Research of human fall detection algorithm based on improved Gaussian mixture model

doi:10.3778/j.issn.1002-8331.1604-0423

Computer Engineering and Applications ›› 2017, Vol. 53 ›› Issue (20): 173-179.DOI: 10.3778/j.issn.1002-8331.1604-0423

Previous Articles Next Articles

Research of human fall detection algorithm based on improved Gaussian mixture model

SUN Peng1，2, XIA Fei1，2，3, ZHANG Hao1，2，3, PENG Daogang1，2, MA Xi1，2, LUO Zhijiang1，2

1.College of Automation Engineering, Shanghai University of Electric Power, Shanghai 200090, China
2.Shanghai Engineering Research Center of Intelligent Management and Control for Power Process, Shanghai 200090, China
3.CMIS Center, Tongji University, Shanghai 200092, China

Online:2017-10-15 Published:2017-10-31

改进混合高斯模型在人体跌倒检测中的应用

孙朋1，2，夏飞1，2，3，张浩1，2，3，彭道刚1，2，马茜1，2，罗志疆1，2

1.上海电力学院自动化工程学院，上海 200090
2.上海发电过程智能管控工程技术研究中心，上海 200090
3.同济大学 CMIS中心，上海 200092

Abstract

Abstract: A hierarchical feature detection based on improved Gaussian mixture model is proposed in this paper to detect the falls. Aiming at the drawbacks of low background updating rate and shadow interference, background model is updated by improving the Gaussian mixture model and the shadow interference is eliminated through the characteristics of shadows in the HSV color space. Human minimum area external rectangle and vertical external rectangle are used to detect the human body, and the characteristics of rectangle ratio, centroid height ratio and inclination angle are analyzed in the paper. A hierarchical feature detection is proposed based on the different sensitivities of the characteristics. Firstly, the non-upright postures are detected by using human inclination angle. Secondly, the falls are confirmed by using the centroid height ratio and rectangle ratio successively. The experimental results show that the fall detection algorithm proposed in this paper has preferable adaptive performance and higher fall detection accuracy compared with different background subtraction algorithms and direct detection algorithms.

Key words: Gaussian mixture model, HSV, hierarchical feature, fall detection

摘要： 提出了一种基于改进混合高斯模型的分级特征检测算法，对人体跌倒状态进行检测。针对实际目标检测过程中背景更新缓慢，阴影干扰等缺点，通过改进混合高斯模型进行背景更新，并根据阴影区域在HSV颜色空间的特征信息消除阴影干扰。利用人体最小面积外接矩形和垂直外接矩形对检测到的人体目标进行标记，分析目标区域的矩形宽高比、人体质心高度比和人体躯干倾斜角的特征变化。根据各个特征对人体不同状态的判断灵敏度，提出了一种分级特征检测的方法。首先通过人体躯干倾角的特征，判断出人处于非站立状态。接下来依次采用人体质心高度比和矩形宽高比的特征，确认人体处于跌倒状态。实验结果证明，采用提出的方法对人体跌倒状态进行检测，其环境适应性和跌倒检测准确率均高于采用背景差分和直接检测的各种方法。

关键词: 混合高斯模型, HSV, 分级特征, 跌倒检测

SUN Peng1，2, XIA Fei1，2，3, ZHANG Hao1，2，3, PENG Daogang1，2, MA Xi1，2, LUO Zhijiang1，2. Research of human fall detection algorithm based on improved Gaussian mixture model[J]. Computer Engineering and Applications, 2017, 53(20): 173-179.

孙朋1，2，夏飞1，2，3，张浩1，2，3，彭道刚1，2，马茜1，2，罗志疆1，2. 改进混合高斯模型在人体跌倒检测中的应用[J]. 计算机工程与应用, 2017, 53(20): 173-179.

[1]	PAN Peixin, PAN Zhongliang. Active Contour Image Segmentation Combined with Saliency [J]. Computer Engineering and Applications, 2021, 57(8): 225-230.
[2]	LEI Henglin, Gulanbaier Tuerhong, Mairidan Wushouer, ZHANG Dongmei. Review of Novelty Detection [J]. Computer Engineering and Applications, 2021, 57(5): 47-55.
[3]	LI Chaochao, ZHANG Yuandong, WANG Changlong. Research on HSV Model for Identifying and Locating Round Plants [J]. Computer Engineering and Applications, 2020, 56(6): 220-223.
[4]	JIA Bingbing, CAO Hui, QIN Chijie. Research on Improving Phoneme Recognition Rate Based on Subspace Gaussian Mixture Model and Deep Neural Network Combination [J]. Computer Engineering and Applications, 2019, 55(24): 117-121.
[5]	ZHU Lianjie, CHEN Zhengyu, TIAN Chenlin. Review of Fall Detection Method Based on Wearable Devices [J]. Computer Engineering and Applications, 2019, 55(18): 8-14.
[6]	CHEN Chao. Target Tracking Algorithm Involving Gaussian Mixture Model and Weighted Likelihood [J]. Computer Engineering and Applications, 2019, 55(12): 124-131.
[7]	LI Chao, SUN Jun. Effective method of weld defect detection and classification based on machine vision [J]. Computer Engineering and Applications, 2018, 54(6): 264-270.
[8]	REN Lei1, ZHOU Jinhai1, WU Xiangfei2, JIN Tao1, JIN Xiaofeng1. Fall detection method based on adaptive position recognition for mobile devices [J]. Computer Engineering and Applications, 2018, 54(21): 7-12.
[9]	QIU Gongda1, HE Ming1, ZHU Chaozheng1, YANG Jie2, LIU Yong1. Fuzzy clustering based on connected point with max density in sparse border [J]. Computer Engineering and Applications, 2018, 54(14): 82-88.
[10]	LIANG Kaibin, GUAN Yihong. Brain MR Images segmentation method based on hidden Gaussian mixture model [J]. Computer Engineering and Applications, 2018, 54(10): 196-203.
[11]	SUN Kai，XIE Linbo. Moving objects detection method based on combination of improved local binary pattern and W4 algorithm [J]. Computer Engineering and Applications, 2017, 53(5): 187-191.
[12]	ZHAO Zhongtang1，3, CHEN Jiguang1，3, MA Qian2. Research on imbalance learning problem in fall detection [J]. Computer Engineering and Applications, 2017, 53(23): 142-146.
[13]	CHEN Hui, HU Likun, HUANG Yuwen. Stereo matching algorithm based on Gaussian mixture model and tree structure [J]. Computer Engineering and Applications, 2017, 53(20): 195-200.
[14]	YANG Xuesong, CAI Biye, ZHANG Jianming, LI Feng. Life jacket detection algorithm based on HSV color feature and contour area [J]. Computer Engineering and Applications, 2016, 52(3): 184-188.
[15]	NIU Yirong, WANG Shitong. Fast image segmentation algorithm based on noise benefit [J]. Computer Engineering and Applications, 2016, 52(21): 195-201.

Research of human fall detection algorithm based on improved Gaussian mixture model

改进混合高斯模型在人体跌倒检测中的应用

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics