CAREER: Statistical modeling and real-time correction of nonuniformity in array sensors

职业：阵列传感器不均匀性的统计建模和实时校正

• 批准号: 9733308
• 负责人: Majeed Hayat
• 金额: $ 21万
• 依托单位: University of Dayton
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1998
• 资助国家: 美国
• 起止时间: 1998-04-01 至 2002-03-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9733308&HistoricalAwards=false
• 关键词: CAREER; Statistical; modeling; real; time

The research component of this Faculty Early Career Development (CAREER) Program Award is addressing the problem of real-time correction of image degradation in infrared and laser imaging systems used in satellite imaging, aerial surveillance and reconnaissance, astronomical imaging, and target recognition. The types of degradation considered are due to the non-ideal characteristics of array sensors and atmospheric turbulence. The most limiting degradation in the performance of array sensors arises from the spatial and temporal nonuniformities in the sensors' responses. This research employs a statistical real-time-correction approach that avoids the need for costly calibration-based correction techniques that require the camera to steer away from the target for the duration of the calibration. The model-based statistical approach undertaken for nonuniformity correction is adapted to restore images degraded by atmospheric turbulence. The fundamental question of determining the optimal exposure time (or number of frames) needed for a given system in the presence of quantum noise is also being addressed. The algorithms developed in this research will be implemented in real systems through the investigator's collaborative efforts with the government and industry. The educational component of this research includes activities designed to motivate and excite undergraduate engineering students and strengthen the structure of the Dayton Area Graduate Studies Institute (DAGSI) Program that is a vital component of the Dayton-Area higher education system. These educational activities include: (1) demonstrating the practical research findings to undergraduate and graduate students by improving certain existing engineering courses; (2) offering periodical seminars available to nine local higher- education institutions and the DAGSI program through an interactive video network; (3) enhancing the practical training of students through collaboration with the local industry an d government laboratory; and, (4) actively involving the research assistants in this research in undergraduate education.

该学院早期职业发展(CALEAR)计划奖的研究部分致力于解决用于卫星成像、空中监视和侦察、天文成像和目标识别的红外和激光成像系统中图像退化的实时校正问题。所考虑的退化类型是由于阵列传感器的非理想特性和大气湍流。阵列传感器性能的最大限度下降来自于传感器响应的空间和时间不一致性。这项研究采用了一种统计实时校正方法，避免了需要昂贵的基于校正的校正技术，这些技术要求相机在校准持续时间内偏离目标。采用基于模型的统计方法进行非均匀校正，适用于恢复因大气湍流而退化的图像。还讨论了在量子噪声存在的情况下确定给定系统所需的最佳曝光时间(或帧数量)的根本问题。通过研究人员与政府和行业的合作，本研究中开发的算法将在实际系统中实现。这项研究的教育部分包括旨在激励和激励工科本科生的活动，并加强代顿地区研究生院(DAGSI)计划的结构，该计划是代顿地区高等教育系统的重要组成部分。这些教育活动包括：(1)透过改善某些现有的工科课程，向本科生及研究生展示实际研究成果；(2)透过互动视像网络，定期为本地9所高等院校举办研讨会及举办DAGSI计划；(3)与本地业界及政府实验室合作，加强学生的实践训练；及(4)积极让研究助理参与本科生的研究工作。