主成分分析方法在空间干涉仪图像处理中的应用
发布时间：2018-11-29 12:27
【摘要】：在干涉仪图像数据处理过程中,目前采用的多行平均图像处理算法会引入较大随机误差,且当CCD相机与狭缝之间存在小转角时,会引入较大系统误差。本文主要探究利用主成分分析(PCA)算法处理空间干涉仪图像的可行性与优势。利用MATLAB模拟空间干涉仪图像,并加入随机噪声和图像旋转,利用PCA方法和多行平均算法处理数据,比较两种算法的得到的结果误差大小。并设计CCD相机小转角实验和干涉图像弱信号实验,评估PCA算法在数据处理过程中纠正CCD相机小转角和重建弱信号图像中的效果。理论和实验均证明,PCA算法较目前多行平均算法,能更有效地降低噪声,分析弱信号图像及纠正CCD相机小转角,消除其带来的系统误差。
[Abstract]:In the process of interferometer image data processing, the current multi-line average image processing algorithm will introduce a large random error, and when there is a small angle between the CCD camera and the slit, a large system error will be introduced. This paper mainly explores the feasibility and advantages of using principal component analysis (PCA) algorithm to process spatial interferometer images. The spatial interferometer image is simulated by MATLAB, random noise and image rotation are added, and the data are processed by PCA method and multi-line average algorithm, and the error between the two algorithms is compared. The small angle experiment of CCD camera and the weak signal experiment of interference image are designed to evaluate the effect of PCA algorithm in correcting the small angle of CCD camera and reconstructing weak signal image in the process of data processing. The theoretical and experimental results show that the PCA algorithm is more effective than the current multiline average algorithm in reducing noise, analyzing weak signal images and correcting the small rotation angle of CCD camera, and eliminating the system error caused by it.
【作者单位】： 中国科学院上海应用物理研究所;中国科学院大学;
【基金】：国家自然科学基金项目(11375255;11075198)
【分类号】：TP391.41
本文编号：2364991
[Abstract]:In the process of interferometer image data processing, the current multi-line average image processing algorithm will introduce a large random error, and when there is a small angle between the CCD camera and the slit, a large system error will be introduced. This paper mainly explores the feasibility and advantages of using principal component analysis (PCA) algorithm to process spatial interferometer images. The spatial interferometer image is simulated by MATLAB, random noise and image rotation are added, and the data are processed by PCA method and multi-line average algorithm, and the error between the two algorithms is compared. The small angle experiment of CCD camera and the weak signal experiment of interference image are designed to evaluate the effect of PCA algorithm in correcting the small angle of CCD camera and reconstructing weak signal image in the process of data processing. The theoretical and experimental results show that the PCA algorithm is more effective than the current multiline average algorithm in reducing noise, analyzing weak signal images and correcting the small rotation angle of CCD camera, and eliminating the system error caused by it.
【作者单位】： 中国科学院上海应用物理研究所;中国科学院大学;
【基金】：国家自然科学基金项目(11375255;11075198)
【分类号】：TP391.41
【相似文献】
相关硕士学位论文 前1条
1 王杰;光学显微层析实验系统设计与研制[D];南京理工大学;2009年
，本文编号：2364991
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2364991.html
