基于神经网络的光照分布预测夜视复原算法

计算机科学 ›› 2019, Vol. 46 ›› Issue (11A): 329-333.

基于神经网络的光照分布预测夜视复原算法

邹鹏, 谌雨章, 陈龙彪, 曾张帆

(湖北大学计算机与信息工程学院武汉430062)

出版日期:2019-11-10 发布日期:2019-11-20
通讯作者: 谌雨章(1984-),男,博士,副教授,硕士生导师,主要研究方向为光电探测、图像处理,E-mail:hubucyz@foxmail.com。
作者简介:邹鹏(1997-),男,主要研究方向为图像处理、深度学习;陈龙彪(1997-),男,主要研究方向为深度学习、图像处理;曾张帆(1983-),男,博士,副教授,主要研究方向为信号处理、系统集成。
基金资助:
本文受国家自然科学基金(61601175),湖北省大学生创新训练项目基金(201810512051,201710512051)资助。

Night Vision Restoration Algorithm Based on Neural Network for Illumination Distribution Prediction

ZOU Peng, CHEN Yu-zhang, CHEN Long-biao, ZENG Zhang-fan

(School of Computer Science and Information Engineering,Hubei University,Wuhan 430062,China)

Online:2019-11-10 Published:2019-11-20

摘要/Abstract

摘要： 夜间图像存在光照不均匀、整体亮度较低、色偏严重的现象,且人工光源附近存在光晕。现有的去模糊模型和算法在光照不均匀情况下,常通过估计光照图来去除光照不均匀的影响。通过使用径向基函数神经网络训练提取光照强度,提出了基于光照估计的夜间图像去模糊算法。针对光照不均匀的问题,通过估计光照分布图来去除不均匀光照的影响,计算得到成像过程中的调制传递函数(MTF)。以计算所得传输图像退化模型的点扩散函数作为先决条件,结合半盲图像复原的数学模型对目标图像进行处理,以提高夜视探测的成像质量。将所提方法与传统盲复原方法及基于深度神经网络的图像复原方法进行主客观比较,实验所得复原图像及数据验证了该方法的有效性,复原图像的质量得到明显提升。

关键词: 半盲图像复原, 调制传递函数, 光照预测, 径向基函数神经网络, 夜间图像复原

Abstract: The illumination of the nighttime image is uneven,the overall brightness is low,the color deviation is large,and there is halo near the artificial light source.Existing deblurring models and algorithms often remove the effects of uneven illumination by estimating the illumination map in the case of uneven illumination.By combining the deep learning method with the radial basis function neural network,the illumination intensity was extracted,and the night image deblurring algorithm based on illumination estimation was proposed.For the problem of uneven illumination,the modulation transfer function (MTF) in the imaging process is calculated by estimating the illumination map.Taking the point diffusion function of the transport degrada-tion model as prior knowledge,combining the mathematical model of semi-blind image restoration method,the target image is processed to improve the quality of night vision imaging.In addition,the effectiveness of this method is verified by comparing with the traditional blind restoration method,and the image quality is improved evidently.

Key words: Lighting prediction, Modulation transfer function, Night vision image restoration, Radial basis function neural network, Semi-blind image restoration

中图分类号:

TP183

邹鹏, 谌雨章, 陈龙彪, 曾张帆. 基于神经网络的光照分布预测夜视复原算法[J]. 计算机科学, 2019, 46(11A): 329-333. https://doi.org/

ZOU Peng, CHEN Yu-zhang, CHEN Long-biao, ZENG Zhang-fan. Night Vision Restoration Algorithm Based on Neural Network for Illumination Distribution Prediction[J]. Computer Science, 2019, 46(11A): 329-333. https://doi.org/

参考文献

[1]杨帆.数字图像处理与分析[M].北京航空航天大学出版社,2015:103-105.
[2]HELSTROM C W.Image Restoration by the Method of Least Squares[J].Journal of the Optical Society of America A,1967,57(3):297-303.
[3]RICHARDSON W H.Bayesian-Based Iterative Method of Image Restoration*[J].Journal of the Optical Society of America,1972,62(1):55-59.
[4]LUCY L B.An iterative technique for the rectification of ob-served distributions[J].J of Astronomy,1974,79(6):745.
[5]PARK S C,PARK M K,KANG M G.Super-resolution image reconstruction:a technical overview[J].IEEE Signal Processing Magazine,2003,20(3):21-36.
[6]BERGEN J B.Hierarchical Model-Based Motion Estimation[C]∥Proceedings of Second European Conference on Computer Vision,1992,11(5):237-252.
[7]BONNIER D,LUTZ Y,LABRANCHE B.Modeling of an active TV system for surveillance operations[C]∥Aerosense.International Society for Optics and Photonics,1999.
[8]JOSHI N,SZELISKI R,KRIEGMAN D J.PSF Estimation using Sharp Edge Prediction[C]∥2008 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2008).Anchorage,Alaska,USA:IEEE,2008.
[9]XU L,JIA J.Two-Phase Kernel Estimation for Robust Motion Deblurring[J].Proceedings of European Conference on ComputerVision,2010,6311(9):157-170.
[10]CHO S,LEE S.Fast motion deblurring[C]∥ACM Transactions on Grphics,2009,28(5):1-8.
[11]方帅,赵育坤,李心科,等.基于光照估计的夜间图像去雾[J].电子学报,2016,44(11):2569-2575.
[12]王一宁,秦品乐,李传朋,等.基于残差神经网络的图像超分辨率改进算法[J].计算机应用,2018,38(1):246-254.
[13]TANG K,YANG J,WANG J.Investigating Haze-Relevant Features in a Learning Framework for Image Dehazing[C]∥2014 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).IEEE,2014.
[14]CAI B,XU X,JIA K,et al.DehazeNet:An End-to-End System for Single Image Haze Removal[J].IEEE Transactions on Image Processing,2016,25(11):5187-5198.
[15]梁中豪,彭德巍,金彦旭,等.基于交通场景区域增强的单幅图像去雾方法[J].计算机应用,2018,38(5):204-210.
[16]隋阳,孟钏楠,董玮,等.基于径向基函数神经网络直接提取布里渊散射谱温度的方法[J].光学学报,2018,38(12):394-400.
[17]张育贵,王义,杨人静.基于径向基神经网络的天气预测模型[J].贵州大学学报(自然版),2018,35(1):69-72.
[18]熊聪聪,潘璇,赵奇,等.多模式集成的RBF神经网络天气预报[J].天津科技大学学报,2014,29(1):75-78.
[19]朱福珍,李金宗,朱兵,等.基于径向基函数神经网络的超分辨率图像重建[J].光学精密工程,2010,18(6):1444-1451.
[20]ABBASCHIAN L,LIMA M S F D.Cracking susceptibility of aluminum alloys during laser welding[J].Materials Research,2003,6(2):273-278.
[21]张健,杨锐.基于径向基函数神经网络的激光焊接熔池光强分布预测[J].中国激光,2010,37(7):1856-1860.
[22]GRABEC I.Extraction of physical laws from joint experimental data[J].The European Physical Journal B - Condensed Matter and Complex Systems,2005,48(2):279-289.
[23]YOU Y L,KAVEH M.Blind image restoration by anisotropic regularization[J].IEEE Transactions on Image Processing A Publication of the IEEE Signal Processing Society,1999,8(3):396-407.
[24]胡海蕾.LED照明光学系统的设计及其阵列光照度分布研究[D].福州:福建师范大学,2005.
[25]熊玲玲,吕百达.描述激光二极管远场光强分布的理论模型[J].强激光与粒子束,2008,20(2):201-206.
[26]CHEN Y,YANG L,ZENG Z,et al.Degradation in LED night vision imaging and recovery algorithms[J].Optik - International Journal for Light and Electron Optics,2017,144(144):240-245.
[27]KATSAGGELOS A K.Digital image restoration[J].Proceed-ings of the IEEE,1978,66(7):812-812.
[28]YANG F,CHOI W,LIN Y.Exploit All the Layers:Fast and Accurate CNN Object Detector with Scale Dependent Pooling and Cascaded Rejection Classifiers[C]∥2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).IEEE Computer Society,2016.
[29]HAN H,ZHANG X,GE W.Performance Evaluation of Underwater Range-gated Viewing Based on Image Quality Metric[C]∥International Conference on Electronic Measurement & Instruments 2009(ICEMI 09).Beijing,2009:441-444.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed