高分一号WFV相机的大气校正与图像质量评估  

Atmospheric Correction and Image Quality Assessment of WFV Camera in GaoFen-1 Satellite

作　　者：李妍敏 舒双宝[1] 李志文 陈志强 陈晶晶[1] Shu Shuangbao;Li Zhiwen;Chen Zhiqiang;Chen Jingjing(School of Instrument Science aud Opto-Electronics Engineering,Hefei,University of Technology,Hefei,Anhui 230009,China)

机构地区：[1]合肥工业大学仪器科学与光电工程学院,安徽合肥230009

出　　处：《光学学报》2020年第20期27-35,共9页Acta Optica Sinica

摘　　要：基于6S大气辐射传输模型和中分辨率成像光谱仪(MODIS)上午星Terra的气溶胶光学厚度数据以及MODIS双向反射分布函数(BRDF)模型参数产品(MCD43A1),对高分一号(GF-1)卫星宽视场相机(WFV)四个波段的大气层顶辐亮度图像进行大气校正,得到校正后的地表反射率图像。而后基于Brenner梯度算子和中频离散余弦变换两种方法统计校正前后图像的清晰度值,分析计算结果可得大气校正后图像清晰度值高于校正前图像的清晰度值,因此校正后图像的边缘纹理比校正前更清晰;基于阈值分割法原理进行信噪比评价,结果表明校正前后每一波段的信噪比随辐亮度呈递增关系,大气对短波波段的影响较大,而对长波波段的影响较小。For radiance images on the top of atmosphere obtained by wide field of view(WFV) camera in GaoFen-1(GF-1) satellite, atmospheric corrections at four bands are made to obtain surface reflectance images based on 6 S atmospheric radiation transmission model, aerosol optical depth from Terra in Moderate-Resolution Imaging Spectroradiometer(MODIS) and the parameter products(MCD43 A1) of MODIS bi-directional reflection distribution function model. Then the calculation methods of the image definition values are based on Brenner gradient operator and medium frequency discrete cosine transform. The calculation results show that the image definition values after correction are higher than those before correction, so the edge texture information of the image after correction is clearer than that before correction. The signal-to-noise ratio(SNR) is evaluated based on the principle of threshold segmentation. The results show that the SNR of each band before and after correction has an increasing relationship with the radiance. The atmosphere has more influence on the short-wave band than the long-wave band.

关 键 词：大气光学 大气辐射传输模型 大气校正 图像清晰度 信噪比 

分 类 号：TP751.1[自动化与计算机技术—检测技术与自动化装置]