压缩成像技术的应用与挑战  

Applications and challenges of compressed imaging

作　　者：张华[1] 曹良才[1] 金国藩[1] 白瑞迪 ZHANG Hua;CAO Liangcai;JIN Guofan;David J. Brady(State Key Laboratory of Precision Measurement Technology and Instrument;Department of Precision Instruments, Tsinghua University, Beijing 100084, China 2. Department of Electronic and Computer Engineering, Duke University, Durham, North Carolina 27708, USA)

机构地区：[1]清华大学精密仪器系精密测试技术及仪器国家重点实验室,北京100084 [2]杜克大学电子与计算机工程系,美国北卡罗莱那州达勒姆27708

出　　处：《科技导报》2018年第10期20-29,共10页Science & Technology Review

基　　金：国家自然科学基金项目(61327902)

摘　　要：概述了压缩成像的发展历史和研究现状。从边际成本的角度分析了压缩成像在不同电磁谱波段上的应用特点,指出压缩成像系统需要综合考虑特定应用场景下的器件成本和计算成本等因素设计采样模型和恢复算法。从全光函数的角度,介绍了利用压缩成像技术实现不同维度数据压缩采样的案例,分析了压缩成像技术的优势。探讨了压缩成像技术在理论突破和应用上面临的挑战,表明高稳定性和系统兼容性的压缩感知系统设计是压缩成像大规模应用面临的一个难题。This paper reviews the development and the applications of the modern compressive imaging in the whole electromagnetic spectra from the perspective of the marginal cost. The values of the compressive imaging need to be considered in the perspective of the hardware cost and the computational expense when the compressive models and the reconstruction algorithms are designed. Based on the plenoptic function, a couple of successful applications are presented to show the advantages in compressing different data. The challenges of the compressive imaging are presented in practical applications. For the large-scale applications of compressive imaging, it is a challenging task to design a compressed sensing model with high stability and good compatibility.

关 键 词：压缩感知 计算成像 全光函数 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]