一种基于卷积神经网络的区域调光技术

doi:10.12068/j.issn.1005-3026.2021.05.003

东北大学学报（自然科学版） ›› 2021, Vol. 42 ›› Issue (5): 624-632.DOI: 10.12068/j.issn.1005-3026.2021.05.003

一种基于卷积神经网络的区域调光技术

张涛，刘天威，杜文丽

(天津大学电气自动化与信息工程学院，天津300072)

修回日期:2020-09-15 接受日期:2020-09-15 发布日期:2021-05-20
通讯作者: 张涛
作者简介:张涛(1975-) ，男，黑龙江北安人，天津大学副教授，博士生导师.
基金资助:
基金项目；(半空) 基金项目.天津市科技计划项目(16YFZCGX00760 ).

A Convolutional Neural Network Based Local Dimming Technology

ZHANG Tao， LIU Tian-wei， DU Wen-li

School of Electrical and Information Engineering， Tianjin University， Tianjin 300072， China.

Revised:2020-09-15 Accepted:2020-09-15 Published:2021-05-20
Contact: DU Wen-li
About author:-
Supported by:
-

摘要/Abstract

摘要： 由于光线串扰，像素补偿算法难以根据提取出的背光信息进行准确补偿，同时，单一补偿曲线难以适应具有不同亮度特点的图像内容，导致补偿图像的平均质量不高.为了提高像素补偿算法对复杂图像内容的适应性，本文引入神经网络中的编码和解码思想，通过编码网络提取图像深层特征，在解码网络中利用浅层特征的信息对深层特征进行解码，提出了一种联合分类回归的液晶像素补偿神经网络.实验结果表明，此网络得到的像素补偿图像不仅可以提高图像的主观质量，还在对比度、峰值信噪比等客观指标上取得了较好的效果.

关键词: 区域调光; 背光提取; 像素补偿; 卷积神经网络; 液晶显示器

Abstract: Due to the light spreading problem， pixel compensation algorithms are difficult to accurately compensate brightness according to the extracted backlight information. Besides， a single compensation curve is difficult to adapt to the complex image content， resulting in unsatisfactory image quality. In order to improve the adaptability of pixel compensation， the idea of encoding and decoding in neural network was introduced. The deep feature of image was extracted by encoding network， and it was decoded by using the information of shallow feature in decoding network. A kind of classification-regression compensation neural network (CRCNN) was proposed. The experimental results show that the pixel compensation image obtained by this network can not only improve the subjective quality of the image， but also achieve good results in contrast， peak signal-to-noise ratio and other objective indicators.

Key words: local dimming; backlight extraction; pixel compensation; convolutional neural network; liquid crystal display

中图分类号:

TN27
TN873

张涛，刘天威，杜文丽. 一种基于卷积神经网络的区域调光技术[J]. 东北大学学报（自然科学版）, 2021, 42(5): 624-632.

ZHANG Tao， LIU Tian-wei， DU Wen-li. A Convolutional Neural Network Based Local Dimming Technology[J]. Journal of Northeastern University(Natural Science), 2021, 42(5): 624-632.

参考文献

[1]Lim K，Wang F X，Li X H，et al.Benefits of emissive display (OLED TV) from a visual perception［J］. SID Symposium Digest of Technical Papers，2020，51(1):76-81.
[2]Kim S E，An J Y，Hong J J.How to reduce light leakage and clipping in local-dimming liquid-crystal displays［J］.Journal of the Society for Information Display，2009，17:1051-1057.
[3]Zhang L，Han D，Feng Q B.et al.Image-classification-based global dimming algorithm for led backlights in LCDS［J］.Optics & Laser Technology，2015，70:106-111.
[4]Shu X，Wu X，Forchhammer S.Optimal local dimming for LC image formation with controllable backlighting［J］.IEEE Transactions on Image Processing:A Publication of the IEEE Signal Processing Society，2013，22(1):166-173.
[5]Nam H，Song E，Kim S K.Weighted roll-off scheme to remove block artifacts for low power local dimming liquid crystal displays［J］.Optics & Laser Technology，2014，58:8-15.
[6]张喜条，吕国强，姚慧慧，等.LED分区背光光扩散性能研究［J］.光电工程，2014，7:72-78.(Zhang Xi-tiao，Lyu Guo-qiang，Yao Hui-hui，et al.Light spread in local-dimming LED-backlight［J］. Opto-Electronic Engineering，2014，7:72-78.)
[7]Liao L Y，Huang Y P，Yeh S C.A real-time image compensation for high dynamic range LCDs［J］.SID Symposium Digest of Technical Papers，2012，39(1):764-767.
[8]Zhang T，Wang Y F，Wu H Y，et al.High-performance local dimming algorithm based on image characteristic and logarithmic function［J］.Journal of the Society for Information Display，2019，27(1/2/3):85-100.
[9]Song S J，Kim Y I，Bae J，et al.Deep-learning-based pixel compensation algorithm for local dimming liquid crystal displays of quantum-dot backlights［J］.Optics Express，2019，27(11):15907.
[10]Ronneberger O，Fischer P，Brox T.U-Net:convolutional networks for biomedical image segmentation［J］.Medical Image Computing and Computer-Assisted Intervention，2015，9351:234-241.
[11]Funamoto T，Kobayashi T，Murao T.High-quality-picture technology for LCD televisions［J］.Matsushita Technical Journal，2000，46(3):12-19.
[12]Liu Y Z，Zheng X R，Chen J B.Dynamic backlight signal extraction algorithm based on threshold of image CDF for LCD-TV and its hardware implementation［J］.Chinese Journal of Liquid Crystals and Displays，2010，25(3):449-453.
[13]Lin F C，Liao C Y，Liao L Y.Inverse of mapping function (IMF) method for image quality enhancement of high dynamic range LCD TVs［J］.SID Symposium Digest of Technical Papers，2007，38(15):1343-1346.
[14]Cho H，Kwon O K.A backlight dimming algorithm for low power and high image quality LCD applications［J］.IEEE Transactions on Consumer Electronics，2009，55(2):839-844.
[15]Zhang X B，Liu X，Liu B，et al.A control algorithm of LCD dynamic backlight based on PSNR［J］.Applied Mechanics & Materials，2013，241/242/243/244:3014-3019.
[16]Chen S L，Tsai H J.A novel adaptive local dimming backlight control chip design based on gaussian distribution for liquid crystal displays［J］.Journal of Display Technology，2017，12(99):1494-1505.
[17]张涛，巫红英，李蒙，等.一种提高图像对比度和视觉质量的新型区域背光算法［J］.工程科学学报，2017，39(12):1888-1897.(Zhang Tao，Wu Hong-ying，Li Meng，et al.A novel local dimming algorithm to improve image contrast and visual quality［J］.Chinese Journal of Engineering，2017，39(12):1888-1897.)
[18]Zhang T，Du W L，Wang H，et al.A stronger adaptive local dimming method with details preservation［J］.Applied Sciences，2020，10(5):1820-1830.
[19]Otsu N.A threshold selection method from gray-level histograms［J］.IEEE Transactions on Systems Man & Cybernetics，2007，9(1):62-66.
[20]张磊.基于图像分析的液晶显示动态调光问题研究［D］.合肥:合肥工业大学，2015.(Zhang Lei.Research on liquid crystal display dimming issues based on image analysis［D］.Hefei:Hefei University of Technology，2015.)
[21]Zhang X B，Wang R，Dong D，et al.Dynamic backlight adaptation based on the details of image for liquid crystal displays［J］.Journal of Display Technology，2012，8(2):108-111.
[22]张涛，潘希豪.一种基于分段式曲线的液晶像素补偿方法:2017110775062［P］.2019-09-27.(Zhang Tao，Pan Xi-hao.A method of LCD pixel compensation based on piecewise curve:2017110775062［P］.2019-09-27.)
[23]Chen L C，Papandreou G，Kokkinos I，et al.Semantic image segmentation with deep convolutional nets and fully connected CRFs［J］.Computer Science，2014，4:357-361.
[24]Horé A，Ziou D.Image quality metrics:PSNR vs.SSIM［C］//20th International Conference on Pattern Recognition，ICPR 2010.Istanbul，Turkey，2010:2366-2369.

一种基于卷积神经网络的区域调光技术

A Convolutional Neural Network Based Local Dimming Technology

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 0

编辑推荐

Metrics

本文评价