Ambulatory activity classification with dendogram-based support vector machine: Application in lower-limb active exoskeleton

Mazumder, Oishee; Kundu, Ananda Sankar; Lenka, Prasanna Kumar; Bhaumik, Subhasis

doi:10.1016/j.gaitpost.2016.08.010

Cited by 14 publications

(12 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, they suffer from the requirement of a large amount of data for training. Discriminative approaches, such as decision tree [22], K-nearest neighbour [23], random forest [24], support vector machine [25] and artificial neural network [26], learn the features mappings to activity labels by creating the decision boundaries in the feature space. For example, decision tree classification models have been successfully adopted for separating static activities from dynamic activities [27].…”

Section: Iirelated Workmentioning

confidence: 99%

Design and assessment of the data analysis process for a wrist-worn smart object to detect atomic activities in the smart home

Cleland

Nugent

et al. 2019

Pervasive and Mobile Computing

View full text Add to dashboard Cite

The ability to accurately identify the different activities of daily living (ADLs) is considered as one of the basis to foster new technological solutions inside the smart home. Current ADL recognition proposals, still however, struggle to accurately and robustly identify the range of different activities that can be performed at home, namely static, dynamic and transient activities, and the high variety of technologies and data analysis possibilities to classify the information gathered by the sensors. In this paper, we describe the methodological approach that we have followed for the processing, analysis and classification of data obtained by a simple and non-intrusive smart object with the objective to detect atomic (i.e. non-divisible) activities inside the smart home. The smart object consists of a wrist-worn 3D accelerometer, which presents as its advantages its customizability and usability. We have performed a set of systematic experiments involving ten people and have followed the steps from data gathering to the comparison of different classification techniques, to find out that it is possible to select a complete succession of data processing steps in order to detect, with high accuracy, a set of atomic activities of daily life with the selected smart object, which performs well with different independent datasets besides ours.

show abstract

Section: Iirelated Workmentioning

confidence: 99%

Design and assessment of the data analysis process for a wrist-worn smart object to detect atomic activities in the smart home

Cleland

Nugent

et al. 2019

Pervasive and Mobile Computing

View full text Add to dashboard Cite

show abstract

“…However, they suffer from the requirement of a large amount of data for training. Discriminative approaches, such as decision tree (DT) [16], K-nearest neighbor [17], conditional random fields [18], support vector machine (SVM) [19] and artificial neural network [20], learn the features mappings to activity labels by creating the decision boundaries in the feature space. For example, DT classification models have been successfully adopted for separating static activities from dynamic activities [21].…”

Section: Related Workmentioning

confidence: 99%

A Heterogeneous Ensemble Approach for Activity Recognition with Integration of Change Point-Based Data Segmentation

Zhang

2018

Applied Sciences

View full text Add to dashboard Cite

One of the main topics of Smart Home (SH) research is the recognition of activities performed by its inhabitants, which is considered to be one of the bases to foster new technological solutions inside the home, including services to prolong independent living of the elderly. However, current activity recognition proposals still find problems when considering all the different types of activities that can be performed at home, namely static, dynamic, and transitional activities. In this paper, we consider recognition of transitional activities, which is often ignored in most studies. In addition, we propose a novel dynamic segmentation method based on change points in data stream and construct an ensemble of heterogeneous classifiers to recognize twelve activities (of all types). The experiment is conducted on the dataset collected over ten hours by a wearable accelerometer placed on the person’s wrist. The base classifiers selected to form this ensemble are support vector machine (SVM), decision tree (DT) and k-nearest neighbors (KNN). As a result, the proposed approach has achieved an overall classification accuracy equal to 96.87% with 10-fold cross-validation. Moreover, all activity types considered have been similarly well identified.

show abstract

“…Ambulatory activity analysis is fundamental for the study of human movement because it can provide rich information for gait analysis [ 1 , 2 , 3 ]. In addition, ambulatory activity analysis is an important tool for fall detection and post-stroke rehabilitation and, thus, can contribute to enhancing the quality of life, especially for the elder population [ 4 , 5 ].…”

Section: Introductionmentioning

confidence: 99%

Adaptive Accumulation of Plantar Pressure for Ambulatory Activity Recognition and Pedestrian Identification

Truong

You

et al. 2021

Sensors

View full text Add to dashboard Cite

In this paper, we propose a novel method for ambulatory activity recognition and pedestrian identification based on temporally adaptive weighting accumulation-based features extracted from categorical plantar pressure. The method relies on three pressure-related features, which are calculated by accumulating the pressure of the standing foot in each step over three different temporal weighting forms. In addition, we consider a feature reflecting the pressure variation. These four features characterize the standing posture in a step by differently weighting step pressure data over time. We use these features to analyze the standing foot during walking and then recognize ambulatory activities and identify pedestrians based on multilayer multiclass support vector machine classifiers. Experimental results show that the proposed method achieves 97% accuracy for the two tasks when analyzing eight consecutive steps. For faster processing, the method reaches 89.9% and 91.3% accuracy for ambulatory activity recognition and pedestrian identification considering two consecutive steps, respectively, whereas the accuracy drops to 83.3% and 82.3% when considering one step for the respective tasks. Comparative results demonstrated the high performance of the proposed method regarding accuracy and temporal sensitivity.

show abstract

Ambulatory activity classification with dendogram-based support vector machine: Application in lower-limb active exoskeleton

Cited by 14 publications

References 15 publications

Design and assessment of the data analysis process for a wrist-worn smart object to detect atomic activities in the smart home

Design and assessment of the data analysis process for a wrist-worn smart object to detect atomic activities in the smart home

A Heterogeneous Ensemble Approach for Activity Recognition with Integration of Change Point-Based Data Segmentation

Adaptive Accumulation of Plantar Pressure for Ambulatory Activity Recognition and Pedestrian Identification

Contact Info

Product

Resources

About