高级检索

  • ISSN 1006-3080
  • CN 31-1691/TQ

基于类别色彩查找表的彩色夜视方法

张玮雯 谷小婧 顾幸生

张玮雯, 谷小婧, 顾幸生. 基于类别色彩查找表的彩色夜视方法[J]. 华东理工大学学报(自然科学版), 2019, 45(6): 954-961. doi: 10.14135/j.cnki.1006-3080.20181018005
引用本文: 张玮雯, 谷小婧, 顾幸生. 基于类别色彩查找表的彩色夜视方法[J]. 华东理工大学学报(自然科学版), 2019, 45(6): 954-961. doi: 10.14135/j.cnki.1006-3080.20181018005
ZHANG Weiwen, GU Xiaojing, GU Xingsheng. Night-Vision Colorization Based on Category-Oriented Color Lookup Tables[J]. Journal of East China University of Science and Technology, 2019, 45(6): 954-961. doi: 10.14135/j.cnki.1006-3080.20181018005
Citation: ZHANG Weiwen, GU Xiaojing, GU Xingsheng. Night-Vision Colorization Based on Category-Oriented Color Lookup Tables[J]. Journal of East China University of Science and Technology, 2019, 45(6): 954-961. doi: 10.14135/j.cnki.1006-3080.20181018005

基于类别色彩查找表的彩色夜视方法

doi: 10.14135/j.cnki.1006-3080.20181018005
基金项目: 国家自然科学基金(61775058,61573144,61773165)
详细信息
    作者简介:

    张玮雯(1994-),女,上海人,硕士生,研究方向为图像彩色夜视。E-mail:vivianuy@126.com

    通讯作者:

    顾幸生,E-mail:xsgu@ecust.edu.cn

  • 中图分类号: TP391

Night-Vision Colorization Based on Category-Oriented Color Lookup Tables

  • 摘要: 针对夜视图像解读困难的问题,提出了一种面向类别的色彩查找表的彩色夜视方法。该方法可以将不同景物的特征色彩映射到多波段夜间图像,使观察者能够更快、更好地解读图像,从而提高态势感知并缩短反应时间。这种映射方法首先采用深度学习中的语义分割技术对图像进行分割,然后使用类别色彩查找表固定不同波段对应色彩,避免了全局彩色化时出现的色彩单一、不自然的缺点。实验结果表明,与其他全局彩色化方法相比,本文方法对于夜视图像的解析能力更强。

     

  • 图  1  基于类别色彩查找表的彩色夜视流程

    Figure  1.  Night-vision colorization framework with category-oriented CLUT

    图  2  并行全卷积神经网络结构

    Figure  2.  Structure of parallel fully convolutional network

    图  3  数据库示例图及其手工标注

    Figure  3.  Examples of database and its label image

    图  4  类别色彩查找表示例

    Figure  4.  Examples of category-oriented CLUT

    图  5  使用预设的类别色彩查找表进行彩色夜视(未进行图像均衡化矫正)

    Figure  5.  Night-vision colorization by the preset category-oriented CLUT (without post processing)

    图  6  加入后处理步骤的彩色夜视图结果对比

    Figure  6.  Night-vision colorization contrast with post processing

    图  7  不同彩色化方案的部分结果示例

    Figure  7.  Part results of different color fusion schemes

    图  8  不同彩色化方案的局部放大示例

    Figure  8.  Examples of partial magnification results of different colorization schemes

    表  1  通用景物和特征色

    Table  1.   General scenery and feature colors

    SceneryFeature color
    Targets (Persons, cars and other warning targets)Red
    VegetationsGreen
    SoilBrown
    Buildings/Roads/RocksGrey
    Sky/Clouds/WaterBlue
    OthersBlack
    下载: 导出CSV

    表  2  不同彩色化方法的图像质量评价指标的平均值(标准差)

    Table  2.   Average (with standard deviation) of image quality assessment metrics of different colorization schemes

    MethodISMICMNC
    MIT0.113 3(0.035 2)0.142 3(0.038 2)505(111)
    TNO0.067 0(0.033 8)0.260 0(0.065 7)524(207)
    BIT0.101 6(0.030 9)0.212 9(0.042 4)819(205)
    SM0.142 1(0.033 8)0.321 6(0.027 8)938(205)
    LUT0.120 3(0.033 9)0.190 3(0.046 3)397(89)
    Colorization (Iizuka 2016)0.133 2(0.040 8)0.095 6(0.048 2)703(235)
    Our method0.198 7(0.035 7)0.190 8(0.017 2)121 5(311)
    下载: 导出CSV
  • [1] MAHMOOD S, KHAN Y D, MAHMOOD M K. A treatise to vision enhancement and color fusion techniques in night vision devices[J]. Multimedia Tools & Applications, 2017, 77(2): 1-49.
    [2] ZHENG Y. An overview of night vision colorization techniques using multispectral images: From color fusion to color mapping[C]// International Conference on Audio, Language and Image Processing. USA: IEEE, 2012: 134-143.
    [3] TOET A, HOGERVORST M A. Progress in color night vision[J]. Optical Engineering, 2012, 51(1): 10901-10921. doi: 10.1117/1.OE.51.1.010901
    [4] WAXMAN A M, FAY D A. Color night vision: Fusion of intensified visible and thermal IR imagery[J]. Proceedings of SPIE, 1995, 2463: 58-68. doi: 10.1117/12.212755
    [5] TOET A, WALRAVEN J. New false color mapping for image fusion[J]. Optical Engineering, 1996, 35(3): 650-658. doi: 10.1117/1.600657
    [6] 王岭雪, 金伟其, 刘广荣, 等. 基于侧抑制特性的夜视图像彩色融合方法研究[J]. 北京理工大学学报, 2003, 23(4): 513-516. doi: 10.3969/j.issn.1001-0645.2003.04.028
    [7] TOET A. Applying daytime colors to multiband nightvision imagery[C]//Proceedings of the Sixth International Conference of Information Fusion. USA: IEEE, 2003: 614-621.
    [8] HOGERVORST M A, TOET A. Improved color mapping methods for multiband nighttime image fusion[J]. Journal of Imaging, 2017, 3(3): 1-25.
    [9] HOGERVORST M A, TOET A. Method for applying daytime colors to nighttime imagery in realtime[J]. Proceedings of SPIE, 2008, 6974: 697403-697412. doi: 10.1117/12.776648
    [10] TOET A, HOGERVORST M A. Perceptual evaluation of color transformed multispectral imagery[J]. Optical Engineering, 2014, 53(4): 1-12.
    [11] IIZUKA S, SIMOSERRA E, ISHIKAWA H. Let there be color!: Joint end-to-end learning of global and local image priors for automatic image colorization with simultaneous classification[J]. ACM Transactions on Graphics, 2016, 35(4): 1-11.
    [12] ZHANG R, ISOLA P, EFROS A A. Colorful image colorization[C]// European Conference on Computer Vision. Cham: Springer, 2016: 649-666.
    [13] LARSSON G, MAIRE M, SHAKHNAROVICH G. Learning representations for automatic colorization[C]// European Conference on Computer Vision. Cham: Springer, 2016: 577-593.
    [14] GU X, HE M, GU X. Thermal image colorization using Markov decision processes[J]. Memetic Computing, 2017, 9(1): 1-8. doi: 10.1007/s12293-017-0223-8
    [15] 吴骏逸, 谷小婧, 顾幸生. 基于可见光/红外图像的夜间道路场景语义分割[J]. 华东理工大学学报(自然科学版), 2019, 45(2): 301-309.
    [16] LONG J, SHELHAMER E, DARRELL T. Fully convolutional networks for semantic segmentation[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2017, 39(4): 640-651.
    [17] CHOI Y, KIM N, PARK K, et al. All-day visual place recognition: Benchmark dataset and baseline[C]// Computer Vision and Pattern Recognition Workshop on Visual Place Recognition in Changing Environments. USA: IEEE, 2015.
    [18] BENNETT E P, MASON J L, MCMILLAN L. Multispectral bilateral video fusion[J]. IEEE Transactions on Image Processing, 2007, 16(5): 1185-1194. doi: 10.1109/TIP.2007.894236
    [19] YUAN Y, ZHANG J, CHANG B. Objective quality evaluation of visible and infrared color fusion image[J]. Optical Engineering, 2011, 50(3): 103-108.
  • 加载中
图(8) / 表(2)
计量
  • 文章访问数:  7560
  • HTML全文浏览量:  2072
  • PDF下载量:  22
  • 被引次数: 0
出版历程
  • 收稿日期:  2018-10-18
  • 网络出版日期:  2019-07-17
  • 刊出日期:  2019-12-01

目录

    /

    返回文章
    返回