基于多核支持向量回归的光谱反射率重建方法

赵丽娟; 王慧琴; 王可; 王展; 刘加林; 杨蕾

doi:10.3788/YJYXS20183312.1008

您当前的位置：

首页 >

文章列表页 >

基于多核支持向量回归的光谱反射率重建方法

图像处理 | 更新时间：2020-08-12

- 基于多核支持向量回归的光谱反射率重建方法
- Spectral reflectance reconstruction based on multi-kernel support vector regression
- 液晶与显示 2018年33卷第12期页码：1008-1018
- 作者机构：
  
  1.西安建筑科技大学信息与控制工程学院, 陕西西安 710055
  2.陕西文物保护研究院, 陕西西安 710075
- 作者简介：
  
  [ "赵丽娟(1994-), 女, 吉林长春人, 硕士研究生, 主要从事光谱图像处理及古代壁画颜色数字化还原、复制方面的研究。E-mail:981629018@qq.com" ]
  王慧琴, E-mail：hqwang@xauat.edu.cn
- 基金信息：
  
  国家自然科学基金青年基金(61701388);教育部归国留学人员科研扶持项目(K05055);住建部科学技术计划资助项目(2017-K2-014);陕西省科技厅国际合作资助项目(2017KW-036);陕西省自然科学基础研究计划(2018JM6080);西安建筑科技大学青年基金(QN1628);西安建筑科技大学青年基金(JC1514);the Youth Fund of National Natural Science Foundation of China(61701388);Research Support Project for Returned Students of Ministry of Education(K05055);Science and Technology Project of Ministry of Housing and Urban-Rural Development of China(2017-K2-014);International Science and Technology Cooperation Project of Science and Technology Department of Shaanxi Province, China(2017KW-036);Natural Science Basic Research Program of Shaanxi Province(2018JM6080);Youth Science and Technology Fund Project of Xi'an University of Architecture And Technology, China(QN1628);Youth Science and Technology Fund Project of Xi'an University of Architecture And Technology(JC1514)
- DOI：10.3788/YJYXS20183312.1008
  中图分类号： O433.4
- 收稿日期：2018-07-05，
  
  录用日期：2018-9-11，
  
  纸质出版日期：2018-12
- 稿件说明：
移动端阅览
赵丽娟, 王慧琴, 王可, 等. 基于多核支持向量回归的光谱反射率重建方法[J]. 液晶与显示, 2018,33(12):1008-1018.

Li-juan ZHAO, Hui-qin WANG, Ke WANG, et al. Spectral reflectance reconstruction based on multi-kernel support vector regression[J]. Chinese journal of liquid crystals and displays, 2018, 33(12): 1008-1018.
赵丽娟, 王慧琴, 王可, 等. 基于多核支持向量回归的光谱反射率重建方法[J]. 液晶与显示, 2018,33(12):1008-1018. DOI： 10.3788/YJYXS20183312.1008.

Li-juan ZHAO, Hui-qin WANG, Ke WANG, et al. Spectral reflectance reconstruction based on multi-kernel support vector regression[J]. Chinese journal of liquid crystals and displays, 2018, 33(12): 1008-1018. DOI： 10.3788/YJYXS20183312.1008.

摘要

针对传统方法重建光谱反射率过程中未考虑多光谱训练数据维度高、冗余大的特点，导致重建精度低、重建模型学习能力和泛化能力较差的问题，提出了多核支持向量回归的光谱反射率重建方法。首先综合全局、局部核函数的特点引入柯西核函数与多项式核函数的乘积作为多核核函数，然后使用试凑法从训练样本中获取提高模型性能的参数。最后使用多核支持向量回归模型对测试样本进行反射率重建，通过光谱误差及适应度等进行评价。实验结果表明：与伪逆、单核支持向量回归法相比，本文重建方法的光谱误差值降低了0.4~0.785，决策系数提高了2.84~5.27%，平均适应度系数值提高了2%~3%。本文方法在颜色复制中重建精度高、色差较小，满足人眼视觉可容忍范围内。

Abstract

In the process of spectral reflectance reconstruction by traditional methods

it doesn't take into account of the characteristics of highly dimensional and redundant multi-spectral training data

which results in low reconstruction accuracy

poor learning ability and generalization ability of the reconstruction model. In order to solve such problems

the spectral reflectance reconstruction of multi-kernel support vector regression was proposed. Firstly

the product of Cauchy kernel function and polynomial kernel function was introduced as multi-kernel function combining with the characteristics of overall situation and part of kernel function

and then the parameters to improve the performance of the model were obtained from the training samples by cut-and-trial. Finally

the multi-kernel support vector regression model was used to reconstruct the reflectance of the test samples

and evaluation was made through spectral error and fitness. The experimental results show that compared with the pseudo-inverse and single-kernel support vector regression

the spectral error of the reconstruction method is reduced by 0.4-0.785

the decision-making coefficient is increased by 2.84-5.27%

and the average fitness coefficient is increased by 2%-3%. In color reproduction

the reconstruction accuracy is high and the chromatic aberration is small

so that it can meet the tolerance range of human vision.

关键词

Keywords

references

刘怡轩, 颜昌翔.在线式分光测色仪发展现状[J].中国光学, 2014, 7(5):712-722.

LIU Y X, YAN C X. Development and status of on-line spectrophotometer[J]. Chinese Optics, 2014, 7(5):712-722. (in Chinese)

刘燕德, 邓清.基于高光谱成像技术的脐橙叶片的叶绿素含量及其分布测量[J].发光学报, 2015, 36(8):957-961.

LIU Y D, DENG Q. Measurement of chlorophyll distribution in navel orange leaves based on hyper-spectral imaging technique[J]. Chinese Journal of Luminescence, 2015, 36(8):957-961. (in Chinese)

闫丽霞, 吴凡.线性与非线性成像系统下的多光谱重建研究[J].计算机应用与软件, 2014, 31(3):207-210.

YAN L X, WU F. Study on multispectral reconstruction based on linear and nonlinear imaging system[J]. Computer Applications and Software, 2014, 31(3):207-210. (in Chinese)

张伟峰.基于支持向量回归的光谱反射率重建方法[J].计算机科学, 2010, 37(12):241-242, 269.

ZHANG W F. Spectral reflectance estimation by support vector regression[J]. Computer Science, 2010, 37(12):241-242, 269. (in Chinese)

SONG H Z, DING Z C, GUO C C, et al. Research on combination kernel function of support vector machine[C]//Proceedings of 2008 International Conference on Computer Science and Software Engineering. Hubei, China: IEEE, 2008: 838-841. https://www.researchgate.net/publication/221196129_Research_on_Combination_Kernel_Function_of_Support_Vector_Machine

李婵, 万晓霞, 梁金星.基于光谱连接空间的彩色相机光谱重构研究[J].发光学报, 2016, 37(12):1571-1578.

LI C, WAN X X, LIANG J X, Spectral reconstruction from trichromatic digital camera responses based on interim connection space for spectral data[J]. Chinese Journal of Luminescence, 2016, 37(12):1571-1578. (in Chinese)

邹纪平, 杨葭, 李宏宁, 等.基于R矩阵法光谱反射率重建研究[J].光学学报, 2014, 34(S2):S233001.

ZOU J P, YANG J, LI H N, et al. Study of spectral reflectance reconstruction based on matrix R method[J]. Acta Optica Sinica, 2014, 34(S2):S233001. (in Chinese)

冯愈钦, 吴龙国, 何建国, 等.基于高光谱成像技术的长枣不同保藏温度的可溶性固形物含量检测方法[J].发光学报, 2016, 37(8):1014-1022.

FENG Y Q, WU L G, HE J G, et al. Detection method of soluble solid of jujube at different preservative temperature based on hyper-spectral imaging technology[J]. Chinese Journal of Luminescence, 2016, 37(8):1014-1022. (in Chinese)

王可, 王慧琴, 龙艳群, 等.基于降维正则化多项式的光谱反射率重建方法[J].激光与光电子学进展, 2018, 55(5):053004.

WANG K, WANG H Q, LONG Y Q, et al. Spectral reflectance reconstruction based on dimension reduction regularization polynomials[J]. Laser & Optoelectronics Progress, 2018, 55(5):053004. (in Chinese)

AWAD M, KHANNA R. Support vector regression[J]. Neural Information Processing Letters & Reviews, 2007, 11(10):203-224.

宋悦, 林志贤, 姚剑敏, 等.基于ε-支持向量回归机的背光源亮度均匀性评价方法[J].液晶与显示, 2015, 30(5):857-863.

SONG Y, LIN Z X, YAO J M, et al. Luminance uniformity evaluation for backlight based on ε-support vector regression[J]. Chinese Journal of Liquid Crystals and Displays, 2015, 30(5):857-863. (in Chinese)

袁其平, 林海杰, 陈志宏, 等.用支持向量回归法实现单帧图像超分辨率重建[J].光学精密工程, 2016, 24(9):2302-2309.

YUAN Q P, LIN H J, CHEN Z H, et al. Single image super-resolution reconstruction using support vector regression[J]. Optics and Precision Engineering, 2016, 24(9):2302-2309. (in Chinese)

杨海燕, 周永权.一种支持向量机的混合核函数[J].计算机应用, 2009, 29(S2):173-175, 178.

YANG H Y, ZHOU Y Q. Mixture kernel function of support vector machines[J]. Journal of Computer Applications, 2009, 29(S2):173-175, 178. (in Chinese)

王国胜.核函数的性质及其构造方法[J].计算机科学, 2006, 33(6):172-174, 178.

WANG G S. Properties and construction methods of kernel in support vector machine[J]. Computer Science, 2006, 33(6):172-174, 178. (in Chinese)

王睿.关于支持向量机参数选择方法分析[J].重庆师范大学学报(自然科学版), 2007, 24(2):36-38, 42.

WANG R. Method analye about support vector machine's parameter[J]. Journal of Chongqing Normal University (Natural Science Edition), 2007, 24(2):36-38, 42. (in Chinese)

龙艳群, 王慧琴, 王可, 等.基于样本优化选取的光谱重建方法研究[J].影像科学与光化学, 2017, 35(1):88-96.

LONG Y Q, WANG H Q, WANG K, et al. Study on spectral reconstruction method based on optimized selected samples[J]. Imaging Science and Photochemistry, 2017, 35(1):88-96. (in Chinese)

WANG K, WANG H Q, WANG Z, et al. Study of spectral reflectance reconstruction based on regularization matrix R method[J]. Cluster Computing, 2017, doi:10.1007/s10586-017-1217-5.

王可, 王慧琴, 王展, 等.基于改进R矩阵方法的光谱反射率重建研究[J].液晶与显示, 2018, 33(4):338-346.

WANG K, WANG H Q, WANG Z, et al. Spectral reflectance reconstruction based on improved matrix R method[J]. Chinese Journal of Liquid Crystals and Displays, 2018, 33(4):338-346. (in Chinese)

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于改进R矩阵方法的光谱反射率重建研究