首页 > 论文 > 光学学报 > 37卷 > 11期(pp:1129001--1)

基于深度神经网络的空间目标常用材质BRDF模型

BRDF Model for Commonly Used Materials of Space Targets Based on Deep Neural Network

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

由于双向反射分布函数( BRDF)经验模型与半经验模型对材质散射特性描述时存在局限性,导致其拟合结果与实测数据的误差较大。针对此问题,基于深度神经网络(DNN)构建了一种适用于具有不同散射特性空间目标材质的BRDF模型。建立的深度神经网络模型基于TensorFlow实现,包含4个隐含层,并采用AdaDelta梯度下降法进行优化,结合Dropout方法进行正则。随机抽取材质测量数据的一部分作为训练样本,最终得到BRDF与入射天顶角、反射天顶角以及观测方位角的映射关系模型。大量的实验结果表明,建立的深度神经网络模型具有良好的材质特性描述能力,且对于相同材质,模型的拟合误差小于经验模型。

Abstract

When the bidirectional reflectance distribution function (BRDF) empirical model and semi-empirical model describe the scattering characteristics of the material, the limitation of these models for the description of different scattering characteristics results in large errors between the fitting result and the measured data. To solve the problem, a BRDF model suitable for commonly used materials on space targets with different characteristics is constructed based on deep neural network (DNN). The DNN model, which contains four hidden layers, is based on TensorFlow implementation. It is optimized by AdaDelta gradient descent method, and combined with Dropout method for regularity. Part of the material measurement data is randomly selected as the training sample, and finally the mapping relationships between the BRDF and the angles of the incident zenith, the reflection zenith and the observation azimuth are obtained. A large number of experimental results show that the DNN model has good ability to describe the scattering characteristics of materials, and the fitting error of the DNN model is less than that of the empirical model for the same material.

投稿润色
补充资料

中图分类号:O436.2

DOI:10.3788/aos201737.1129001

所属栏目:散射

基金项目:国家863计划(2015AA7046104)

收稿日期:2017-07-03

修改稿日期:2017-07-20

网络出版日期:--

作者单位    点击查看

刘程浩:装备学院研究生院, 北京 101416
李 智:装备学院航天指挥系, 北京 101416
徐 灿:装备学院航天装备系, 北京 101416
田琪琛:装备学院研究生院, 北京 101416

联系人作者:李智(lizhizys@263.com)

备注:刘程浩(1993-),男,硕士研究生,主要从事空间目标光学散射特性测量与建模方面的研究。

【1】Yuan Yan, Sun Chengming, Zhang Xiubao. Measuring and modeling the spectral bidirectional reflection distribution function of space target’s surface material[J]. Acta Physica Sinica, 2010, 59(3): 2097-2103.
袁艳, 孙成明, 张修宝. 空间目标表面材料光谱双向反射分布函数测量与建模[J]. 物理学报, 2010, 59(3): 2097-2103.

【2】Xu Can, Zhang Yasheng, Li Peng, et al. Calculation of optical cross section areas of spatial objects based on OpenGL picking technique[J]. Acta Optica Sinica, 2017, 37(7): 0720001.
徐灿, 张雅声, 李鹏, 等. 基于OpenGL拾取技术的空间目标OCS计算方法[J]. 光学学报, 2017, 37(7): 0720001.

【3】Gou Ruixin, Du Xiaoping, Liu Hao. Advances in attitude inversion of space object based on photometric data[J]. Laser & Optoelectronics Progress, 2016, 53(10): 100002.
苟瑞新, 杜小平, 刘浩. 光度数据反演空间目标姿态的研究进展[J]. 激光与光电子学进展, 2016, 53(10): 100002.

【4】Han Yi, Sun Huayan. Advances in space target optical scattering character research[J]. Infrared and Laser Engineering, 2013, 42(3): 758-766.
韩意, 孙华燕. 空间目标光学散射特性研究进展[J]. 红外与激光工程, 2013, 42(3): 758-766.

【5】Torrance K E, Sparrow E M. Theory for off-specular reflection from roughened surfaces[J]. Journal of the Optical Society of America, 1967, 57(9): 1105-1114.

【6】Maxwell J R, Beard J, Weiner S, et al. Bidirectional reflectance model validation and utilization[R]. Ann Arbor: Environmental Research Institute of Michigan, 1973: 10-35.

【7】Phong B T. Illumination for computer generated pictures[J]. Communications of the ACM, 1975, 18(6): 311-317.

【8】Wu Zhensen, Xie Donghui, Xie Pinhua, et al. Modeling reflectance function from rough surface and algorithms[J]. Acta Optica Sinica, 2002, 22(8): 897-901.
吴振森, 谢东辉, 谢品华, 等. 粗糙表面激光散射统计建模的遗传算法[J]. 光学学报, 2002, 22(8): 897-901.

【9】Minnaert M. The reciprocity principle in lunar photometry[J]. Astrophysical Journal, 1941, 93(3): 403-410.

【10】吴岸城. 神经网络与深度学习[M]. 北京: 电子工业出版社, 2016: 81-87.

【11】李玉鑑, 张婷. 深度学习导论及案例分析[M]. 北京: 机械工业出版社, 2016: 4-8.

引用该论文

Liu Chenghao,Li Zhi,Xu Can,Tian Qichen. BRDF Model for Commonly Used Materials of Space Targets Based on Deep Neural Network[J]. Acta Optica Sinica, 2017, 37(11): 1129001

刘程浩,李 智,徐 灿,田琪琛. 基于深度神经网络的空间目标常用材质BRDF模型[J]. 光学学报, 2017, 37(11): 1129001

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF