1 中国科学院太空应用重点实验室, 北京 100094
2 中国科学院空间应用工程与技术中心, 北京 100094
3 中国科学院大学, 北京 100049
近年来随着高光谱成像技术的快速发展, 航天高光谱遥感数据在各领域应用研究中取得了良好的发展与突破。文中回顾了国内外航天高光谱成像技术的发展历程, 介绍了有代表性的航天高光谱成像仪的主要应用技术指标, 较为系统地总结和分析了近五年来航天高光谱遥感数据在国土资源、农林遥感、海洋湖泊遥感、城市环境、灾害监测及其他方面等各领域的最新应用研究进展。对基于AI技术的高光谱信息提取与应用、基于高光谱遥感的多源数据融合与应用以及面向深空探测领域的高光谱数据分析与应用等发展趋势做了展望, 未来航天高光谱成像仪技术的进一步突破和应用研究需求的牵引将会推动高光谱应用领域更大范围的创新与发展。
航天高光谱 成像光谱仪 高光谱数据应用 aerospace hyperspectral imaging spectrometer application of hyperspectral data 红外与激光工程
2019, 48(3): 0303001
Author Affiliations
Abstract
1 Department of Electrical and Systems Engineering, Washington University in St. Louis, St. Louis, Missouri 63130, USA
2 Department of Electrical Engineering, Pennsylvania State University, University Park, Pennsylvania 16802, USA
Conventionally, metallic nanostructures are used for surface-enhanced Raman spectroscopy (SERS), but recently there has been increasing interest in the enhancement of Raman scattering from dielectric substrates due to their improved stability and biocompatibility compared with metallic substrates. Here, we report the observation of enhanced Raman scattering from rhodamine 6G molecules coated on silica microspheres. We excite the whispering gallery modes (WGMs) supported in the microspheres with a tapered fiber coupler for efficient WGM excitation, and the Raman enhancement can be attributed to the WGM mechanism. Strong resonance enhancement in pump laser intensity and modified Raman emission from the Purcell effect in the microsphere resonator are observed from the experiment and compared with theoretical results. A total Raman enhancement factor of 1.4×104 is observed, with contribution mostly from the enhancement in pump laser intensity. Our results show that, with an efficient pumping scheme, dielectric microspheres are a viable alternative to metallic SERS substrates.
Spectroscopy, Raman Surface-enhanced Raman scattering Resonators Photonics Research
2018, 6(5): 05000346
北京理工大学 信息与电子学院, 北京 100081
为提高步态识别准确率, 提出了基于空-频域特征和线性判别分析的视频步态识别方法。利用离散余弦变换、Contourlet变换分别提取步态能量图的频率特征和多尺度多方向轮廓特征; 融合得到空-频域特征, 并通过线性判别分析映射到最佳鉴别矢量空间; 根据相似性距离实现身份识别。在中科院自动化所提供的数据库中进行实验, 结果表明, 提出的特征提取方法优于现有常用方法。空-频域特征能够有效地区分步态中的高低频分量,并捕捉丰富的细节信息, 线性判别分析在降维的同时进一步增强特征的判别能力, 有助于提高识别精度。
步态识别 步态能量图 离散余弦变换 Contourlet变换 线性判别分析 gait recognition gait energy image discrete cosine transform contourlet transform linear discriminant analysis
