《计算机应用》唯一官方网站 ›› 2022, Vol. 42 ›› Issue (3): 757-763.DOI: 10.11772/j.issn.1001-9081.2021040857

• 2021年中国计算机学会人工智能会议(CCFAI 2021) • 上一篇    

基于Res2Net-YOLACT和融合特征的室内跌倒检测算法

张璐, 方春, 祝铭()   

  1. 山东理工大学 计算机科学与技术学院,山东 淄博 255049
  • 收稿日期:2021-05-25 修回日期:2021-06-30 接受日期:2021-07-06 发布日期:2021-11-09 出版日期:2022-03-10
  • 通讯作者: 祝铭
  • 作者简介:张璐(1996—),男,山东潍坊人,硕士研究生,主要研究方向:计算机视觉、深度学习
    方春(1981—),女,山东淄博人,讲师,博士,主要研究方向:智能计算、模式识别;
  • 基金资助:
    国家自然科学基金资助项目(61602280);山东省高等学校优秀青年创新团队支持计划项目(2019KJN048);淄博市校城融合发展计划项目(2019ZBXC114)

Indoor fall detection algorithm based on Res2Net-YOLACT and fusion feature

Lu ZHANG, Chun FANG, Ming ZHU()   

  1. School of Computer Science and Technology,Shandong University of Technology,Zibo Shandong 255049,China
  • Received:2021-05-25 Revised:2021-06-30 Accepted:2021-07-06 Online:2021-11-09 Published:2022-03-10
  • Contact: Ming ZHU
  • About author:ZHANG Lu, born in 1996, M. S. candidate. His research interests include computer vision, deep learning.
    FANG Chun, born in 1981, Ph. D., lecturer. Her research interests include intelligence computation, pattern recognition.
  • Supported by:
    National Natural Science Foundation of China(61602280);Outstanding Youth Innovation Team Support Program of Shandong Colleges and Universities(2019KJN048);University and City Integration Development Program of Zibo City(2019ZBXC114)

摘要:

为了加强对老年人的监护、降低跌倒带来的安全风险,提出了一种新的基于Res2Net-YOLACT和融合特征的室内跌倒检测算法。首先,通过融入Res2Net模块的YOLACT网络来提取视频图像序列中的人体轮廓;然后,利用两级判断的方法做出跌倒决策,其中一级判别通过运动速度特征粗略判断是否发生异常状态,二级通过融合人体形状特征和深度特征的模型结构对人体姿势进行判别;最后,当检测出跌倒且发生时间大于阈值时,发出跌倒报警。实验结果表明,该跌倒检测算法可以在复杂的场景下很好地提取到人体轮廓,对光照的鲁棒性较好,并且检测速度可达每秒28帧,能满足实时检测要求。此外,融入手工特征后的算法分类性能表现更优,分类准确率达98.65%,比卷积神经网络(CNN)特征算法提升了1.03个百分点。

关键词: 健康监护, YOLACT, 融合特征, 卷积神经网络, 跌倒检测

Abstract:

In order to strengthen the monitoring of old people and reduce the safety risks caused by falls, a new indoor fall detection algorithm based on Res2Net-YOLACT and fusion feature was proposed. For the video image sequences, firstly, the YOLACT network integrated with Res2Net module was used to extract the human body contour, and then a two-level judgment method was used to make a fall decision. In the first level, whether an abnormal state occurs was judged roughly through the movement speed feature, and in the second level, the human body posture was determined through the model structure that combines the body shape features and the depth feature. Finally, when fall posture was detected and the occurrence time was greater than the threshold, a fall alarm was given. Experimental results show that the proposed fall detection algorithm can extract the human body contour well in complex scenes, which has good robustness to illumination as well as a real-time performance of up to 28 fps (frames per second). In addition, the classification performance of the algorithm after adding manual features is better, the classification accuracy is 98.65%, which is 1.03 percentage points higher than that of the algorithm with original CNN (Convolutional Neural Network) features.

Key words: health care, YOLACT, fusion feature, Convolutional Neural Network (CNN), fall detection

中图分类号: