提高人脸识别率的重光照算法

doi:10.3969/j.issn.1006-2475.2015.06.011

摘要/Abstract

摘要： 在逃犯追缉领域，基于人脸识别的逃犯鉴别技术受到高度重视。但是，各种恶劣的光照条件严重影响了对目标人脸的辨识。因此，如何排除光照干扰成为一个迫切需要解决的问题。本文提出一种新的重光照算法，利用自商图像算法求出人脸纹理特征，再利用辅助光照集合来归一化人脸光照分量，最后合成新的人脸图像用于人脸识别。在AR、CMU_PIE、CSA_Lighting库上进行人脸识别测试，结果表明改进的新算法有效地减轻了光照影响，提高了人脸正确识别率。在实际的追逃图片处理中也明显改善了追逃效率。

关键词: 重光照算法, 人脸识别, 朗伯体模型

Abstract: In escapee hunting, face recognition is applied widely. However, face recognition rate is affected by complex lighting environment. So it is a key problem that how to diminish the effect of illumination for face recognition. A re-lighting algorithm is proposed in this paper. First, the inherent texture features of the input image is extracted by selfquotient image algorithm. Then the illumination of the image is normalized based on an illumination bootstrap set. Finally the re-lighting image is produced by the combining texture image and the normalized illumination image. Experiments on AR, CMU_PIE and CAS_Lighting Face Database demonstrate that the proposed algorithm can improve the recall rate of face recognition algorithm, and further help to chase escapee.

Key words: re-lighting algorithm, face recognition, Lambertain model

中图分类号:

TP391

罗光明. 提高人脸识别率的重光照算法[J]. 计算机与现代化, doi: 10.3969/j.issn.1006-2475.2015.06.011.

LUO Guang-ming. A Re-Lighting Algorithm to Improve Recall Rate of Face Recognition[J]. Computer and Modernization, doi: 10.3969/j.issn.1006-2475.2015.06.011.

参考文献［18］

［1］	Adini Y, Moses Y, Ullman S. Face recognition: The problem of compensating for changes in illumination direction［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1997,19(7):721-732.
［2］	Lades M, Vorbruggen J C, Buhmann J, et al. Distortion invariant object recognition in the dynamic link architecture［J］. IEEE Transactions on Computers, 1993,42(3):300-311.
［3］	Ojala T, Pietikainen M, Maenpaa T. Multiresolution grayscale and rotation invariant texture classification with local binary patterns［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002,24(7):971-987.
［4］	Pizer S M, Amburn E P, Austin J D, et al. Adaptive histogram equalization and its variations［J］. Computer Vision, Graphics, and Image Processing, 1987,39（3）：355-368.
［5］	Shan Shiguang, Gao Wen, Cao Bo, et al. Illumination normalization for robust face recognition against varying lighting conditions［C］// IEEE International Workshop on Analysis and Modeling of Faces and Gestures 2003:157-164.
［6］	Jobson D J, Rahman Z, Woodell G A. Properties and performance of a center/surround retinex［J］. IEEE Transactions on Image Processing, 1997,6(3):451-462.
［7］	Amnon Shashua, Tammy RiklinRaviv. The quotient image: Class-based re-rendering and recognition with varying illuminations［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2014，23（2）：129-139
［8］	Riklin-Raviv T, Shashua A. The quotient image: Class based recognition and synthesis under varying illumination［C］// IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 1999,2:566-571.
［9］	Wang Haitao, Li S Z, Wang Yangsheng. Face recognition under varying lighting conditions using self quotient image［C］// The 6th IEEE International Conference on Automatic Face and Gesture Recognition. 2004:819-824.
［10］	Tan Xiaoyang, Triggs B. Enhanced local texture feature sets for face recognition under difficult lighting conditions［J］. IEEE Transactions on Image Processing, 2010,19(6):1635-1650.
［11］	Chen T, Yin W, Zhou X S, et al. Total variation models for variable lighting face recognition［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006,28(9):1519-1524.
［12］	Liu Jianyi, Zheng Nanning, Xiong Lei, et al. Illumination transition image:Parameter-based illumination estimation and rerendering［C］// The 19th International Conference on Pattern Recognition. 2008:1-4.
［13］	Xie Xiaohua, Zheng Wei-shi, Lai Jianhuang, et al. Normalization of face illumination based on large-and small-scale features［J］. IEEE Transaction on Image Processing, 2011,20(7):1807-1821.
［14］	Turk M, Pentland A. Eigenfaces for recognition［J］. Journal of Cognitive Neuroscience, 1991,3(1):71-86.
［15］	Belhumeur P N, Hespanha J P, Kriegman D J. Eigenfaces vs. fisherfaces:Recognition using class specific linear projection［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1997,19(7):711-720.
［16］	Ke Yan, Chen Youbin, Zhang David. Gabor surface feature for face recognition［C］// IEEE 2011 1st Asian Conference on Pattern Recognition (ACPR). 2011:288-292.
［17］	Sim T, Baker S, Bsat M. The CMU pose, illumination, and expression database［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003,25(12):1615-1618.
［18］	Gao Wen, Cao Bo, Shan Shiguang, et al. The CASPEAL largescale Chinese face database and baseline evaluations［J］. IEEE Transactions on Systems,Man and CyberneticsPart A：Systems and Humans, 2008,38(1):149-161.

[1]	王鑫, 肖韬睿. 基于生成对抗网络的人脸识别对抗攻击[J]. 计算机与现代化, 2023, 0(10): 115-120.
[2]	盛江岸, 陈淑荣. 融合双重注意力机制的戴口罩人脸识别方法[J]. 计算机与现代化, 2023, 0(02): 72-77.
[3]	陈凯,祖莉,欧屹. 基于YOLOv3与ResNet50的摄影机器人人脸识别跟踪系统[J]. 计算机与现代化, 2020, 0(04): 30-.
[4]	李硕1，卞青山2，刘传文2，刘鸣涛1，张林涛1 . 基于DCNN的证件照人脸验证及应用研究[J]. 计算机与现代化, 2020, 0(02): 104-.
[5]	崔娟娟1，张蕾1,侯谢炼2，陈才扣2，张海燕1. 基于鉴别性低秩表示的2阶段人脸识别算法[J]. 计算机与现代化, 2019, 0(12): 55-.
[6]	李垒，任越美. 基于多任务联合判别稀疏表示的人脸识别[J]. 计算机与现代化, 2019, 0(10): 72-.
[7]	黄武飞1，曾吉昌1，黄锦权1，李思寒1，陈育均1，陈钊1，郑志硕1,2. 基于实验室安保服务型机器人的人脸识别[J]. 计算机与现代化, 2019, 0(07): 61-.
[8]	郑健,王志明,张宁. 一种基于深度学习的改进人脸识别算法[J]. 计算机与现代化, 2018, 0(12): 90-.
[9]	潘峥嵘，王震. 一种结合Gabor小波与深度学习的人脸识别方法[J]. 计算机与现代化, 2018, 0(11): 46-.
[10]	郑秋梅，谢换丽，王风华，苏政，刘真. 一种基于局部三值模式的深度学习人脸识别算法[J]. 计算机与现代化, 2018, 0(02): 112-.
[11]	陈湘，王玲. 权值融合原始和虚拟样本的人脸识别方法[J]. 计算机与现代化, 2017, 0(3): 91-.
[12]	高亮1，潘积远2，于佳平3. 多元线性回归在非负矩阵分解人脸识别中的应用[J]. 计算机与现代化, 2017, 0(11): 41-45+54.
[13]	范引娣. 基于高阶结构约束的稀疏人脸识别算法[J]. 计算机与现代化, 2016, 0(9): 73-76.
[14]	汪方良，施慧彬. 基于OpenCL的人脸识别与隐藏系统[J]. 计算机与现代化, 2016, 0(1): 16-19,58.
[15]	刘铎. 基于CUDA的实时人脸识别系统[J]. 计算机与现代化, 2015, 0(6): 122-126.