基于支持向量机预测偏振模式

汪方斌; 洪津; 孙晓兵; 王羿; 胡亚东

doi:10.3788/OPE.20142211.2914

您当前的位置：

首页 >

文章列表页 >

基于支持向量机预测偏振模式

现代应用光学 | 更新时间：2020-08-13

- 基于支持向量机预测偏振模式
- Prediction of polarization pattern by SVM
- 光学精密工程 2014年22卷第11期页码：2914-2922
- 作者机构：
  
  1. 安徽建筑大学机械与电气工程学院,安徽合肥,230601
  2. 中国科学院安徽光学精密机械研究所光学遥感中心,安徽合肥,230031
- 作者简介：
- 基金信息：
  
  国家自然科学基金资助项目(No.40971196)()
- DOI：10.3788/OPE.20142211.2914
  中图分类号： P407
- 收稿日期：2013-10-28，
  
  修回日期：2013-11-30，
  
  纸质出版日期：2014-11-25
- 稿件说明：
移动端阅览
汪方斌, 洪津, 孙晓兵等. 基于支持向量机预测偏振模式[J]. 光学精密工程, 2014,22(11): 2914-2922

WANG Fang-bin, HONG Jin, SUN Xiao-bing etc. Prediction of polarization pattern by SVM[J]. Editorial Office of Optics and Precision Engineering, 2014,22(11): 2914-2922
汪方斌, 洪津, 孙晓兵等. 基于支持向量机预测偏振模式[J]. 光学精密工程, 2014,22(11): 2914-2922 DOI： 10.3788/OPE.20142211.2914.

WANG Fang-bin, HONG Jin, SUN Xiao-bing etc. Prediction of polarization pattern by SVM[J]. Editorial Office of Optics and Precision Engineering, 2014,22(11): 2914-2922 DOI： 10.3788/OPE.20142211.2914.

摘要

基于多光谱多角度偏振辐射探测研究了大气偏振模式预测方法。首先

介绍了航空多角度偏振辐射计的探测原理与支持向量机回归算法;然后

从矢量传输模型出发

说明大气状况不变时

偏振模式主要取决于地表特征与观测几何

并介绍了观测几何与姿态之间的关系以及地表特征的表达形式;最后

在考虑平台姿态与地表特征的情况下

利用支持向量机回归算法预测了航空多角度偏振辐射计的探测偏振度

并对预测与实际试验探测的偏振度进行了比较。结果显示:偏振度预测误差小于1%

影响模型精度的主要因素不是姿态变化本身

而是姿态改变造成的观测地表特性变化。

Abstract

On the basis of multi-spectral multi-angular radiative polarization measurements

a prediction method for atmospheric polarization patterns is investigated. Firstly

the principle of an Aviation Multi-angle Polarimetric Radiometer (AMPR) and the regression algorithm of a Support Vector Machine (SVM) are introduced. Then

according to Vector Radiative Transfer Equation (VRTE)

it points out that the atmospheric polarization pattern is primarily dependent on view geometry and surface features when atmosphere condition is invariant. Meanwhile

it introduces the relationship between view geometry and platform attitude and the expressing form of surface features. Finally

how to use the regression algorithm of SVM to predict the detected polarization degree of the APMR and to validate the application process is introduced in the consideration of the surface features and platform attitudes. Furthermore

the predicted degrees of polarization is also compared to that of real measurements. The results indicate that the error of polarization degree predicted is less than 1%. It concludes that the serious factor to affect the model accuracy is not the change of attitude but the variation of underlying surface property caused by the changed attitudes.

关键词

Keywords

references

LITVINOV P, HASEKAMP O, CAIRNS B,et al. Reflection models for soil and vegetation surfaces from multiple-viewing angle photopolarimetric measurements[J]. Journal of Quantitative Spectroscopy & Radiative Transfer,2010, 111: 529-539.

赵开春, 卢皓, 尤政. 天空光偏振模式自动探测装置[J]. 光学精密工程,2013,21(2): 239-245. ZHAO K CH, LU H, YOU ZH. Automatic detection system for skylight polarized pattern[J]. Opt. Precision Eng., 2013, 21(2): 239-245. (in Chinese)

CAIRNS B, TRAVIS L, MISHCHENKO M I, et al.. Aerosol retrievals over land surfaces (the advantages of polarization)[C]. Proc. AMS Annu. Meeting, 2001:119-122.

WAQUET F, LÉON J F, CAIRNS B, et al.. Analysis of the spectral and angular response of the vegetated surface polarization for the purpose of aerosol remote sensing over land[J]. Applied Optics, 2009,48(6):1228-1236.

褚金奎, 王威, 崔岩. 气溶胶对天空偏振辐射影响的测量[J]. 光学精密工程,2012,20(3):520-526. CHU J K, WANG W, CUI Y. Measurement for influence of aerosols on polarized sky radiance[J]. Opt. Precision Eng., 2012,20(3):520-526. (in Chinese)

LITVINOV P, HASEKAMP O, CAIRNS B, et al.. Semi-empirical BRDF and BPDF models applied to the problem of aerosol retrievals over land: testing on airborne data and implications for modeling of top-of-atmosphere measurements[C]. NATO Science for Peace and Security Series C: Environmental Security, 2011: 313-340.

KNOBELSPIESSE K D, CAIRNS B, SCHMID B, et al.. Surface BRDF estimation from an aircraft compared to MODIS and ground estimates at the Southern Great Plains site[J]. Journal of Geophysical Research,2008:113, D20105.

卢鸿谦,尹航,黄显林. 偏振光/地磁/GPS/SINS组合导航方法[J]. 宇航学报,2007,28(4):897-902. LU H Q, YING H, HUANG X L. Combination navigation method with polarized light/geomagnetic/GPS/SINS[J]. Journal of Astronautics, 2007, 28(4): 897-902. (in Chinese)

赵开春,褚金奎,姚弘轶,等. Rayleigh大气天空光偏振分布仿真与预测[J]. 四川大学学报:工程科学版,2007,39(s1): 288-291. ZHAO K CH, CHU J K, YAO H Y, et al.. Simulation and prediction for Rayleih skylight polarization distribution[J]. Journal of Sichuan University:Engineering Science Edition,2007,39(s1): 288-291. (in Chinese)

李淑军,姜会林,朱京平,等. 偏振成像探测技术发展现状及关键技术[J]. 中国光学,2013,6(6):803-809. LI SH J, JIANG H L, ZHU J P, et al.. Development status and key technologies of polarization imaging detection[J]. Chinese Optics, 2013, 6(6): 803-809. (in Chinese)

周军,刘莹莹. 基于自然偏振光的自主导航新方法研究进展[J]. 宇航学报, 2009, 30(2): 409-414. ZHOU J, LIU Y Y. Research progress of new method on autonomous navigation based on natural polarimetric light[J]. Journal of Astronautics, 2009, 30(2):409-414. (in Chinese)

ABLE K P. Skylight polarization patterns and the orientation of migratory birds[J]. The Journal of Experimental Biology,1989, 141:241-256.

王玲妹, 高隽, 谢昭. 光的地表反射偏振特性分析及空间偏振模式计算方法[J]. 中国科学:物理学·力学·天文学, 2013, 43(7): 833-843. WANG L M, GAO J, XIE ZH. Analysis on polarized lights by ground-reflection with calculating its spatial patterns[J]. Sci. Sin-Phys. Mech. Astron., 2013, 43(7): 833-843. (in Chinese)

宋茂新,孙斌,孙晓兵,等. 航空多角度偏振辐射计的偏振定标[J]. 光学精密工程,2012,20(6): 1153-1158. SONG M X, SUN B, SUN X B, et al.. Polarization calibration of airborne muti-angle polarimetric radiometer[J]. Opt. Precision Eng., 2012,20(6): 1153-1158. (in Chinese)

MISHCHENKO M I, ROSENBUSH V K, KISELEV N N, et al.. Polarimetric Remote Sensing of Solar System Objects[M]. Kyiv:Akademperiodyka, 2010.

邓乃扬,田英杰. 数据挖掘中的新方法-支持向量机[M]. 北京:科学出版社,2004. DENG N Y, TIAN Y J. New Method on Data Mining-Support Vector Machine[M]. Beijing:Science Press, 2004. (in Chinese)

MISHCHENKO M I. Microphysical approach to polarized radiative transfer: extension to the case of an external observation point[J]. Appl. Optics, 2003,42(24):4963-4967.

许永森,丁亚林,田海英,等. 航空遥感器姿态像移仿真分析[J]. 激光与红外, 2007, 37(12): 1303-1306. XU Y S, DING Y L, TIAN H Y, et al.. The simulation of aerial remote sensor's gesture image motion[J]. Laser & Infrared,2007, 37(12):1303-1306. (in Chinese)

WQQUET F, CAIRNS B, KNOBELSPIESSE K, et al.. Polarimetric remote sensing of aerosols over land[J]. Journal of Geophysical Research, 2009:114, D01206.

SPURR R J D. A new approach to the retrieval of surface properties from earthshine measurements[J]. Journal of Quantitative Spectroscopy & Radiative Transfer, 2004, 83(1): 15-46.

KAUFMAN Y J, TANRE D. Atmospherically resistant vegetation index (ARVI) for EOS-MODIS[C]. Proc. IEEE Int. Geosci. and Remote Sensing Symp.,New York,1992:261-270.

WANG F B, LI D H. A method of parameters selection with higher accuracy for SVM[C]. ICACI2012, 2012:248-252.

浏览量

270

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于稀疏编码空间金字塔匹配和GA-SVM的列车故障自动识别

基于高光谱的微藻生物膜生长特性研究

高速切削钛合金TC4的表面粗糙度预测与控制方法

统计回归大气透过率廓线的实时计算

基于拉曼光谱的三组分食用调和油快速定量检测