Corpus ID: 7213449

Evaluating the effects of signal segmentation on activity recognition

@inproceedings{Baos2014EvaluatingTE,
  title={Evaluating the effects of signal segmentation on activity recognition},
  author={Oresti Ba{\~n}os and Juan Manuel G{\'a}lvez and Miguel Damas and A. Guill{\'e}n and Luis Javier Herrera and H{\'e}ctor Pomares and Ignacio Rojas},
  booktitle={IWBBIO},
  year={2014}
}
On-body activity recognition systems are becoming more and more frequent in people's lives. These systems normally register body motion signals through small sensors that are placed on the user. To per- form the activity detection the signals must be adequately partitioned, however no clear consensus exists on how this should be done. More specically, considered the sliding window technique the most widely used approach for segmentation, it is unclear which window size must be applied. This… Expand

Figures and Topics from this paper

Investigation on the Effect of Different Window Size in Segmentation for Common Sport Activity
TLDR
It was found that 2.5 seconds window size represents the best trade-off in recognition of common sports activity, with an obtained accuracy above 90% and the preferably employed window size in detecting thecommon sports activity is determined. Expand
Adaptive sliding window segmentation for physical activity recognition using a single tri-axial accelerometer
TLDR
A novel approach to activity signal segmentation for physical activity recognition where the window size is adaptively adjusted according to the probability of the signal belongs to a particular activity to achieve the most effective segmentation. Expand
Hierarchical Signal Segmentation and Classification for Accurate Activity Recognition
TLDR
The paper presents the techniques employed in the team's (The Drifters) submission to the SHL recognition challenge, and proposes a hierarchical signal segmentation approach to deal with the problem of fixed-size windows. Expand
Human Activity Recognition with Convolutional Neural Networks
TLDR
This paper proposes to use Convolutional Neural Networks (CNNs) to classify human activities, and explores several combinations of activities and sensors, showing how motion signals can be adapted to be fed into CNNs by using different network architectures. Expand
Segmentation Study of Aged Gait Based on FFT
TLDR
The human action sequence segmentation method has been proposed based on the motion characteristics and can effectively segment the daily actions of the aged, extract valid action sequence, and is particularly effective in the segmentation of regular actions. Expand
A Preliminary Study on Hyperparameter Configuration for Human Activity Recognition
TLDR
Experimental results show that it is possible to maintain classification accuracy by adjusting hyperparameters, like window size and windows overlap factor, depending on user and activity performed, and motivate the development of a system able to automatically adapt hyperparameter settings for the activity performed by each user. Expand
A Study on Hyperparameter Configuration for Human Activity Recognition
TLDR
Experimental results show that it is possible to maintain classification accuracy by adjusting hyperparameters, like window size and window overlap, depending on the person and activity performed, and motivate the development of a system able to automatically adapt hyperparameter settings for the activity performed by each person. Expand
Vision-based human activity recognition using CNN
Human Activity Recognition (HAR) is a commonly discussed topic in computer vision. HAR implementations include representations such as health care and contact between the human and computer systems.Expand
Human Activity Recognition: A Comparison of Machine Learning Approaches
TLDR
These ML techniques are evaluated against classification performance metrics such as accuracy, precision, recall, F1 score, support, and run time on multiple HAR datasets to highlight the importance to tailor the selection of ML technique based on the specific HAR requirements and the characteristics of the associated HAR dataset. Expand
Multi-sensor fusion based on multiple classifier systems for human activity identification
TLDR
The study proposes a multi-view ensemble algorithm to integrate predicted values of different motion sensors to resolve major challenges in human activity identification and suggests a promising potential of hybrid feature selection approach, diversity-based multiple classifier systems to improve mobile and wearable sensor-based human activity detection and health monitoring system. Expand
...
1
2
...

References

SHOWING 1-10 OF 27 REFERENCES
Activity Recognition from User-Annotated Acceleration Data
TLDR
This is the first work to investigate performance of recognition algorithms with multiple, wire-free accelerometers on 20 activities using datasets annotated by the subjects themselves, and suggests that multiple accelerometers aid in recognition. Expand
Human activity recognition based on a sensor weighting hierarchical classifier
TLDR
This paper presents a fusion classification methodology which takes into account the potential of the individual decisions yielded at both activity and sensor classification levels, and systematically outperforms the results obtained by traditional multiclass models which otherwise may require a high-dimensional feature space to acquire a similar performance. Expand
Daily living activity recognition based on statistical feature quality group selection
TLDR
Satisfactory results are achieved in both laboratory and semi-naturalistic activity living datasets for real problems using several classification models, thus proving that any body sensor location can be suitable to define a simple one-feature-based recognition system, with particularly remarkable accuracy and applicability in the case of the wrist. Expand
Activity recognition and monitoring using multiple sensors on different body positions
TLDR
The design of an activity recognition and monitoring system based on the eWatch, multi-sensor platform worn on different body positions, is presented and the tradeoff between recognition accuracy and computational complexity is analyzed. Expand
Feature Selection and Activity Recognition from Wearable Sensors
TLDR
In this experiment, data was from 17 daily life examples from male and female subjects, and Interestingly, acceleration mean values from the necklace were selected as important features. Expand
Activity recognition using a single accelerometer placed at the wrist or ankle.
TLDR
A classification algorithm using 13 features shows good classification into the four classes given the complexity of the activities in the original data set, and could be implemented in real time on mobile devices with only 4-s latency. Expand
A Comparison of Feature Extraction Methods for the Classification of Dynamic Activities From Accelerometer Data
TLDR
The findings show that, although the wavelet transform approach can be used to characterize nonstationary signals, it does not perform as accurately as frequency-based features when classifying dynamic activities performed by healthy subjects. Expand
Ambulatory system for human motion analysis using a kinematic sensor: monitoring of daily physical activity in the elderly
TLDR
The ambulatory system showed a very high accuracy (> 99%) in identifying the 62 transfers or rolling out of bed, as well as 144 different posture changes to the back, ventral, right and left sides, in both first and second studies. Expand
A benchmark dataset to evaluate sensor displacement in activity recognition
TLDR
This work introduces an open benchmark dataset to investigate inertial sensor displacement effects in activity recognition, and introduces a concept of gradual sensor displacement conditions, including ideal, self-placement of a user, and mutual displacement deployments. Expand
Accelerometry: providing an integrated, practical method for long-term, ambulatory monitoring of human movement.
TLDR
An integrated approach is described in which a single, waist-mounted accelerometry system is used to monitor a range of different parameters of human movement in an unsupervised setting. Expand
...
1
2
3
...