An Improved Feature-based Method to Fall Detection by Fusing Machine Learning Technology
Before the time of the popularization of deep learning technology, a lot of handcrafted features emerged to solve pattern recognition problems in the field of fall detection.This paper uses machine learning technology,Support Vector Machine(SVM), to improve the traditional feature-based methods.Our method provides two major improvements.First, the existed classic features, which were always indiscreetly abandoned when they meet deep learning methods, were adopted and together with machine learning technology form an improved and efficient fall detection method.Second, the definition of a threshold which needs massive experiments was now learned by the program itself.Compare with the current popular end-to-end deep learning methods, the improved feature-based method fusing machine learning technology shows great advantages in time efficiency because of the significant reduce of the input parameters.Additionally, with the help of SVM, the thresholds need no manual definitions, which saves a lot of time and makes it more precise.Our approach is evaluated on a public dataset, TST fall detection dataset v2.The results show that our approach achieves an accuracy of 93.56%,which is better than other typical methods.Furthermore, the approach can be used in real-time video surveillance because of its time efficiency and robustness.
fall detection support vector machine SVM computer vision handcrafted feature feature-based method
Leiyue Yao Wei Yang
The Center of Collaboration and Innovation Jiangxi University of Technology Nan Chang,China
国际会议
郑州
英文
456-462
2018-09-21（万方平台首次上网日期，不代表论文的发表时间）