scholarly journals Recognizing human activities from smartphone sensors using hierarchical continuous hidden Markov models

2017 ◽  
Vol 13 (1) ◽  
pp. 155014771668368 ◽  
Author(s):  
Charissa Ann Ronao ◽  
Sung-Bae Cho
Keyword(s):  
Hierarchical Structure ◽  
Markov Models ◽  
Hidden Markov ◽  
Classification Performance ◽  
Fine Tuning ◽  
Sensor Data ◽  
Model Framework ◽  
The Hierarchical Structure ◽  
Smartphone Sensors ◽  
And Training

Human activity recognition has been gaining more and more attention from researchers in recent years, particularly with the use of widespread and commercially available devices such as smartphones. However, most of the existing works focus on discriminative classifiers while neglecting the inherent time-series and continuous characteristics of sensor data. To address this, we propose a two-stage continuous hidden Markov model framework, which also takes advantage of the innate hierarchical structure of basic activities. This kind of system architecture not only enables the use of different feature subsets on different subclasses, which effectively reduces feature computation overhead, but also allows for varying number of states and iterations. Experiments show that the hierarchical structure dramatically increases classification performance. We analyze the behavior of the accelerometer and gyroscope signals for each activity through graphs, and with added fine tuning of states and training iterations, the proposed method is able to achieve an overall accuracy of up to 93.18%, which is the best performance among the state-of-the-art classifiers for the problem at hand.

scholarly journals Hidden Markov models identify major movement modes in accelerometer and magnetometer data from four albatross species

2021 ◽  
Vol 9 (1) ◽  
Author(s):  
Melinda G. Conners ◽  
Théo Michelot ◽  
Eleanor I. Heywood ◽  
Rachael A. Orben ◽  
Richard A. Phillips ◽  
...  
Keyword(s):  
Hidden Markov Models ◽  
Markov Models ◽  
Hidden Markov ◽  
Flapping Flight ◽  
Sensor Data ◽  
Accelerometer Data ◽  
Fine Scale ◽  
Magnetometer Data ◽  
Behavioral Classification ◽  
Soaring Flight

AbstractBackgroundInertial measurement units (IMUs) with high-resolution sensors such as accelerometers are now used extensively to study fine-scale behavior in a wide range of marine and terrestrial animals. Robust and practical methods are required for the computationally-demanding analysis of the resulting large datasets, particularly for automating classification routines that construct behavioral time series and time-activity budgets. Magnetometers are used increasingly to study behavior, but it is not clear how these sensors contribute to the accuracy of behavioral classification methods. Development of effective  classification methodology is key to understanding energetic and life-history implications of foraging and other behaviors.MethodsWe deployed accelerometers and magnetometers on four species of free-ranging albatrosses and evaluated the ability of unsupervised hidden Markov models (HMMs) to identify three major modalities in their behavior: ‘flapping flight’, ‘soaring flight’, and ‘on-water’. The relative contribution of each sensor to classification accuracy was measured by comparing HMM-inferred states with expert classifications identified from stereotypic patterns observed in sensor data.ResultsHMMs provided a flexible and easily interpretable means of classifying behavior from sensor data. Model accuracy was high overall (92%), but varied across behavioral states (87.6, 93.1 and 91.7% for ‘flapping flight’, ‘soaring flight’ and ‘on-water’, respectively). Models built on accelerometer data alone were as accurate as those that also included magnetometer data; however, the latter were useful for investigating slow and periodic behaviors such as dynamic soaring at a fine scale.ConclusionsThe use of IMUs in behavioral studies produces large data sets, necessitating the development of computationally-efficient methods to automate behavioral classification in order to synthesize and interpret underlying patterns. HMMs provide an accessible and robust framework for analyzing complex IMU datasets and comparing behavioral variation among taxa across habitats, time and space.

scholarly journals Probabilistic Independence Networks for Hidden Markov Probability Models

1997 ◽  
Vol 9 (2) ◽  
pp. 227-269 ◽  
Author(s):  
Padhraic Smyth ◽  
David Heckerman ◽  
Michael I. Jordan
Keyword(s):  
Speech Recognition ◽  
Graphical Models ◽  
Statistical Physics ◽  
Graphical Model ◽  
Markov Models ◽  
Hidden Markov ◽  
Probability Models ◽  
Model Framework ◽  
Estimation Algorithms ◽  
Special Cases

Graphical techniques for modeling the dependencies of random variables have been explored in a variety of different areas, including statistics, statistical physics, artificial intelligence, speech recognition, image processing, and genetics. Formalisms for manipulating these models have been developed relatively independently in these research communities. In this paper we explore hidden Markov models (HMMs) and related structures within the general framework of probabilistic independence networks (PINs). The paper presents a self-contained review of the basic principles of PINs. It is shown that the well-known forward-backward (F-B) and Viterbi algorithms for HMMs are special cases of more general inference algorithms for arbitrary PINs. Furthermore, the existence of inference and estimation algorithms for more general graphical models provides a set of analysis tools for HMM practitioners who wish to explore a richer class of HMM structures. Examples of relatively complex models to handle sensor fusion and coarticulation in speech recognition are introduced and treated within the graphical model framework to illustrate the advantages of the general approach.

Machinery Faults Detection and Forecasting Using Hidden Markov Models

2006 ◽  
Author(s):  
Paolo Calefati ◽  
Biagio Amico ◽  
Antonella Lacasella ◽  
Emanuel Muraca ◽  
Ming J. Zuo
Keyword(s):  
Data Acquisition ◽  
Hidden Markov Models ◽  
Viterbi Algorithm ◽  
Markov Models ◽  
Hidden Markov ◽  
Experimental Results ◽  
Sensor Data ◽  
Residual Lifetime ◽  
Multiple Sensor ◽  
Useful Lifetime

The present work describes an automatic procedure for diagnostics and prognostic issues, and its application to the evaluation of gearboxes residual lifetime. The Hidden Markov Models — HMM — technique has been used to create quasistationary and stationary models and to take advantages of the multiple sensor data acquisition architecture. At first, Markov models for diagnostics have been defined. The main advantage of the HMMs approach is that all vibration raw data measured by a multisensor architecture can be used without any preprocessing. An effort to adapt the HMMs technique to the prognostic issue has also been carried out. To create Markov Models suitable for prognostics, the Viterbi Algorithm has been used to define the best sequence of model states and to optimize residual useful lifetime computation. Finally, experimental results are discussed, which encourage further research efforts according to the proposed approach.

scholarly journals Detection of Ships at Mooring Dolphins with Hidden Markov Models

2019 ◽  
Vol 2673 (4) ◽  
pp. 439-447
Author(s):  
Maurits Waterbolk ◽  
Jasper Tump ◽  
Rianne Klaver ◽  
Rosalie van der Woude ◽  
Daniel Velleman ◽  
...  
Keyword(s):  
Evaluation Method ◽  
Markov Models ◽  
Hidden Markov ◽  
Economic Value ◽  
Descriptive Statistics ◽  
Sensor Data ◽  
Automatic Identification ◽  
Identification System ◽  
Value Range ◽  
Lock In

IJpalen near the lock in IJmuiden are of great economic value to the Port of Amsterdam. These mooring dolphins have to endure a considerable amount of kinetic forces which can have an impact on the condition of the dolphins. These forces are created by either mooring or already moored ships. Any irregularities taking place at the IJpalen can have disastrous results unless timely addressed. The Port of Amsterdam has attached sensors to the poles and the plates, which measure changes in the dimensions regarding the dolphins. This report explores whether combining sensor data from the IJpalen and automatic identification system (AIS) data can produce beneficial insights into the dolphins’ states. We have used the sensor dataset to build a hidden Markov model (HMM) which predicts whether a ship is moored. We evaluated these results using the AIS data, in which can be discovered when a ship was moored at the IJpalen, producing remarkable results. We analyzed the sensor values using descriptive statistics to discover the normal and problem values. This research has obtained the following findings. First, descriptive statistics indicate a normal value range for the sensor values. Whenever a value out of this range is observed, it could be a problem case. Finally, it is possible to detect whether a ship is moored in the sensor data. An HMM on the z-angle of the plate of the east dolphin produces the best prediction, i.e., the highest accuracy of 90.2% according to the evaluation method, of a moored ship at the IJpalen.

Sign in / Sign up

Export Citation Format

Share Document