Machine learning based canine posture estimation using inertial data
The aim of this study was to design a new canine posture estimation system specifically for working dogs. The system was composed of Inertial Measurement Units (IMUs) that are commercially available, and a supervised learning algorithm which was developed for different behaviours. Three IMUs, each c...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2023-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10284380/?tool=EBI |
| _version_ | 1797795968120258560 |
|---|---|
| author | Marinara Marcato Salvatore Tedesco Conor O’Mahony Brendan O’Flynn Paul Galvin |
| author_facet | Marinara Marcato Salvatore Tedesco Conor O’Mahony Brendan O’Flynn Paul Galvin |
| author_sort | Marinara Marcato |
| collection | DOAJ |
| description | The aim of this study was to design a new canine posture estimation system specifically for working dogs. The system was composed of Inertial Measurement Units (IMUs) that are commercially available, and a supervised learning algorithm which was developed for different behaviours. Three IMUs, each containing a 3-axis accelerometer, gyroscope, and magnetometer, were attached to the dogs’ chest, back, and neck. To build and test the model, data were collected during a video-recorded behaviour test where the trainee assistance dogs performed static postures (standing, sitting, lying down) and dynamic activities (walking, body shake). Advanced feature extraction techniques were employed for the first time in this field, including statistical, temporal, and spectral methods. The most important features for posture prediction were chosen using Select K Best with ANOVA F-value. The individual contributions of each IMU, sensor, and feature type were analysed using Select K Best scores and Random Forest feature importance. Results showed that the back and chest IMUs were more important than the neck IMU, and the accelerometers were more important than the gyroscopes. The addition of IMUs to the chest and back of dog harnesses is recommended to improve performance. Additionally, statistical and temporal feature domains were more important than spectral feature domains. Three novel cascade arrangements of Random Forest and Isolation Forest were fitted to the dataset. The best classifier achieved an f1-macro of 0.83 and an f1-weighted of 0.90 for the prediction of the five postures, demonstrating a better performance than previous studies. These results were attributed to the data collection methodology (number of subjects and observations, multiple IMUs, use of common working dog breeds) and novel machine learning techniques (advanced feature extraction, feature selection and modelling arrangements) employed. The dataset and code used are publicly available on Mendeley Data and GitHub, respectively. |
| first_indexed | 2024-03-13T03:26:03Z |
| format | Article |
| id | doaj.art-2fb92dbad04b442490ac0ecd1c71b2c5 |
| institution | Directory Open Access Journal |
| issn | 1932-6203 |
| language | English |
| last_indexed | 2024-03-13T03:26:03Z |
| publishDate | 2023-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj.art-2fb92dbad04b442490ac0ecd1c71b2c52023-06-25T05:31:12ZengPublic Library of Science (PLoS)PLoS ONE1932-62032023-01-01186Machine learning based canine posture estimation using inertial dataMarinara MarcatoSalvatore TedescoConor O’MahonyBrendan O’FlynnPaul GalvinThe aim of this study was to design a new canine posture estimation system specifically for working dogs. The system was composed of Inertial Measurement Units (IMUs) that are commercially available, and a supervised learning algorithm which was developed for different behaviours. Three IMUs, each containing a 3-axis accelerometer, gyroscope, and magnetometer, were attached to the dogs’ chest, back, and neck. To build and test the model, data were collected during a video-recorded behaviour test where the trainee assistance dogs performed static postures (standing, sitting, lying down) and dynamic activities (walking, body shake). Advanced feature extraction techniques were employed for the first time in this field, including statistical, temporal, and spectral methods. The most important features for posture prediction were chosen using Select K Best with ANOVA F-value. The individual contributions of each IMU, sensor, and feature type were analysed using Select K Best scores and Random Forest feature importance. Results showed that the back and chest IMUs were more important than the neck IMU, and the accelerometers were more important than the gyroscopes. The addition of IMUs to the chest and back of dog harnesses is recommended to improve performance. Additionally, statistical and temporal feature domains were more important than spectral feature domains. Three novel cascade arrangements of Random Forest and Isolation Forest were fitted to the dataset. The best classifier achieved an f1-macro of 0.83 and an f1-weighted of 0.90 for the prediction of the five postures, demonstrating a better performance than previous studies. These results were attributed to the data collection methodology (number of subjects and observations, multiple IMUs, use of common working dog breeds) and novel machine learning techniques (advanced feature extraction, feature selection and modelling arrangements) employed. The dataset and code used are publicly available on Mendeley Data and GitHub, respectively.https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10284380/?tool=EBI |
| spellingShingle | Marinara Marcato Salvatore Tedesco Conor O’Mahony Brendan O’Flynn Paul Galvin Machine learning based canine posture estimation using inertial data PLoS ONE |
| title | Machine learning based canine posture estimation using inertial data |
| title_full | Machine learning based canine posture estimation using inertial data |
| title_fullStr | Machine learning based canine posture estimation using inertial data |
| title_full_unstemmed | Machine learning based canine posture estimation using inertial data |
| title_short | Machine learning based canine posture estimation using inertial data |
| title_sort | machine learning based canine posture estimation using inertial data |
| url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10284380/?tool=EBI |
| work_keys_str_mv | AT marinaramarcato machinelearningbasedcaninepostureestimationusinginertialdata AT salvatoretedesco machinelearningbasedcaninepostureestimationusinginertialdata AT conoromahony machinelearningbasedcaninepostureestimationusinginertialdata AT brendanoflynn machinelearningbasedcaninepostureestimationusinginertialdata AT paulgalvin machinelearningbasedcaninepostureestimationusinginertialdata |