U.S.-Egypt Cooperative Research: High Performance Techniques for Remote Sensing

描述：该项目支持康涅狄格大学斯托尔斯分校计算机科学与工程系Sanguthevar Rajasekaran博士的合作研究。合作的埃及科学家是埃及开罗电子研究所的Salwa Nassar博士。他们计划研究高性能的遥感技术。遥感在国土安全和气候研究等各个领域都具有重要意义。遥感问题经常处理海量数据集，这些数据集的处理需要很高的计算运行时间，这可能是不可接受的。获得可接受的运行时间(最好是实时性能)的一种方法是借助高性能技术。通用并行编译器可能不会有太大帮助，因为涉及的问题性质复杂。私人投资机构计划解决对高性能遥感理解方面的这一差距。特别是，他们计划开发高性能的算法来解决遥感中的基本问题，如成像、图像分割和反射率模型的开发。将考虑三种实现并行性的方法：使用超级计算机、使用本地集群和利用网格计算。将在基本方面做出贡献，包括算法的开发、调度和软件性能分析。在解决遥感中的具体问题时，也将同样重视开发的算法的实施。智力优势：发挥高性能计算在解决遥感问题中的作用。本项目涉及的研究问题需要一套互补的专门知识。拟议的国际集团将成为一个可行的模式，今后可以被其他国家效仿。受过这个项目培训的学生将能够在他们的职业生涯中建立这样的合作。广泛的影响：在过去的三十年里，算法领域产生了出色的研究和聪明的算法，并解决了许多基本问题。然而，这些进步的全部影响并没有那么广泛地感受到，因为向现实世界应用程序的传递机制充其量是不系统的。在这个项目中，重点是将算法进步传输到对埃及和美国乃至整个世界至关重要的遥感应用领域。该项目的研究成果将极大地惠及这些国家的产业和教育机构。受训的学生将在合作方面获得独特的技能和专业知识。所取得的进展将在短期内影响这一做法。因此，该项目的成果在不久的将来可能会在准确的天气预报等领域对社会产生影响。该项目得到了美国-埃及联合基金计划的支持，该计划为两国的科学家和工程师提供资助，以开展这些合作活动。

0710900RajasekaranDescription: This project supports collaborative research by Dr. Sanguthevar Rajasekaran, Department of Computer Science and Engineering, University of Connecticut at Storrs, Connecticut. The collaborating Egyptian scientist is Dr. Salwa Nassar, Electronics Research Institute, Cairo, Egypt. They plan a study of high performance techniques for remote sensing. Remote sensing is of great importance in varied domains such as homeland security and climate studies. Remote sensing problems often deal with voluminous datasets whose processing demands high computational run times that may not be acceptable. One way of achieving acceptable run times (preferably real-time performance) is with the help of high-performance techniques. General purpose parallel compilers may not be of much help because of the complex nature of the problems involved. The PIs plan to address this gap in the understanding of high-performance remote sensing. In particular, they plan to develop high-performance algorithms for solving fundamental problems in remote sensing such as imagoedesy, image segmentation and development of reflectance models. Three ways of achieving parallelism will be considered: using supercomputers, employing local clusters, and utilizing grid computing. Contributions will be made on the fundamental aspects including the development of algorithms, scheduling, and software performance analysis. An equal emphasis will also be placed in implementing the developed algorithms to solve specific problems in remote sensing.Intellectual Merits: The role of high-performance computing in solving remote sensing problems will be brought out. The research problems addressed in this project need complementary sets of expertise. The proposed international group will serve as a viable model that can be emulated by others in future. Students trained in this project will be able to setup collaborations of this kind in their career.Broader Impacts: Over the past three decades, the area of algorithms has produced excellent research and clever algorithms and has solved numerous fundamental problems. Yet the full impact of these advances is not as widely felt since the transmission mechanism to real world applications has been unsystematic at best. In this project, the focus is on the transmission of algorithmic advances to remote sensing application areas that are of vital importance to Egypt and the US and also to the world at large. Research outcomes of this project will greatly benefit industries and educational institutes in these countries. Students trained will acquire unique skills and expertise in the aspects of collaborations. Developments made will impact the practice in the short run. Thus the outcomes of this project can impact the society in the near future, in areas such as in accurate weather forecasting. This project is being supported under the US-Egypt Joint Fund Program, which provides grants to scientists and engineers in both countries to carry out these cooperative activities.