A Combined One-Class SVM and Template-Matching Approach for User-Aided Human Fall Detection by Means of Floor Acoustic Features
The primary cause of injury-related death for the elders is represented by falls. The scientific community devoted them particular attention, since injuries can be limited by an early detection of the event. The solution proposed in this paper is based on a combined One-Class SVM (OCSVM) and templat...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Hindawi Limited
2017-01-01
|
Series: | Computational Intelligence and Neuroscience |
Online Access: | http://dx.doi.org/10.1155/2017/1512670 |
id |
doaj-a44ff57825e9432caf4b2f8e7313bb27 |
---|---|
record_format |
Article |
spelling |
doaj-a44ff57825e9432caf4b2f8e7313bb272020-11-24T22:46:31ZengHindawi LimitedComputational Intelligence and Neuroscience1687-52651687-52732017-01-01201710.1155/2017/15126701512670A Combined One-Class SVM and Template-Matching Approach for User-Aided Human Fall Detection by Means of Floor Acoustic FeaturesDiego Droghini0Daniele Ferretti1Emanuele Principi2Stefano Squartini3Francesco Piazza4Department of Information Engineering, Università Politecnica delle Marche, Via Brecce Bianche, 60131 Ancona, ItalyDepartment of Information Engineering, Università Politecnica delle Marche, Via Brecce Bianche, 60131 Ancona, ItalyDepartment of Information Engineering, Università Politecnica delle Marche, Via Brecce Bianche, 60131 Ancona, ItalyDepartment of Information Engineering, Università Politecnica delle Marche, Via Brecce Bianche, 60131 Ancona, ItalyDepartment of Information Engineering, Università Politecnica delle Marche, Via Brecce Bianche, 60131 Ancona, ItalyThe primary cause of injury-related death for the elders is represented by falls. The scientific community devoted them particular attention, since injuries can be limited by an early detection of the event. The solution proposed in this paper is based on a combined One-Class SVM (OCSVM) and template-matching classifier that discriminate human falls from nonfalls in a semisupervised framework. Acoustic signals are captured by means of a Floor Acoustic Sensor; then Mel-Frequency Cepstral Coefficients and Gaussian Mean Supervectors (GMSs) are extracted for the fall/nonfall discrimination. Here we propose a single-sensor two-stage user-aided approach: in the first stage, the OCSVM detects abnormal acoustic events. In the second, the template-matching classifier produces the final decision exploiting a set of template GMSs related to the events marked as false positives by the user. The performance of the algorithm has been evaluated on a corpus containing human falls and nonfall sounds. Compared to the OCSVM only approach, the proposed algorithm improves the performance by 10.14% in clean conditions and 4.84% in noisy conditions. Compared to Popescu and Mahnot (2009) the performance improvement is 19.96% in clean conditions and 8.08% in noisy conditions.http://dx.doi.org/10.1155/2017/1512670 |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Diego Droghini Daniele Ferretti Emanuele Principi Stefano Squartini Francesco Piazza |
spellingShingle |
Diego Droghini Daniele Ferretti Emanuele Principi Stefano Squartini Francesco Piazza A Combined One-Class SVM and Template-Matching Approach for User-Aided Human Fall Detection by Means of Floor Acoustic Features Computational Intelligence and Neuroscience |
author_facet |
Diego Droghini Daniele Ferretti Emanuele Principi Stefano Squartini Francesco Piazza |
author_sort |
Diego Droghini |
title |
A Combined One-Class SVM and Template-Matching Approach for User-Aided Human Fall Detection by Means of Floor Acoustic Features |
title_short |
A Combined One-Class SVM and Template-Matching Approach for User-Aided Human Fall Detection by Means of Floor Acoustic Features |
title_full |
A Combined One-Class SVM and Template-Matching Approach for User-Aided Human Fall Detection by Means of Floor Acoustic Features |
title_fullStr |
A Combined One-Class SVM and Template-Matching Approach for User-Aided Human Fall Detection by Means of Floor Acoustic Features |
title_full_unstemmed |
A Combined One-Class SVM and Template-Matching Approach for User-Aided Human Fall Detection by Means of Floor Acoustic Features |
title_sort |
combined one-class svm and template-matching approach for user-aided human fall detection by means of floor acoustic features |
publisher |
Hindawi Limited |
series |
Computational Intelligence and Neuroscience |
issn |
1687-5265 1687-5273 |
publishDate |
2017-01-01 |
description |
The primary cause of injury-related death for the elders is represented by falls. The scientific community devoted them particular attention, since injuries can be limited by an early detection of the event. The solution proposed in this paper is based on a combined One-Class SVM (OCSVM) and template-matching classifier that discriminate human falls from nonfalls in a semisupervised framework. Acoustic signals are captured by means of a Floor Acoustic Sensor; then Mel-Frequency Cepstral Coefficients and Gaussian Mean Supervectors (GMSs) are extracted for the fall/nonfall discrimination. Here we propose a single-sensor two-stage user-aided approach: in the first stage, the OCSVM detects abnormal acoustic events. In the second, the template-matching classifier produces the final decision exploiting a set of template GMSs related to the events marked as false positives by the user. The performance of the algorithm has been evaluated on a corpus containing human falls and nonfall sounds. Compared to the OCSVM only approach, the proposed algorithm improves the performance by 10.14% in clean conditions and 4.84% in noisy conditions. Compared to Popescu and Mahnot (2009) the performance improvement is 19.96% in clean conditions and 8.08% in noisy conditions. |
url |
http://dx.doi.org/10.1155/2017/1512670 |
work_keys_str_mv |
AT diegodroghini acombinedoneclasssvmandtemplatematchingapproachforuseraidedhumanfalldetectionbymeansofflooracousticfeatures AT danieleferretti acombinedoneclasssvmandtemplatematchingapproachforuseraidedhumanfalldetectionbymeansofflooracousticfeatures AT emanueleprincipi acombinedoneclasssvmandtemplatematchingapproachforuseraidedhumanfalldetectionbymeansofflooracousticfeatures AT stefanosquartini acombinedoneclasssvmandtemplatematchingapproachforuseraidedhumanfalldetectionbymeansofflooracousticfeatures AT francescopiazza acombinedoneclasssvmandtemplatematchingapproachforuseraidedhumanfalldetectionbymeansofflooracousticfeatures AT diegodroghini combinedoneclasssvmandtemplatematchingapproachforuseraidedhumanfalldetectionbymeansofflooracousticfeatures AT danieleferretti combinedoneclasssvmandtemplatematchingapproachforuseraidedhumanfalldetectionbymeansofflooracousticfeatures AT emanueleprincipi combinedoneclasssvmandtemplatematchingapproachforuseraidedhumanfalldetectionbymeansofflooracousticfeatures AT stefanosquartini combinedoneclasssvmandtemplatematchingapproachforuseraidedhumanfalldetectionbymeansofflooracousticfeatures AT francescopiazza combinedoneclasssvmandtemplatematchingapproachforuseraidedhumanfalldetectionbymeansofflooracousticfeatures |
_version_ |
1725685054101782528 |