SBIR Phase II: High Resolution, Synthetic Satellite Imagery of the Earth

点击翻译按钮获取中文摘要

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase II project will be to address a strong commercial and scientific need for operational synthesis of spatially consistent and temporally relevant historical and current high resolution satellite imagery for analytical purposes. The technology will contribute to the advancement of scientific knowledge especially in the geospatial arena and to market spillovers. By dramatically simplifying access to accurate imagery for any time and place, this technology will provide companies, researchers, educators, students, and regular citizens with a valuable tool for visualizing and exploring our changing planet and will contribute to increasing public engagement with science and technology. Further, the analytical capabilities offered by the imagery have great potential in scientific applications thus contributing to partnerships between academia and industry and improving datasets for research and education. Finally, this technology will be valuable in operational settings at the large providers of commercial satellite imagery, to individual users, and enable a wide variety of new visualization, analysis, and data mining applications. The examples of commercial applications for the technology are virtual earth, insurance and reinsurance, agriculture, emergency, change detection.This Small Business Innovation Research (SBIR) Phase II project will operationally synthesize accurate regional-to-global, high spatial / high temporal satellite imagery of the Earth. The technology will utilize advanced data fusion algorithms to combine various sources of imagery while preserving the best spatial and temporal attributes of the data sources. The complexity of accessing, processing, and analyzing various sources of satellite imagery creates a significant barrier to its use. Synthesis of regionally and globally continuous high spatial / high temporal resolution imagery is a challenge as in addition to inherent differences in spatial and temporal resolutions of the source data, the new models need to account for enormous data volumes and sparse coverage of high spatial resolution imagery. Existing techniques to handle these challenges have severe limitations which curtail their use outside of the research arena. The technology will overcome these limitations by implementing algorithms that are robust, automated, scalable, deliver accurate data, and are usable in operational settings. They will provide spatially consistent and temporally relevant imagery which will empower businesses with regional and global outreach to make better decisions with better data.