Advanced techniques for human activity classification
Autor(es) y otros:
Director(es):
Palabra(s) clave:
Imu
Hmm
Markov
Mtx
Walk
Fecha de publicación:
Serie:
Máster Universitario en Soft Computing y Análisis Inteligente de Datos
Resumen:
The purpose of this work is to find out whether a system based on a conventional Hidden Markov Model (HMM) is able to detect the action that the person is carrying out, and be able to differentiate it from other similar actions. Thus, at some point in the future, a system can be developed that checks the data read from the sensors, compares it with some different models that represent a bunch of different actions, and find out if any of them correspond to what the person is actually doing.
The purpose of this work is to find out whether a system based on a conventional Hidden Markov Model (HMM) is able to detect the action that the person is carrying out, and be able to differentiate it from other similar actions. Thus, at some point in the future, a system can be developed that checks the data read from the sensors, compares it with some different models that represent a bunch of different actions, and find out if any of them correspond to what the person is actually doing.
Colecciones
- Trabajos Fin de Máster [5253]