基于智能视频监控的老人跌倒识别算法研究  

The Elderly Man Fall Recognition Algorithms Based on Video Monitoring

作　　者：万航[1] 陈立潮[1] 潘理虎[1] 谢斌红[1] 张英俊[1] 

机构地区：[1]太原科技大学计算机学院,太原030024

出　　处：《太原科技大学学报》2013年第4期245-249,共5页Journal of Taiyuan University of Science and Technology

基　　金：山西省自然科学基金项目(2009011022-1);山西省自然基金项目(2012011011-5);太原科技大学博士创新基金(20102030)

摘　　要：如何识别老年人跌倒是图像处理的一个技术难题,为了达到识别老人跌倒行为的目的,提出了一种新的老年人跌倒识别算法。利用三帧差法提取视频中的目标图像轮廓,然后对模糊轮廓进行去噪,膨胀腐蚀,Canny边缘检测等处理,然后求二值图像重心,利用重心的位置和加速度大小来判断老年人的行为是否为跌倒,跌倒则发出报警信息。实验结果表明,该算法能达到较好的识别结果。How to identify the elderly man falls is a difficult technical problem in image processing. In order to a-chieve the aim of identifying the elderly man fall behaviors, this paper proposes a new people fall identification al- gorithm. Three frame differential methods were used to extract the contour of the target image in the video, and then the fuzzy outline denoising, expansion, corrosion, such as the Canny edge detection processing were done, and then the center of gravity of the binary image was obtained, and the center of gravity location and magnitude of the accel- eration was used to determine the behaviors of older people whether they fall or not, the alarm information will be sent if they fall. The experimental results show that the algorithm is able to achieve better recognition results.

关 键 词：跌倒检测 三帧差法 重心 边缘检测 数学形态学 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]