LUO Kai,JIN Xiaofeng*.Research on fall behavior recognition based on Kinect[J].Journal of Yanbian University,2016,42(02):156-160.
基于Kinect的跌倒行为识别算法
- Title:
- Research on fall behavior recognition based on Kinect
- 分类号:
- TP391.41
- 文献标志码:
- A
- 摘要:
- 根据人体跌倒时的骨架特征,提出了一种人体跌倒行为识别方法.首先,依据跌倒行为的定义,将人体的头部和重心节点作为表征跌倒行为的特征参数,通过Kinect传感器获取人体骨架信息; 其次,采用滑动窗口和阈值方法确定行为的发生阶段,并提取其运动特征向量; 最后,通过人工神经网络对本文提取的跌倒行为特征进行训练和识别.实验结果表明,本文提出的方法高效准确,识别率达到90.5%.
- Abstract:
- According to the human skeleton feature in fall down, a method of human fall behavior recognition is proposed in this paper. Following fall behavior definition, joints head and hip-center are considered as feature to describe behavior of fall. First, we apply the Kinect sensor to obtain the human skeleton information. Then, we use slide-window and threshold for behavior initiation stage, motion feature vector is extracted. Finally, fall behaviors are trained and recognized by improved ANN. Experimental results show that the method proposed in this paper is high-efficiency and accurate, and its accuracy rate reaches 90.5%.
参考文献/References:
[1] Bogdan K, Michal K. Improving fall detection by the use of depth sensor and accelerometer[J]. Neurocomputing, 2015,168:637-645.
[2] Anastasiia K, Cyrille P L, Vasilli A G, et al. Falling in the elderly: do statistical models matter for performance criteria of fall prediction? Results from two large population-based studies[J]. European Journal of Internal Medicine, 2016,27:48-56.
[3] 王荣,章韵,陈建新.基于三轴加速度传感器的人体跌倒检测系统设计与实现[J].计算机应用,2012,32(5):1450-1452.
[4] 杨帆,谢静,周余,等.基于头部运动轨迹和3D视觉的跌倒检测系统[J].现代电子技术,2012,35(2):54-57.
[5] 邓小园.基于Kinect运动捕捉的高尔夫挥杆分析与辅助训练系统的研制[D].北京:北京邮电大学,2013:20-36.
[6] 刘飞.基于Kinect骨架信息的人体动作识别[D].上海:东华大学,2014:16-47.
[7] 郑立国.基于Kinect的动作捕捉系统的实现[J].吉林大学学报(工学版),2013,43:249-255.
[8] Buchner D M, Hornbrook M C, Kutner N G, et al. Development of the common data base for the FICSIT trials[J]. Journal of the American Geriatrics Society, 1993,41(3):297-308.
[9] 韩云,钟圣伦,叶正圣,等.以人体骨架为基础的室内实时动作侦测[C]//第32届中国控制会议论文集.西安.2013:3965-3969.
[10] 韩旭.应用Kinect的人体行为识别方法研究与系统设计[D].上海:山东大学,2013:31-36.
[11] Lv Q. A poselet-based approach for fall detection[C]//2011 IEEE International Symposium on IT in Medicine and Education, 2011,2:209-212.
[12] Laila A, Hussein Z, Ali A B. The implementation of an intelligent and video-based fall detection system using a neural network[J]. Applied Soft Computing, 2014,18(17):59-69.
[13] 邓万宇,郑庆华,陈琳,等.神经网络极速学习方法研究[J].计算机学报,2010,33(2):279-287.
相似文献/References:
[1]李思含,罗凯,金小峰.基于Kinect信息融合的移动平台目标定位算法研究[J].延边大学学报(自然科学版),2018,44(01):69.
LI Sihan,LUO Kai,JIN Xiaofeng*.Research on mobile platform target localization algorithmbased on Kinect information fusion[J].Journal of Yanbian University,2018,44(02):69.
备注/Memo
收稿日期: 2016-04-22 基金项目: 吉林省科技厅自然科学基金资助项目(20140101225JC)*通信作者: 金小峰(1970—),男,副教授,研究方向为智能信息处理.