to specify geographical and temporal selection and the multiple data types to be mashed up, and the analytical functions that may be developed or uploaded by the user. The Instrument will be built on top of TerraFly, inheriting its modules of Data Store, Mapping, High Performance Query, API, Spatial Join, Rooftop Geocoding, Data Drilldown, and Spatiotemporal Visualizer. The eventual Instrument will be an integrated system, comprised of: a server farm; a massive database containing contributed and created datasets; a fleet of balloon-borne aerial platforms streaming multi-spectral imagery and environmental measurements; modules for batch and ad-hoc data analytics and knowledge extraction; and a TerraFly extension to display and query super-resolution aerial imagery in the context of thousands of other datasets. The present Demonstration Prototype of the Instrument will include a large database containing sample contributed and created datasets; acquisition of imagery from several balloon-borne sensors, and a TerraFly extension to display and query super-resolution aerial imagery in the context of several other datasets. While the Demonstration Prototype will involve only several balloon-borne sensors in the airspace of the Institution, a design will be elaborated to prevent any air traffic and environmental problems in the broader deployment of untethered balloons in the eventual Instrument. The Demonstration Prototype will acquire, load, store, analyze, process, create, and serve geospatial data, including: traditional raster and vector geospatial data, sample geo-projected video streams, high-resolution data contributed by NASA, sample super-resolution aerial imagery contributed by committed industrial partners, and sample analyzed geo-referenced documents. Under prior funding, the TerraFly Geospatial Data Management project pioneered a number of technologies that became role models for later deployment by major mapping providers. The proposed Prototype is of an Instrument that would deploy an order-of-magnitude sharper aerial imagery, which will enable vast new applications and analytics. Due to its non-invasive low-altitude super-resolution sensing and automatic knowledge extraction, the Instrument will enable multi-disciplinary research of twelve groups.