A HMM-Based Structured Model for Human Behavior

DOI

10.2991/fmsmt-17.2017.311How to use a DOI?

Keywords

Behavior Recognition, Model of Behavior, Human Behavioral Analysis and Understanding

Abstract

A structured recognition method of human behavior based on HMM is proposed to build the human action in this paper, only one camera which does not need to be adjust should be used in this approach. The robustness of the human behavior recognition model which is established by the free view angle of the camera is improved. The human behavior outline feature is segmented in video frame, some algorithm is used to establish the structure rob posture model and then to be standardized, at the last, the standardized rob postures of the same behavior are used to build the human body hidden Markov model (HMM) which has the independent position, independent translation, independent scale, independent rotating and the advantages of the free view angle, this model is verified by experiment has higher recognition rate.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Weihua Wang
PY  - 2017/04
DA  - 2017/04
TI  - A HMM-Based Structured Model for Human Behavior
BT  - Proceedings of the 2017 5th International Conference on Frontiers of Manufacturing Science and Measuring Technology (FMSMT 2017)
PB  - Atlantis Press
SP  - 1599
EP  - 1602
SN  - 2352-5401
UR  - https://doi.org/10.2991/fmsmt-17.2017.311
DO  - 10.2991/fmsmt-17.2017.311
ID  - Wang2017/04
ER  -