傅里叶望远镜成像质量分析与评价  

Image Quality Evaluation and Analysis in Fourier Telescopy for Laboratory Demonstration

作　　者：张羽[1,2] 罗秀娟[1] 夏爱利[1] 曹蓓[1] 程志远[1,2] 曾志红[1,2] 司庆丹[1,2] 王保峰[1,2] 

机构地区：[1]中国科学院西安光学精密机械研究所,西安710119 [2]中国科学院大学,北京100049

出　　处：《光子学报》2014年第3期160-165,共6页Acta Photonica Sinica

基　　金：总装备部重大研究项目(No.7131130)资助

摘　　要：斯特列尓比通常被作为评价傅里叶望远镜成像质量的方法.该方法通过求得复原图像与理想图像的相关度来评价成像质量,主要用于计算机仿真分析.然而,在真实环境中往往因为不能预知被测目标的形貌特征,使斯特列尓比像质评价方法失去作用.本文针对实验室内傅里叶望远镜成像数据,采用了两种不需要参考图象对比的方法:灰度平均梯度法和拉普拉斯梯度模法描述复原图像包含细节信息的程度,分析了非参照图像评价方法的可行性.对同一复杂程度的目标,分别利用计算机仿真及室内实测数据成像,分析灰度平均梯度法、拉普拉斯梯度模法与斯特列尓比描述成像质量的一致性.结果表明,分析灰度平均梯度法、拉普拉斯梯度模法这样的评价与斯特列尓比评价趋势一致,证明了分析灰度平均梯度法、拉普拉斯梯度模法无参照评价方法用于分析与评价傅里叶望远镜图像质量的可行性.Strehl ratio is often used as a method for evaluation of the Fourier telescope imaging quality. This method computes the correlation of the reconstructed image and the ideal image to evaluate the image quality, which is mainly applicable in the computer simulation analysis. However, the object to be observed may be unknow in the real world, which means there is no ideal image for use. Based on laboratory imaging data, the feasibility of the non-reference image evaluation method applied to Fourier telescopy is proposed and analyzed. The gray mean grads （GMG） method and Laplacian Sum（LS） are adopted to describe how much detail information the reconstructed image has. For the same object, Employ Strehl ratio and GMG, LS to evaluate the simulated image and the laboratory imaging result, respectively. Statistical results show that the two kinds of evaluations are identical in changing tendency. It proves the GMG and LS methods are applicable in Fourier telescopy as a non-reference image evaluation.

关 键 词：傅里叶望远术 像质评价 灰度平均梯度 拉普拉斯梯度模 斯特列尓比 激光拍频 

分 类 号：O438.2[机械工程—光学工程]