计算机科学 ›› 2016, Vol. 43 ›› Issue (2): 105-108.doi: 10.11896/j.issn.1002-137X.2016.02.024

• 2015年中国计算机学会人工智能会议 • 上一篇    下一篇

改进Retinex的光照变化人脸图像增强算法

杜明,赵向军   

  1. 江苏师范大学计算机科学与技术学院 徐州221116,江苏师范大学计算机科学与技术学院 徐州221116
  • 出版日期:2018-12-01 发布日期:2018-12-01
  • 基金资助:
    本文受国家自然科学基金(61272297),江苏师范大学自然科学基金(13XLB03)资助

Face Enhancement Algorithm with Variable Illumination Based on Improved Retinex

DU Ming and ZHAO Xiang-jun   

  • Online:2018-12-01 Published:2018-12-01

摘要: 为了提高可变光照条件下的人脸图像整体效果,提出一种基于改进单尺度Retinex的光照变化人脸增强算法。首先对人脸图像进行对数变换,经过曲波变换得到高频和低频两部分;然后采用双边滤波对高频进行去噪处理,同时采用Kimmel变分模型对低频部分进行光滑滤波;最后对人脸图像进行重构,并对图像进行伽马校正处理。在Yale B光照人脸库上的实验结果表明,该算法能较好地防止“光晕”现象出现,可以还原出人脸图像的本来面貌,使人脸图像更加适合人眼观察。

关键词: 人脸识别,可变光照,图像增强,色彩恒常理论

Abstract: In order to improve the overall effect of face images with variable illumination,this paper proposed a novel face enhancement algorithm with variable illumination based on single scale Retinex.Firstly,the face images are logarithmically transformed,and the image is transformed into frequency and low frequency part by curvelet transform.Se-condly,the bilateral filtering is used to denoise the high frequency while Kimmel variation model is used to smooth filtering low frequency part.Finally,the image is reconstructed,and Gamma is used to correct the image.The experimental results on Yale B database show that the proposed algorithm can prevent the “halos” phenomenon,and can restore the original face image,so the face image is more suitable for human eye observation.

Key words: Face recognition,Variable illumination,Image enhancement,Retinex

[1] Chen T,Yin Wo-tao,Zhou Xiang-sean,et al.Total variationmodels for variable lighting face recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2006,8(9):1519-1524
[2] Cheng Yong,Zhao Chun-xia,Wang Huan,et al.Illumination Invariant and Face Recognition Based on NSCT and SQI[J].Journal of Nanjing University of Science And Technology,2010,4(4):425-430(in Chinese) 程勇,赵春霞,王欢,等.基于NSCT和SQI的光照不变量及人脸识别[J].南京理工大学学报(自然科学版),2010,4(4):425-430
[3] Provenzi E,Gatta C,Fierro M,et al.A spatially variant white-patch and gray-world method for color image enhancement dri-ven by local contrast[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2008,30(10):1757-1770
[4] Tan X,Triggs B.Enhanced local texture feature sets for face recognition under difficult lighting conditions [J].IEEE Trans on Image Processing,2010,19(6):1635-1650
[5] Young K P,Seokl P,Joong K K.Retinex method based on adaptive smoothing for illumination base on adaptive smoothing for illumination invariant face recognition[J].Signal Processing,2008,8:1929-1945
[6] Xie Xiao-hua,Lai Jian-huang,Zheng Wei-shi,et al.Illumination Normalization of Human Face Based on the Second-Order Polynomial Model[J].Acta Electronica Sinica,2010,8(8):1791-1797(in Chinese) 谢晓华,赖剑煌,郑伟诗,等.基于二次多项式模型的人脸光照归一化[J].电子学报,2010,8(8):1791-1797
[7] Wang Xiao-ming,Fang Xiao-ying,Liu Jin-gao.Adaptive face image enhancement under complex illumination[J].Computer Engineering and Applications,2011,7(2):15-18(in Chinese) 王小明,方晓颖,刘锦高.可变光照下的自适应人脸增强[J].计算机工程与应用,2011,7(2):15-18
[8] Jiang Ju-lang,Zhang You-sheng,Xue Feng,et al.Local Histogram Equalization with Br ightness Preservation[J].Acta Electronica Sinica,2006,4(5):861-866(in Chinese) 江巨浪,张佑生,薛峰,等.保持图像亮度的局部直方图均衡算法[J].电子学报,2006,34(5):861-866
[9] Song Shu-Lin,Zhang Yan,Wang Xian,et al.Illumination processing method based on Curvelet transform and Retinex[J].Computer Engineering and Applications,2013,9(3):171-173(in Chinese) 宋书林,张彦,王宪,等.基于曲波变换和Retinex人脸光照处理算法[J].计算机工程与应用,2013,9(3):171-173
[10] Li Fu-wen,Jin Wei-qi,Chen Wei-li,et al.Global Color Image Enhancement Algorithm Based on Retinex Mode[J].Transactions of Beijing Institute of Technology,2010,0(8):947-951(in Chinese) 李福文,金伟其,陈伟力,等.基于Retinex模型的彩色图像全局增强算法[J].北京理工大学学报,2010,0(8):947-951
[11] Yan Jun-xuan,Zhang Ke.HSI space based on luminance information in multi-Scale Retinex image enhancement algorithm[J].Computer Engineering and Aplications,2010,46(23),31-33(in Chinese) 闫骏宣,张科.HSI空间亮度信息的多尺度 Retinex 图像增强研究[J].计算机工程与应用,2010,46(23):31-33
[12] Wu Zhuang-zhi,Qi Li-na,Luo Pei.A Retinex Image Processing Framework Based on Estimation of Piecewise Smooth Illumination[J].Journal of National University of Defense Technology,2011,3(3):83-88(in Chinese) 吴壮志,齐立娜,骆沛.基于分片光滑照度估计的 Retinex 图像处理框架[J].国防科技大学学报,2011,33(3):83-88
[13] Cheng Yong,Hou Ying-kun,Zhao Chun-xia,et al.Robust face recognition based on illumination invariant in nonsubsampled Contourlet transform domain[J].Neurocomputing,2010,3(10):2217-2224
[14] Sun J G,Li Y,Meng X F,et al.Improved monoscale Retinex and application in face recognition[J].Application Research of Computers,2011,8(12):4790-4793(in Chinese) 孙劲光,李扬,孟祥福,等.改进的单尺度Retinex 及其在人脸识别中的应用[J].计算机应用研究,2011,8(12):4790-4793
[15] Ge Wei,Li Gui-ju,et al.Face image illumination processingbased on improved Retinex[J].Optics and Precision Engineering,2010,8(4):1011-1020(in Chinese) 葛微,李桂菊,等.利用改进的 Retinex 进行人脸图像光照处理 [J].光学精密工程,2010,8(4):1011-1020
[16] Vu N,Caplier A.Illumination-robust face recognition using retina modeling [C]∥16th IEEE International Conference on Image Processing (ICIP).2009:3289-3292
[17] Elad M.On the bilateral filter and ways to improve it [J].IEEE Transactions on Image Processing,2012,11(10):1141-1151
[18] Fan Chun-nian,Zhang Fu-yan.Illumination Invariant Extracting Algorithm Based on Nonsubsampled Contourlet Transform[J].Signal Processing,2012,8(4):507-513(in Chinese) 范春年,张福炎.基于非下采样 Contourlet 变换的光照不变量提取算法[J].信号处理,2012,8(4):507-513

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!