Denoising UWB Radar Data for Human Activity Recognition Using Convolutional Autoencoders
Human Activity Recognition (HAR) is one of the most popular research topics thanks to its usefulness in providing targeted, meaningful assistance to older adults. Because of the aging of the population in first-world countries, it becomes increasingly important to find innovative solutions that redu...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10197389/ |
| _version_ | 1797749635003973632 |
|---|---|
| author | Virgile Lafontaine Kevin Bouchard Julien Maitre Sebastien Gaboury |
| author_facet | Virgile Lafontaine Kevin Bouchard Julien Maitre Sebastien Gaboury |
| author_sort | Virgile Lafontaine |
| collection | DOAJ |
| description | Human Activity Recognition (HAR) is one of the most popular research topics thanks to its usefulness in providing targeted, meaningful assistance to older adults. Because of the aging of the population in first-world countries, it becomes increasingly important to find innovative solutions that reduce risks associated with aging-in-place policies. HAR proposes solutions that are based on Ambient Intelligence (AmI) to alleviate those risks. In this work, we exploited three UWB radars to recognize 14 activities performed by 19 participants in a prototype smart-home apartment. The main contribution of this paper is UWB radar data cleaning on a practical dataset. The UWB radar data has been filtered using an unsupervised deep convolutional autoencoder (CNN-AE) that learns background noise from the data. This filtering method is compared to the unfiltered data using a Convolutional Neural Network (CNN) classifier in a Leave-One-Subject-Out (LOSO) classification. Performances attest that the CNN-AE unsupervised filtering is efficient for HAR. In addition, we tested the generalization potential of this architecture when the dataset is comprised of a lower number of participants (1, 5, 10, and all 19 participants). Generalization in HAR is difficult as the results show the importance of data quantity and number of subjects. We obtained 69.9% top-1 accuracy when using our filtering architecture compared to 48.4% without it. To conclude, we show that an unsupervised CNN-AE can efficiently filter and generalize UWB radar data in a HAR setting while providing easier learning constraints and implementation on a practical dataset. |
| first_indexed | 2024-03-12T16:22:38Z |
| format | Article |
| id | doaj.art-94679d89210f4a69ae523daeddb45f72 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-03-12T16:22:38Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-94679d89210f4a69ae523daeddb45f722023-08-08T23:00:34ZengIEEEIEEE Access2169-35362023-01-0111812988130910.1109/ACCESS.2023.330022410197389Denoising UWB Radar Data for Human Activity Recognition Using Convolutional AutoencodersVirgile Lafontaine0https://orcid.org/0009-0004-9261-599XKevin Bouchard1https://orcid.org/0000-0002-5227-6602Julien Maitre2Sebastien Gaboury3Laboratoire d’Intelligence Ambiante pour la Reconnaissance d’Activités (LIARA), Université du Québec à Chicoutimi, Saguenay, CanadaLaboratoire d’Intelligence Ambiante pour la Reconnaissance d’Activités (LIARA), Université du Québec à Chicoutimi, Saguenay, CanadaLaboratoire d’Intelligence Ambiante pour la Reconnaissance d’Activités (LIARA), Université du Québec à Chicoutimi, Saguenay, CanadaLaboratoire d’Intelligence Ambiante pour la Reconnaissance d’Activités (LIARA), Université du Québec à Chicoutimi, Saguenay, CanadaHuman Activity Recognition (HAR) is one of the most popular research topics thanks to its usefulness in providing targeted, meaningful assistance to older adults. Because of the aging of the population in first-world countries, it becomes increasingly important to find innovative solutions that reduce risks associated with aging-in-place policies. HAR proposes solutions that are based on Ambient Intelligence (AmI) to alleviate those risks. In this work, we exploited three UWB radars to recognize 14 activities performed by 19 participants in a prototype smart-home apartment. The main contribution of this paper is UWB radar data cleaning on a practical dataset. The UWB radar data has been filtered using an unsupervised deep convolutional autoencoder (CNN-AE) that learns background noise from the data. This filtering method is compared to the unfiltered data using a Convolutional Neural Network (CNN) classifier in a Leave-One-Subject-Out (LOSO) classification. Performances attest that the CNN-AE unsupervised filtering is efficient for HAR. In addition, we tested the generalization potential of this architecture when the dataset is comprised of a lower number of participants (1, 5, 10, and all 19 participants). Generalization in HAR is difficult as the results show the importance of data quantity and number of subjects. We obtained 69.9% top-1 accuracy when using our filtering architecture compared to 48.4% without it. To conclude, we show that an unsupervised CNN-AE can efficiently filter and generalize UWB radar data in a HAR setting while providing easier learning constraints and implementation on a practical dataset.https://ieeexplore.ieee.org/document/10197389/Activity of daily livingdata filteringdata processingdeep learninghuman activity recognitionunsupervised learning |
| spellingShingle | Virgile Lafontaine Kevin Bouchard Julien Maitre Sebastien Gaboury Denoising UWB Radar Data for Human Activity Recognition Using Convolutional Autoencoders IEEE Access Activity of daily living data filtering data processing deep learning human activity recognition unsupervised learning |
| title | Denoising UWB Radar Data for Human Activity Recognition Using Convolutional Autoencoders |
| title_full | Denoising UWB Radar Data for Human Activity Recognition Using Convolutional Autoencoders |
| title_fullStr | Denoising UWB Radar Data for Human Activity Recognition Using Convolutional Autoencoders |
| title_full_unstemmed | Denoising UWB Radar Data for Human Activity Recognition Using Convolutional Autoencoders |
| title_short | Denoising UWB Radar Data for Human Activity Recognition Using Convolutional Autoencoders |
| title_sort | denoising uwb radar data for human activity recognition using convolutional autoencoders |
| topic | Activity of daily living data filtering data processing deep learning human activity recognition unsupervised learning |
| url | https://ieeexplore.ieee.org/document/10197389/ |
| work_keys_str_mv | AT virgilelafontaine denoisinguwbradardataforhumanactivityrecognitionusingconvolutionalautoencoders AT kevinbouchard denoisinguwbradardataforhumanactivityrecognitionusingconvolutionalautoencoders AT julienmaitre denoisinguwbradardataforhumanactivityrecognitionusingconvolutionalautoencoders AT sebastiengaboury denoisinguwbradardataforhumanactivityrecognitionusingconvolutionalautoencoders |