|
Title:
|
VIDEO ACTION RECOGNITION BASED ON HIDDEN MARKOV MODEL COMBINED WITH PARTICLE SWARM |
|
Author(s):
|
Haiyi Zhang, Yang Yi, XiaoXing Li |
|
ISBN:
|
978-972-8939-68-7 |
|
Editors:
|
Miguel Baptista Nunes, Pedro Isaías and Philip Powell |
|
Year:
|
2012 |
|
Edition:
|
Single |
|
Keywords:
|
Action recognition, behavior modeling, HMM (Hidden Markov Model), event probability sequence, PSO (Particle Swarm Optimization) |
|
Type:
|
Full Paper |
|
First Page:
|
133 |
|
Last Page:
|
140 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
A novel learning algorithm for a Hidden Markov Model (HMM) based on Particle Swarm Optimization (PSO) is addressed, and a new video action recognition approach is proposed. First, we extract the features of the action's trajectories from each of the target objects. We show them with the semantic event probability produced in the HMM. Secondly, we improve the HMM by modifying its parameters learning algorithm based on PSO, which has the advantage of changing the computational learning level of the HMM from a result that is locally optimal to one that is globally optimal, meanwhile, it can avoid the common computational errors associated with data overflow. Finally, we identify the objective activity' patterns using the Time Warping Method by matching event probability sequence. Practical data experiments show that the presented algorithm is efficient by reflecting the real activities and can influence how the problem is attacked. The comparative experiments also show that it has advantages over the Baum-Welch algorithm and some other famous methods. |
|
|
|
|
|
|
