《计算机应用研究》|Application Research of Computers

基于场景识别的夜视图像彩色融合方法

Night vision image color fusion method based on scene recognition

免费全文下载 (已被下载 次)  
获取PDF全文
作者 瞿哲,肖刚,徐宁文,刁卓然
机构 上海交通大学 航空航天学院,上海 200240
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2018)03-0944-05
DOI 10.3969/j.issn.1001-3695.2018.03.064
摘要 将红外与微光图像进行彩色融合可以使目标检测和识别更精确。针对现有的使用固定参考图像的彩色传递融合方法,提出了一种新的结合场景分类、融合质量评价以及彩色传递的融合方法。首先提取输入图像的GIST特征并利用SVM分类器进行场景的分类,随后利用彩色融合质量评价方法在相应类别的图像库中匹配最佳的参考图像,最终使用彩色传递的方法将红外与微光图像融合为彩色图像。仿真结果表明,相较于其他彩色融合方法,该方法在无须观测环境先验信息的情况下,能够在线性以及非线性彩色空间使融合图像接近自然真实的色彩感觉,更易于分辨识别目标,从而达到提高机器视觉效率的目的。
关键词 彩色融合;红外图像;微光图像;场景分类;质量评价
基金项目 国家“973”计划资助项目(2014CB744903)
中国航天科技集团公司航天科技创新基金资助项目(HTKJCX2015CAAA09)
上海市浦江人才计划资助项目
上海市军民融合专项资助项目
本文URL http://www.arocmag.com/article/01-2018-03-064.html
英文标题 Night vision image color fusion method based on scene recognition
作者英文名 Qu Zhe, Xiao Gang, Xu Ningwen, Diao Zhuoran
机构英文名 SchoolofAeronautics&Astronautics,ShanghaiJiaoTongUniversity,Shanghai200240,China
英文摘要 Infrared and low light level images color fusion can make the target detection and recognition more accurate. Aimed at the existing color fusion methods based on color transfer, this paper proposed a new fusion method that combined scene classification, fusion quality evaluation and color transfer. Firstly, it extracted GIST features of the input image and used SVM classifier to do the scene classification, then it used the color fusion quality evaluation method to find the best matched reference image in the image library of the corresponding category, finally it got the color fusion result of infrared and low light level image using color transfer method. The simulation results show that compared with other color fusion methods, this method doesn’t need prior information of observation environment, and can display the fusion result in natural colors under both linear and non-linear color space, which makes the target identification more accurate, therefore improving the efficiency of the machine vision system.
英文关键词 color fusion; infrared image; low light level image; scene classification; quality evaluation
参考文献 查看稿件参考文献
  [1] Toet A, Hogervorst M A. Progress in color night vision[J] . Optical Engineering, 2012, 51(1):010901.
[2] Zheng Yufeng, Essock E A. A local-coloring method for night-vision colorization utilizing image analysis and fusion[J] . Information Fusion, 2008, 9(2):186-199.
[3] Lee H, Kim H S, Kim J I. Background subtraction using background sets with image-and color-space reduction[J] . IEEE Trans on Multimedia, 2016, 18(10):2093-2103.
[4] 禹晶, 李大鹏, 廖庆敏. 基于颜色恒常性的低照度图像视见度增强[J] . 自动化学报, 2011, 37(8):923-931.
[5] Van De Sande K E A, Gevers T, Snoek C G M. Evaluating color descriptors for object and scene recognition[J] . IEEE Trans on Pattern Analysis and Machine Intelligence, 2010, 32(9):1582-1596.
[6] Oliva A, Torralba A. Modeling the shape of the scene:a holistic representation of the spatial envelope[J] . International Journal of Computer Vision, 2001, 42(3):145-175.
[7] Ji Aibing, Chen Songcan, Hua Qiang. Fuzzy classifier based on fuzzy support vector machine[J] . Journal of Intelligent & Fuzzy Systems, 2014, 26(1):421-430.
[8] Gondal I, Murshed M. A novel color image fusion QoS measure for multi-sensor night vision applications[C] //Proc of IEEE Symposium on Computers and Communications. 2010:399-404.
[9] Panetta K, Bao Long, Agaian S. A human visual “no-reference” image quality measure[J] . IEEE Instrumentation & Measurement Magazine, 2016, 19(3):34-38.
[10] Xue Wufeng, Zhang Lei, Mou Xuanqin, et al. Gradient magnitude similarity deviation:a highly efficient perceptual image quality index[J] . IEEE Trans on Image Processing, 2014, 23(2):684-695.
[11] Fay D A, Waxman A M, Aguilar M, et al. Fusion of multi-sensor imagery for night vision:color visualization, target learning and search[C] // Proc of the 3rd International Conference on Information Fusion. 2000.
收稿日期 2016/10/8
修回日期 2016/11/29
页码 944-948
中图分类号 TP391
文献标志码 A