利用編碼孔徑之高亮度高光譜成像系統;High-Brightness Hyperspectral Imaging System with Use of Coded Aperture

NCU Institutional Repository > 理學院 > 光電科學研究所 > 博碩士論文 > Item 987654321/83480

jsp.display-item.identifier=請使用永久網址來引用或連結此文件: http://ir.lib.ncu.edu.tw/handle/987654321/83480

题名:	利用編碼孔徑之高亮度高光譜成像系統;High-Brightness Hyperspectral Imaging System with Use of Coded Aperture
作者:	董澄逸;Tung, Cheng-Yi
贡献者:	光電科學與工程學系
关键词:	高光譜影像;編碼孔徑;快照式高光譜成像;Hyperspectral imaging;Coded aperture;Snapshot hyperspectral imaging
日期:	2020-08-11
上传时间:	2020-09-02 15:43:10 (UTC+8)
出版者:	國立中央大學
摘要:	不同於傳統針孔陣列式積分場光譜儀，需要在系統進光量及頻譜解析度之間做選擇。在快照式高光譜影像(Snapshot Hyperspectral Imaging)架構下，本團隊提出新型屏幕設計，利用編碼孔徑(Coded aperture)之編碼方式，取代傳統針孔陣列，以提升系統進光量。我們在相機前端加上閃耀光柵，以實現快照式高光譜影像之目的。此外，利用兩台相機同時擷取彩色影像與其對應的高光譜立方體。亦對編碼孔徑系統提出解碼演算法，並針對系統之波長定位、系統響應校正提出相對應適合之解決方案，透過結合編碼孔徑與高光譜概念，我們在不犧牲光譜解析度前提下建立了高亮度高光譜系統。並且，我們提出一套理論極限公式，連結了空間帶寬乘積與頻譜解析數量間之關係，可藉其快速設計出符合需求之編碼孔徑快照式高光譜影像系統。 ;In contrast to traditional Integral Field Spectroscopy with pinhole array, it trades off light collection for spectrum resolution. Our team proposed a novel screen design to replace the traditional pinhole array. We utilized coded aperture based on snapshot hyperspectral imaging system in order to enhance the light throughput. We mounted the blazed grating in front of the camera to achieve snapshot hyperspectral system. In addition, we used two cameras to acquire color image and its corresponding hyperspectral cube at the same time. We also find an algorithm to decode the coded aperture system, and seek the suitable solution to wavelength localization and system response calibration. Through combining the concept of coded aperture and hyperspectral, we build a high-brightness hyperspectral system without sacrificing spectral resolution. Then we promoted a theory limit formula, connecting space-bandwidth product and number of wavelength channels. So that we can design the coded aperture snapshot hyperspectral system according to our demand.
显示于类别:	[光電科學研究所] 博碩士論文

文件中的档案:

档案	描述	大小	格式	浏览次数
index.html		0Kb	HTML	179	检视/开启

在NCUIR中所有的数据项都受到原著作权保护.

社群 sharing

数据加载中.....