首页 > 论文 > 激光与光电子学进展 > 56卷 > 2期(pp:21101--1)

基于多模型融合的高光谱图像质量评价

Hyperspectral Image Quality Evaluation Based on Multi-Model Fusion

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

摘要

针对单模型评价图像质量容易产生过拟合的问题, 提出基于多模型融合的高光谱图像质量评价算法。以图像噪声、模糊度和云含量为降质特征, 建立遥感图像主观评价库, 分别选用支持向量回归方法和集成决策树方法对带有评价值的训练集图像建立质量评价单模型。将两个单模型评价结果线性回归拟合, 得到模型融合的图像质量评价结果。同时, 以广义回归神经网络模型作为参照, 分别从均方误差、回归拟合指标、分类准确率、训练时间4个方面对几种模型进行对比。实验结果表明, 所提模型融合算法具有较高的拟合精度、较强的泛化能力, 并且所需的训练时间相对较少。

Abstract

In order to solve the problem that image quality is easily overfitted by a single model, a hyperspectral image quality evaluation algorithm is proposed based on multi-model fusion. Taking image noise, ambiguity and cloud content as the degraded features, a remote sensing image subjective evaluation database is established. The support vector regression method and the integrated decision tree method are respectively selected to establish a quality evaluation model for training set images with evaluation values. The image quality evaluation results based on model fusion are obtained via linear regression fitting of the two single model evaluation results. At the same time, the generalized regression neural network model is introduced as a reference, and several models are compared from four aspects of mean square error, regression fitting index, classification accuracy and training time. The experimental results show that the proposed model fusion algorithm has relatively high fitting accuracy, relatively strong generalization ability and relatively little training time.

Newport宣传-MKS新实验室计划
补充资料

中图分类号:TP751.1

DOI:10.3788/lop56.021101

所属栏目:成像系统

基金项目:国家自然科学基金(61671408)、教育部联合基金(6141A02022314)、上海航天科技创新基金(SAST2015041)

收稿日期:2018-07-01

修改稿日期:2018-07-06

网络出版日期:2018-07-26

作者单位    点击查看

徐冬宇:浙江大学电气工程学院, 浙江 杭州 310027
厉小润:浙江大学电气工程学院, 浙江 杭州 310027
赵辽英:杭州电子科技大学计算机应用技术研究所, 浙江 杭州 310018
舒锐:上海卫星工程研究所, 上海 200240
唐琪佳:上海卫星工程研究所, 上海 200240

联系人作者:厉小润(lxr@zju.edu.cn)

【1】Xu D. Hyperspectral remote sensing image denoising based on multiple linear regression and sparse representation [D]. Changsha: National University of Defense Technology, 2013.
徐冬. 基于多元线性回归和稀疏化表示的高光谱遥感图像去噪[D]. 长沙: 国防科学技术大学, 2013.

【2】Tan Z G, Ou J P, Zhang J, et al. A laminar denoising algorithm for depth image[J]. Acta Optica Sinica, 2017, 37(5): 0510002.
谭志国, 欧建平, 张军, 等. 一种层析深度图像去噪算法[J]. 光学学报, 2017, 37(5): 0510002.

【3】Zhang X, Jing X L. A method for image denoising based on normal inverse Gaussian model using Bayesian estimation[J]. Acta Optica Sinica, 2010, 30(1): 70-74.
张鑫, 井西利. 一种基于正态反高斯模型的贝叶斯图像去噪方法[J]. 光学学报, 2010, 30(1): 70-74.

【4】Pang S L. Research on image ambiguity evaluation [D]. Xi′an: Xidian University, 2010.
庞胜利. 图像模糊度评价研究[D]. 西安: 西安电子科技大学, 2010.

【5】Lu H M, Xu M, Li X. Image deblurring with adaptive signal-noise ratio estimation for computational imaging system[J]. Acta Optica Sinica, 2014, 34(8): 0810002.
卢惠民, 徐明, 李迅. 用于计算成像系统的基于信噪比自适应估计的图像去模糊研究[J]. 光学学报, 2014, 34(8): 0810002.

【6】Feng S Y, Zhang N, Shen J, et al. Method of cloud detection with hyperspectral remote sensing image based on the reflective characteristics [J]. China Optics, 2015, 8(2): 198-204.
冯书谊, 张宁, 沈霁, 等. 基于反射率特性的高光谱遥感图像云检测方法研究[J]. 中国光学, 2015, 8(2): 198-204.

【7】Chen Y, Fan R S, Wang J X, et al. Cloud detection of ZY-3 satellite remote sensing images based on deep learning[J]. Acta Optica Sinica, 2018, 38(1): 0128005.
陈洋, 范荣双, 王竞雪, 等. 基于深度学习的资源三号卫星遥感影像云检测方法[J]. 光学学报, 2018, 38(1): 0128005.

【8】Jiang X H, Wang X H, Ye H H, et al. Cloud contaminated satellite data processing method in CO2 retrieving [J]. Acta Optica Sinica, 2015, 35(8): 0801001.
江新华, 王先华, 叶函函, 等. CO2反演中卫星数据的云污染处理方法研究[J]. 光学学报, 2015, 35(8): 0801001.

【9】Gao L L, Liu J J, Ren X, et al. Image quality evaluation of panoramic camera steropair based on structural similarity [J]. Laser & Optoelectronics Progress, 2014, 51(7): 071004.
高露露, 刘建军, 任鑫, 等. 基于结构相似度的全景相机立体像对图像质量评价[J]. 激光与光电子学进展, 2014, 51(7): 071004.

【10】Hou C P, Ma T T, Yue G H, et al. Multiply-distorted image quality assessment based on high-order phase congruency[J]. Laser & Optoelectronics Progress, 2017, 54(7): 071001.
侯春萍, 马彤彤, 岳广辉, 等. 基于高阶相位一致性的混合失真图像质量评价[J]. 激光与光电子学进展, 2017, 54(7): 071001.

【11】Zhang F, Zhang N Y, Li Z Z. Image quality evaluation method based on symmetric phase consistency [J]. Laser & Optoelectronics Progress, 2017, 54(10): 101003.
张帆, 张偌雅, 李珍珍. 基于对称相位一致性的图像质量评价方法[J]. 激光与光电子学进展, 2017, 54(10): 101003.

【12】Wang Z M. Review of no-reference image quality assessment[J]. Acta Automatica Sinica, 2015, 41(6): 1062-1079.
王志明. 无参考图像质量评价综述[J]. 自动化学报, 2015, 41(6): 1062-1079.

【13】Yin Y. Non-reference blur image quality assessment based on general regression neural network[J]. Laser & Infrared, 2013, 43(4): 466-470.
殷莹. 基于广义回归神经网络无参考模糊图像质量评价[J].激光与红外, 2013, 43(4): 466-470.

【14】Zhu X L, Xu M Z, Cong M. Assessment method for no-reference remote sensing image quality based on support vector machine[J]. Spacecraft Recovery & Remote Sensing, 2014, 35(6): 83-90.
朱晓玲, 许妙忠, 丛铭.基于支持向量机的无参考遥感图像质量评价方法[J]. 航天返回与遥感, 2014, 35(6): 83-90.

【15】Wu Y Q, Zhou Y, Long Y L. Small target detection in hyperspectral remote sensing image based on adaptive parameter SVM[J]. Acta Optica Sinica, 2015, 35(9): 0928001.
吴一全, 周杨, 龙云淋. 基于自适应参数支持向量机的高光谱遥感图像小目标检测[J].光学学报, 2015, 35(9): 0928001.

【16】Fan C L. Hyperspectral remote sensing image classification algorithm based on decision tree[D]. Qinhuangdao: Yanshan University, 2014.
范成龙. 基于决策树的高光谱遥感图像分类算法研究[D]. 秦皇岛: 燕山大学, 2014.

【17】Zhou Y J, Tian Q J. Image quality evaluation of EO-1 hyperion sensor[J]. Geo-Information Science, 2008, 10(5): 678-683.
周雨霁, 田庆久. EO-1 Hyperion高光谱数据的质量评价[J]. 地球信息科学, 2008, 10(5): 678-683.

【18】Yan L, Hu X B. Image quality assessment of IKONOS images based on contourlet-SIIM model[J]. Geomatics and Information Science of Wuhan University, 2014, 39(1): 12-16.
闫利, 胡晓斌. 利用Contourlet-SSIM视觉模型的IKONOS图像质量评价研究[J].武汉大学学报(信息科学版), 2014, 39(1): 12-16.

【19】Zhang X L, Liu Z F, Dai J B, et al. Image quality assessment based on visual perception and gradient structural similarity[J]. Application Research of Computers, 2011, 28(6): 2348-2351.
张晓琳, 刘直芳, 代金波, 等. 基于视觉感知的梯度结构相似度图像质量评价[J]. 计算机应用研究, 2011, 28(6): 2348-2351.

【20】Qian F, Sun T, Guo J, et al. Dynamic assessment of laser-dazzling effects based on the character of laser-spot and image feature[J]. Chinese Journal of Lasers, 2014, 41(6): 0609004.
钱方, 孙涛, 郭劲, 等. 基于光斑与图像特征的动态激光干扰效果评估[J]. 中国激光, 2014, 41(6): 0609004.

【21】Wang F, Ni J P, Dong T, et al. No-reference image quality assessment method based on visual attention mechanism and sharpness metric approach[J]. Journal of Applied Optics, 2018, 39(1): 51-56.
王凡, 倪晋平, 董涛, 等. 结合视觉注意力机制和图像锐度的无参图像质量评价方法[J]. 应用光学, 2018, 39(1): 51-56.

【22】Li S. Research on remote sensing image quality evaluation method based on statistical distribution[D]. Xi′an: Xidian University, 2017.
李帅. 基于统计分布的遥感图像质量评价方法研究[D]. 西安: 西安电子科技大学, 2017.

引用该论文

Xu Dongyu,Li Xiaorun,Zhao Liaoying,Shu Rui,Tang Qijia. Hyperspectral Image Quality Evaluation Based on Multi-Model Fusion[J]. Laser & Optoelectronics Progress, 2019, 56(2): 021101

徐冬宇,厉小润,赵辽英,舒锐,唐琪佳. 基于多模型融合的高光谱图像质量评价[J]. 激光与光电子学进展, 2019, 56(2): 021101

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