基于低分辨率红外阵列传感器的人体身份和动作识别

Abstract
Figure/Table
References
Related Citation (1)

Download: PDF (32815 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract Aiming at the problem of human identity and motion recognition, a method based on low resolution infrared array sensor and using convolutional neural network for classification and recognition is proposed, which can identify the identity of people and actions of falling, sitting and walking. The convolutional neural network used in this paper is based on VGGNet. It consists of input layer, five-layer convolutional layer, three-layer pooling layer, one layer of fully connected layer and output layer. It automatically extracts information features in infrared thermal images, and classifies actions, avoids the cumbersome manual extraction features under good privacy protection. After experimental testing, the average accuracy of convolutional neural network algorithm recognition is 93.3%, of which the walking recognition accuracy rate is 100%, the sitting recognition accuracy is 90%, the fall recognition accuracy is 90%, and the identity recognition accuracy is 96.7%.

Key words： low resolution infrared array sensor convolutional neural network motion recognition identification

Received: 15 May 2019 Published: 19 November 2019

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	Wang Zhaojun
	Xu Zhimeng

Cite this article:

Wang Zhaojun,Xu Zhimeng. Human identity and motion recognition based on low resolution infrared array sensor[J]. Electrical Engineering, 2019, 20(11): 6-10.

URL:

http://dqjs.cesmedia.cn/EN/Y2019/V20/I11/6

[1] 王刚. 基于多传感器的可穿戴跌倒检测系统的设计与实现[D]. 北京: 北京工业大学, 2015.
[2] 左常玲. 基于视频的自动摔倒检测研究与实现[D]. 合肥: 安徽大学, 2012.
[3] 李仲年, 臧春华, 杨刚, 等. 基于噪声嵌入的跌倒检测系统设计[J]. 微处理机, 2017, 38(2): 74-76, 81.
[4] Mashiyama S, Hong J, Ohtsuki T.A fall detection system using low resolution infrared array sensor[C]// 2014 IEEE 25th Annual International Symposium on Personal, Indoor and Mobile Radio Communication, 2014: 2109-2113.
[5] 杨任兵, 程文播, 钱庆, 等. 红外图像中基于多特征提取的跌倒检测算法研究[J]. 红外技术, 2017, 39(12): 1131-1138.
[6] 李倩玉, 蒋建国, 齐美彬. 基于改进深层网络的人脸识别算法[J]. 电子学报, 2017, 45(3): 619-625.
[7] 陈虹旭, 李晓坤, 郑永亮, 等. 基于深度学习的虹膜识别方法研究[J]. 智能计算机与应用, 2018, 8(2): 108-111.
[8] 田光见, 赵荣椿. 基于连续隐马尔可夫模型的步态识别[J]. 中国图象图形学报, 2006, 11(6): 867-871.
[9] Adhikari K, Bouchachia H, Nait-Charif H.Activity recognition for indoor fall detection using con- volutional neural network[C]//2017 Fifteenth IAPR International Conference on Machine Vision Appli- cations (MVA), 2017: 81-84
[10] Feng Pengming, Yu Miao, Naqvi S M, et al.Deep learning for posture analysis in fall detection[C]//2014 19th International Conference on Digital Signal Pro- cessing (DSP), 2014: 12-17.
[11] Vidigal M, Lima M, De Almeida Neto A. Elder falls detection based on artificial neural networks[C]//2015 Fourteenth Mexican International Conference on Artificial Intelligence (MICAI), 2015: 226-230.
[12] Kerdegari H, Mokaram S, Samsudin K A, et al.A pervasive neural network based fall detection system on smart phone[J]. Journal of Ambient Intelligence and Smart Environments, 2015, 7(2): 221-230.
[13] Kwapisz J R, Weiss G M, Moore S A.Activity recognition using cell phone accelerometers[J]. ACM SIGKDD Explorations Newsletter, 2010, 12(2): 74-82.
[14] Sebastian R. An overview of gradient descent opti- mization algorithms[EB/OL]. [2019-05-23]. http://sebasti- anruder.com/optimizing-gradient-descent/index.html.