【作者】 宗靖国；

【导师】 张建奇；

【作者基本信息】 西安电子科技大学 ， 光学工程， 2011， 博士

【摘要】 近年来,随着光谱探测技术和光电成像技术的发展,红外成像光谱技术以其在资源勘探、自然灾害评估、军事目标检测和识别等方面的突出优势而受到了广泛的关注。但目前针对宽谱段的红外成像光谱技术的研究还很不足,数据源相对缺乏,制约了红外成像光谱技术在各方面研究和应用的开展。因此研究红外成像光谱获取技术及其在关键领域的应用,对于推动红外成像光谱技术的发展,进而促进该技术的应用具有非常重要的意义。本文的工作主要是研究红外成像光谱数据获取技术和基于实测光谱数据的红外光谱场景仿真方法。开展了如下工作:1.分析了当前光谱获取技术的发展现状和特点,设计并实现了一种基于单元型红外光谱辐射计的成像光谱数据获取系统。该系统是基于单元非成像型傅立叶红外光谱辐射计的基础而建立的。利用在红外光谱辐射计的视场光轴上添加二维光机扫描机构,通过控制二维光机扫描机构运动与光谱获取操作的协同,完成场景的二维空间分解与红外光谱辐射数据获取。该系统不但具有单元型傅立叶变换红外光谱辐射计的优点,而且还可以得到准确的物体空间信息。2.以系统的工作方式为基础,建立了本系统获取的光谱数据处理与重建体系架构。对获取的光谱数据进行光谱反演、标定、光谱维消噪和光谱数据空间二维重建等处理。首先实现了基于干涉光谱学基本原理的光谱数据反演和数据标定。接着利用小波分析理论对标定后的光谱数据进行了光谱维消噪。最后根据实际场景的空间分解方式,将一维时间轴上的光谱数据组重新映射到二维空间,完成数据的空间二维重建,进而获得光谱数据立方体。3.对红外成像光谱数据的应用展开了研究。利用获取的红外光谱数据,建立一种基于实测红外光谱数据的光谱场景仿真框架。采用通过实测获得的光谱成像数据立方体,结合场景建模,实现了市区停车场和旷野两个红外光谱场景的仿真。并对仿真结果进行分析。更多还原

【Abstract】 With the development of Electro-Optical Imaging and spectral detection technology, imaging infrared spectral technology and its outstanding advantages in the field of resources exploration, natural disaster assessment, and military target detection etc. attracts more and more attentions in recent years. However, at the present time, the wide band imaging infrared spectral technology is insufficiently researched, which is lack of data resources. Consequently, this shortage restricts the research and application of imaging infrared spectral technology in certain areas. Therefore,studying of imaging and spectral detection acquisition technology and applying to key sectors, is of a great significance for promoting of Imaging and spectral detection technology development, moreover, accelerating the technology further application.In this article we will investigate infrared imaging spectral data acquisition techniques and the method of synthetic infrared spectral scene generation based on measured data. The works are carried out will be described in the following aspects.First and foremost, analyze the features of spectral data acquisition techniques and the development status. Design and manufacture of a new imaging FTIR spectroradiometer which is established on the base of cell-based Fourier transform infrared spectroradiometer. In the system,a two-dimensional scanning mechanism was added in the axis of Fourier transform infrared spectroradiometer. Meanwhile, by controlling the motion of the two-dimensional scanning mechanism and the cooperation of spectral acquisition operations, it is complete data collection for the scene of the two-dimensional space decomposition and infrared spectral radiation. The system not only has advantages of cell-based Fourier transform infrared spectroradiometer but also can obtain accurate spatial information of objects.Secondly, according to the operation mode of the system, design software of spectral data management. The spectral data is processed by a series of procedure, such as spectrum recovery, calibration, denoising and Two-dimensional reconstruction of spectral data. Firstly, spectrum recovery and calibration was applied based on the basic principles of interference spectrum. Moreover, processing spectrum denoising utilizes the principle of wavelet analysis for the calibrated spectral data. Eventually, based on space decomposition method of the actual scene, we can get the spectral data cube.Finally, focus on the research of application for infrared imaging spectral data. By use of acquired infrared spectral data, establish the framework of the scene simulation based on the measured infrared spectral data. Apply imaging spectral data cube through measured and processed data acquisition method, combined with scene modeling, achieving two simulated infrared scene which is parking and wild scene. Meanwhile, the research analyzes simulation results.更多还原