基于网络摄像头的活体人脸识别系统设计

doi:10.16180/j.cnki.issn1007-7820.2020.06.012

电子科技 ›› 2020, Vol. 33 ›› Issue (6): 63-68.doi: 10.16180/j.cnki.issn1007-7820.2020.06.012

基于网络摄像头的活体人脸识别系统设计

王春江¹,张猛²,张建飞¹

^1. 中国空空导弹研究院, 河南洛阳 471000
^2. 西安电子科技大学人工智能学院, 陕西西安 710071

收稿日期:2019-06-11 出版日期:2020-06-15 发布日期:2020-06-18
作者简介:王春江(1982-),男,工程师。研究方向:图像跟踪与信息处理技术。|张猛(1993-),男,硕士研究生。研究方向:嵌入式开发,图像处理。|张建飞(1986-),男,硕士研究生。研究方向:信号与信息处理。
基金资助:
国家自然科学基金(61372071)

Design of Living Face Recognition System Based on Web Camera

WANG Chunjiang¹,ZHANG Meng²,ZHANG Jianfei¹

^1. China Air-borne Missile Academy, Luoyang 471000, China
^2. School of Artificial Antelligence, Xidian University, Xi’an 170071, China;

Received:2019-06-11 Online:2020-06-15 Published:2020-06-18
Supported by:
National Natural Science Foundation of China(61372071)

摘要/Abstract

摘要：

针对当前监控系统平台体积大、功能单一、造价高等问题,文中基于海思3518E平台设计了一套在线活体人脸识别系统。该系统包含图像预处理、图像获取、人脸检测、人脸活体检测、人脸识别共5个部分,并针对传统算法在嵌入式平台的不足进行了改进与优化。系统通过摄像头采集的人脸图像对图像做预处理,为后续图像特征提取提供保障,利用扩展Haar特征训练分类器,并使用Adaboost算法级联分类器进行人脸检测,将检测到的人脸利用HSV和Ycbcr多色彩空间下提取的COALBP和LPQ融合特征训练SVM模型,并进行活体人脸检测。最后,对人脸图像分块提取LBP特征进行人脸识别,将识别结果通过微信小程序显示。实验结果表明,基于海思网络摄像头人脸识别系统的可行性,具有一定的实用价值。

关键词: 人脸识别, 活体检测, 海思, 人脸检测, ARM平台, Haar, Adaboost

Abstract: Aim

ing at the problems of large size, single function and high cost of the current monitoring system platform, this paper designs an online living face recognition system based on the HiSilicon 3518E platform. The system includes five parts: image pre-processing, image acquisition, face detection, face live detection, and face recognition. It also improved and optimized the deficiencies of traditional algorithms in embedded platforms. The system collected the face image through the camera and preprocessed the image to provided guarantee for subsequent image feature extraction. The extended Haar feature was used to train the classifier and the Adaboost algorithm was used to cascade face detection. The detected face used HSV and Ycbcr multi-color space the COALBP and LPQ fusion features extracted below were used to train the SVM model, perform live face detection, and finally extract LBP features from the face image to perform face recognition, and the recognition results were displayed through the WeChat applet. The experimental results showed that the feasibility of the face recognition system based on the HiSilicon web camera had certain practical value.

Key words: face recognition, live detection, Hisilicon, face detection, ARM platform, Haar, Adaboost

中图分类号:

TN919.5

王春江,张猛,张建飞. 基于网络摄像头的活体人脸识别系统设计[J]. 电子科技, 2020, 33(6): 63-68.

WANG Chunjiang,ZHANG Meng,ZHANG Jianfei. Design of Living Face Recognition System Based on Web Camera[J]. Electronic Science and Technology, 2020, 33(6): 63-68.

图/表 17

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

图11

图12

图13

图14

图15

图16

图17

参考文献 14

[1]	李侃, 廖启征 .基于 S3C2410 平台与嵌入式 Linux 的图像采集应用[J]. ARM 开发与应用, 2006,22(8):125-168.
	Li Kan, Liao Qizheng . Image acquisition application based on S3C2410 platform and embedded Linux[J]. ARM Development and Application, 2006,22(8):125-168.
[2]	Hill H, Bruce V . Effects of lighting on matching facial surfaces[J]. Journal of Experimental Psychology: Human Perception and Performance, 1996(22):986-1004.
[3]	Wang Jing, Su Guangda . Sparse representation for face recognition based on constraint sampling and face alignment[J]. IEEE Transactions on Computer Science, 2013,18(1):62-67.
[4]	Viola P, Jones M J. Robust real-time face detection[C]. International Journal of Computer Vision, 2004,57(2):137-154.
[5]	Paul Viola, Jones M . Fast and robust classification using asymmetric[J]. NIPS, 2001(9):36-48.
[6]	Lienhart R, Maydt J . An extended set of Haar-like features for rapid object detection[J]. IEEE Image Processing, 2002(3):900-903.
[7]	Viola P, Jones M . Rapid object detection using a boosted cascade of simple features[C]. Boston:Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 200l(1):511-518.
[8]	Kim G, Eum S, Suhr J K, et al. Face liveness detection based on texture and frequency analyses[C]. New York:5 ^th International Conference on IAPR,IEEE , 2012.
[9]	Qi X, Xiao R, Li C G , et al. Pairwise rotation invariant cooccurrence local binary pattern[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2014,36(11):2199-2213.
[10]	Ojansivu V, Janne Heikkilä. Blur insensitive texture classification using local phase quantization[C]. Berlin:International Conference on Image & Signal Process,Springer-Verlag, 2008.
[11]	Ctores C, Vapnik V N . Support Vector Network[J]. Machine Learning, 1995(20):1-25.
[12]	Deng Zhang, Shingo Mabu, Kotaro Hirasawa . Robust intelligent PCA-based face recognition framework using GNP-fuzzy data mining[J]. IEEE Transactions on Electrical and Electronic Engineering, 2013,8(3):253-262.
[13]	Hi Silicon Technologies.Hi3518E datasheet[EB/OL].( 2012- 07- 16)[2019-03-11]http://www.datasheetcafe.com/hi3518-datasheet-pdf .
[14]	Toshiba America Electronic Components.SC2235 datasheet[EB/OL].( 2013- 11- 23)[2019-03-15]http://www.smartsenstech.com/product.html .

[1]	程晓雅,张雷. 基于深度学习的有遮挡人脸识别方法研究[J]. 电子科技, 2022, 35(1): 35-39.
[2]	彭荣杰,彭亚雄,陆安江. 基于改进PCA+SVM的人脸识别系统[J]. 电子科技, 2021, 34(12): 56-61.
[3]	杨思燕,苗凯彬,王锋,苗启广. 视频图像中人脸自动检测与统计算法[J]. 电子科技, 2020, 33(8): 1-9.
[4]	司琴,李菲菲,陈虬. 基于深度学习与特征融合的人脸识别算法[J]. 电子科技, 2020, 33(4): 18-22.
[5]	汪欣,吴薇,曾照. 基于视频的人脸检测算法研究[J]. 电子科技, 2020, 33(2): 25-31.
[6]	田雄,吴薇,刘晓尚,吴秀. 多人脸跟踪与最佳人脸提取[J]. 电子科技, 2019, 32(9): 32-37.
[7]	苏饶,李菲菲,陈虬. 基于多特征融合的人脸识别算法[J]. 电子科技, 2019, 32(7): 43-48.
[8]	刘芬. 基于线性和非线性算法的高效人脸识别系统[J]. 电子科技, 2019, 32(7): 82-86.
[9]	胡少聪. 基于深度学习的人脸识别方法研究[J]. 电子科技, 2019, 32(6): 82-86.
[10]	周柱,甘屹,孙福佳. 一种融合Gabor+SIFT特征的人脸识别算法[J]. 电子科技, 2019, 32(4): 1-5.
[11]	殷涛,崔佳冬. 视频监控行人流量统计系统的设计[J]. 电子科技, 2019, 32(12): 48-52.
[12]	于世强，应捷. 基于Android平台的实时人脸检测与跟踪[J]. , 2016, 29(7): 78-.
[13]	肖儿良,韩超. 应用于人脸自动实时跟踪的改进Struck算法[J]. , 2016, 29(3): 42-.
[14]	焦淑红,孙志帅. 基于GPCA的KNNY与SVM融合的人脸识别方法[J]. , 2016, 29(2): 74-.
[15]	黄华盛,杨阿庆. 基于PCA算法的人脸识别[J]. , 2015, 28(8): 98-.

基于网络摄像头的活体人脸识别系统设计

Design of Living Face Recognition System Based on Web Camera

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 17

参考文献 14

相关文章 15

Metrics

本文评价

推荐阅读 10