|本期目录/Table of Contents|

[1]郭 婷,张天序*,郭诗嘉.一种红外图像和宽光谱融合的人脸识别算法[J].武汉工程大学学报,2022,44(03):320-324.[doi:10.19843/j.cnki.CN42-1779/TQ.202111015]
 GUO Ting,ZHANG Tianxu*,GUO Shijia.Face Recognition Algorithm Based on Fusion of Infrared Image and Wide Spectrum Features[J].Journal of Wuhan Institute of Technology,2022,44(03):320-324.[doi:10.19843/j.cnki.CN42-1779/TQ.202111015]
点击复制

一种红外图像和宽光谱融合的人脸识别算法(/HTML)
分享到:

《武汉工程大学学报》[ISSN:1674-2869/CN:42-1779/TQ]

卷:
44
期数:
2022年03期
页码:
320-324
栏目:
机电与信息工程
出版日期:
2022-06-30

文章信息/Info

Title:
Face Recognition Algorithm Based on Fusion of Infrared Image and Wide Spectrum Features
文章编号:
1674 - 2869(2022)03 - 0320 - 05
作者:
郭 婷1张天序*2郭诗嘉1
1.武汉工程大学电气信息学院,湖北 武汉 430205;
2.华中科技大学图像识别与人工智能研究所,湖北 武汉 430074
Author(s):
GUO Ting1ZHANG Tianxu*2GUO Shijia1
1. School of Electrical and Information Engineering, Wuhan Institute of Technology, Wuhan 430205,China;
2. Institute of Image Recognition and Artificial Intelligence, Huazhong University of Science and Technology,Wuhan 430074,China

关键词:
人脸识别红外图像宽光谱自适应权重特征融合
Keywords:
face recognition thermal infrared image wide spectrum adaptive weight feature fusion
分类号:
TP391
DOI:
10.19843/j.cnki.CN42-1779/TQ.202111015
文献标志码:
A
摘要:
针对可见光人脸识别算法难以适应弱光照、面部涂装、夸张表情等场景的问题,提出了一种基于热红外图像和宽光谱特征融合的人脸识别算法。采用了热红外和宽光谱两个特征模块提取鲁棒的模态特征;设计了一种自适应权重的特征融合方法,该方法可以自主学习各模态特征的融合权重,从而引导分类模型更多的关注具有更好判别性的特征。结果表明:基于图谱融合的人脸识别算法达到了96%的人脸识别率;相比于基于单模态信息的人脸识别,融合人脸识别算法有效的提升了识别的精确度和稳定性,同时,受益于红外图像和宽光谱模态信息的特有属性,图谱融合人脸识别算法提升了人脸识别技术对复杂场景的适应性,扩宽了其应用场景。
Abstract:
Aiming at the problem that the visible light face recognition algorithm is difficult to adapt to the scenes of weak light, facial painting, exaggerated expression and photo deception,a face recognition algorithm based on the fusion of thermal infrared image and wide spectrum features was proposed. Two feature modules, thermal infrared and wide spectrum, were used to extract robust modal features. A feature fusion method with adaptive weight was designed, which can independently learn the fusion weight of each modal feature, so as to guide the classification model to pay more attention to the features with better discrimination. The results show that the face recognition algorithm based on atlas fusion achieves 96% face recognition rate; compared with face recognition based on single-mode information, the fusion face recognition algorithm effectively improves the accuracy and stability of recognition. At the same time, benefiting from the unique attributes of infrared image and wide spectrum modal information, the atlas fusion face recognition algorithm improves the adaptability of face recognition technology to complex scenes and widens its application scenarios.

参考文献/References:

[1] 庄薪霖. 人脸识别方法综述[J]. 科技创新与应用, 2022, 12(2):130-132.

[2] KORTL Y, JRIDI M, Al FALOU A, et al. Face recognition systems: a survey[J]. Sensors, 2020, 20(2):342.
[3] 陈希彤, 卢涛. 基于全局深度分离卷积残差网络的高效人脸识别算法[J]. 武汉工程大学学报,2019,41(3):276-282.
[4] 曾华. 红外人脸图像识别研究[D]. 武汉: 武汉工程大学, 2011.
[5] POSTER D D, HU S, SHORT N J, et al. Visible-to-thermal transfer learning for facial landmark detection[J]. IEEE Access, 2021(9):52759-52772.
[6] MUNIR R, KHAN R A. An extensive review on spectral imaging in biometric systems: challenges & advancements[J]. Journal of Visual Communication and Image Representation, 2019, 65: 102660.
[7] SINGH R, AHMED T, SINGH R, et al. Identifying tiny faces in thermal images using transfer learning[J]. Journal of Ambient Intelligence and Humanized Computing, 2020, 11(5): 1957-1966.
[8] 栗科峰, 黄全振, 卢金燕. 基于热红外光谱的人脸特征提取算法[J]. 激光与光电子学进展,2019,56(2): 1-8.
[9] 马娜. 基于高光谱图像的人脸识别算法和实验研究[D]. 济南: 山东师范大学, 2017.
[10] 谢志华, 李毅, 牛杰一. 联合分块谱带优选和深度特征的高光谱人脸识别[J]. 中国图象图形学报, 2021, 26(12): 2870-2878.
[11] 袁璞. 基于多任务和特征融合的显著性目标检测研究[D]. 赣州: 江西理工大学, 2021.
[12] 仁泽裕, 王振超, 柯尊旺. 多模态数据融合综述[J]. 计算机工程与应用, 2021, 57(18): 49-64.
[13] 褚玉晓. 基于融合深度卷积神经网络的人脸识别方法研究[J]. 信息与电脑, 2021, 33(10):173-175.
[14] KANMANI M, NARASIMHAN V. Optimal fusion aided face recognition from visible and thermal face images[J]. Multimedia Tools and Applications, 2020, 79(25): 17859-17883.
[15] 何承鹏. 基于红外人脸识别的汽车防盗系统[D]. 长春: 长春工业大学, 2020.

相似文献/References:

[1]夏平平,吕太之.动态人脸识别系统的设计与实现[J].武汉工程大学学报,2011,(10):107.
 XIA Ping ping,LV Tai zhi.Design and implementation of dynamic faces recognition system[J].Journal of Wuhan Institute of Technology,2011,(03):107.
[2]韦 仙,康睿丹.基于降维压缩法的图像重构[J].武汉工程大学学报,2015,37(12):69.[doi:10. 3969/j. issn. 1674-2869. 2015. 12. 015]
 -.Image reconstruction based on dimension reduction and compression technology[J].Journal of Wuhan Institute of Technology,2015,37(03):69.[doi:10. 3969/j. issn. 1674-2869. 2015. 12. 015]
[3]陈希彤,卢 涛*.基于全局深度分离卷积残差网络的高效人脸识别算法[J].武汉工程大学学报,2019,(03):276.[doi:10. 3969/j. issn. 1674-2869. 2019. 03. 014]
 CHEN Xitong,LU Tao *.Efficient Face Recognition Algorithm Using Global Deep Separable Convolutional and Residual Network[J].Journal of Wuhan Institute of Technology,2019,(03):276.[doi:10. 3969/j. issn. 1674-2869. 2019. 03. 014]

备注/Memo

备注/Memo:
收稿日期:2021-11-17
作者简介:郭 婷,硕士研究生。E-mail: 524276702@qq.com
*通讯作者:张天序,博士,教授。E-mail: txzhang@hust.edu.cn
引文格式:郭婷, 张天序, 郭诗嘉. 一种红外图像和宽光谱融合的人脸识别算法[J]. 武汉工程大学学报,2022,44(3):320-324.

更新日期/Last Update: 2022-06-29