Classifying Goliath Grouper (<i>Epinephelus itajara</i>) Behaviors from a Novel, Multi-Sensor Tag
Inertial measurement unit sensors (IMU; i.e., accelerometer, gyroscope and magnetometer combinations) are frequently fitted to animals to better understand their activity patterns and energy expenditure. Capable of recording hundreds of data points a second, these sensors can quickly produce large d...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-09-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/21/19/6392 |
| _version_ | 1797515798406758400 |
|---|---|
| author | Lauran R. Brewster Ali K. Ibrahim Breanna C. DeGroot Thomas J. Ostendorf Hanqi Zhuang Laurent M. Chérubin Matthew J. Ajemian |
| author_facet | Lauran R. Brewster Ali K. Ibrahim Breanna C. DeGroot Thomas J. Ostendorf Hanqi Zhuang Laurent M. Chérubin Matthew J. Ajemian |
| author_sort | Lauran R. Brewster |
| collection | DOAJ |
| description | Inertial measurement unit sensors (IMU; i.e., accelerometer, gyroscope and magnetometer combinations) are frequently fitted to animals to better understand their activity patterns and energy expenditure. Capable of recording hundreds of data points a second, these sensors can quickly produce large datasets that require methods to automate behavioral classification. Here, we describe behaviors derived from a custom-built multi-sensor bio-logging tag attached to Atlantic Goliath grouper (<i>Epinephelus itajara</i>) within a simulated ecosystem. We then compared the performance of two commonly applied machine learning approaches (random forest and support vector machine) to a deep learning approach (convolutional neural network, or CNN) for classifying IMU data from this tag. CNNs are frequently used to recognize activities from IMU data obtained from humans but are less commonly considered for other animals. Thirteen behavioral classes were identified during ethogram development, nine of which were classified. For the conventional machine learning approaches, 187 summary statistics were extracted from the data, including time and frequency domain features. The CNN was fed absolute values obtained from fast Fourier transformations of the raw tri-axial accelerometer, gyroscope and magnetometer channels, with a frequency resolution of 512 data points. Five metrics were used to assess classifier performance; the deep learning approach performed better across all metrics (Sensitivity = 0.962; Specificity = 0.996; <i>F</i><sub>1</sub>-score = 0.962; Matthew’s Correlation Coefficient = 0.959; Cohen’s Kappa = 0.833) than both conventional machine learning approaches. Generally, the random forest performed better than the support vector machine. In some instances, a conventional learning approach yielded a higher performance metric for particular classes (e.g., the random forest had a <i>F</i><sub>1</sub>-score of 0.971 for backward swimming compared to 0.955 for the CNN). Deep learning approaches could potentially improve behavioral classification from IMU data, beyond that obtained from conventional machine learning methods. |
| first_indexed | 2024-03-10T06:52:22Z |
| format | Article |
| id | doaj.art-adc95225860046c89698763eeb5a5fcc |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T06:52:22Z |
| publishDate | 2021-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-adc95225860046c89698763eeb5a5fcc2023-11-22T16:45:23ZengMDPI AGSensors1424-82202021-09-012119639210.3390/s21196392Classifying Goliath Grouper (<i>Epinephelus itajara</i>) Behaviors from a Novel, Multi-Sensor TagLauran R. Brewster0Ali K. Ibrahim1Breanna C. DeGroot2Thomas J. Ostendorf3Hanqi Zhuang4Laurent M. Chérubin5Matthew J. Ajemian6Harbor Branch Oceanographic Institute, Florida Atlantic University, Fort Pierce, FL 34946, USAHarbor Branch Oceanographic Institute, Florida Atlantic University, Fort Pierce, FL 34946, USAHarbor Branch Oceanographic Institute, Florida Atlantic University, Fort Pierce, FL 34946, USAHarbor Branch Oceanographic Institute, Florida Atlantic University, Fort Pierce, FL 34946, USADepartment of Electrical Engineering and Computer Science, Florida Atlantic University, Boca Raton, FL 33431, USAHarbor Branch Oceanographic Institute, Florida Atlantic University, Fort Pierce, FL 34946, USAHarbor Branch Oceanographic Institute, Florida Atlantic University, Fort Pierce, FL 34946, USAInertial measurement unit sensors (IMU; i.e., accelerometer, gyroscope and magnetometer combinations) are frequently fitted to animals to better understand their activity patterns and energy expenditure. Capable of recording hundreds of data points a second, these sensors can quickly produce large datasets that require methods to automate behavioral classification. Here, we describe behaviors derived from a custom-built multi-sensor bio-logging tag attached to Atlantic Goliath grouper (<i>Epinephelus itajara</i>) within a simulated ecosystem. We then compared the performance of two commonly applied machine learning approaches (random forest and support vector machine) to a deep learning approach (convolutional neural network, or CNN) for classifying IMU data from this tag. CNNs are frequently used to recognize activities from IMU data obtained from humans but are less commonly considered for other animals. Thirteen behavioral classes were identified during ethogram development, nine of which were classified. For the conventional machine learning approaches, 187 summary statistics were extracted from the data, including time and frequency domain features. The CNN was fed absolute values obtained from fast Fourier transformations of the raw tri-axial accelerometer, gyroscope and magnetometer channels, with a frequency resolution of 512 data points. Five metrics were used to assess classifier performance; the deep learning approach performed better across all metrics (Sensitivity = 0.962; Specificity = 0.996; <i>F</i><sub>1</sub>-score = 0.962; Matthew’s Correlation Coefficient = 0.959; Cohen’s Kappa = 0.833) than both conventional machine learning approaches. Generally, the random forest performed better than the support vector machine. In some instances, a conventional learning approach yielded a higher performance metric for particular classes (e.g., the random forest had a <i>F</i><sub>1</sub>-score of 0.971 for backward swimming compared to 0.955 for the CNN). Deep learning approaches could potentially improve behavioral classification from IMU data, beyond that obtained from conventional machine learning methods.https://www.mdpi.com/1424-8220/21/19/6392accelerometermagnetometergyroscopeclassificationrandom forestsupport vector machine |
| spellingShingle | Lauran R. Brewster Ali K. Ibrahim Breanna C. DeGroot Thomas J. Ostendorf Hanqi Zhuang Laurent M. Chérubin Matthew J. Ajemian Classifying Goliath Grouper (<i>Epinephelus itajara</i>) Behaviors from a Novel, Multi-Sensor Tag Sensors accelerometer magnetometer gyroscope classification random forest support vector machine |
| title | Classifying Goliath Grouper (<i>Epinephelus itajara</i>) Behaviors from a Novel, Multi-Sensor Tag |
| title_full | Classifying Goliath Grouper (<i>Epinephelus itajara</i>) Behaviors from a Novel, Multi-Sensor Tag |
| title_fullStr | Classifying Goliath Grouper (<i>Epinephelus itajara</i>) Behaviors from a Novel, Multi-Sensor Tag |
| title_full_unstemmed | Classifying Goliath Grouper (<i>Epinephelus itajara</i>) Behaviors from a Novel, Multi-Sensor Tag |
| title_short | Classifying Goliath Grouper (<i>Epinephelus itajara</i>) Behaviors from a Novel, Multi-Sensor Tag |
| title_sort | classifying goliath grouper i epinephelus itajara i behaviors from a novel multi sensor tag |
| topic | accelerometer magnetometer gyroscope classification random forest support vector machine |
| url | https://www.mdpi.com/1424-8220/21/19/6392 |
| work_keys_str_mv | AT lauranrbrewster classifyinggoliathgrouperiepinephelusitajaraibehaviorsfromanovelmultisensortag AT alikibrahim classifyinggoliathgrouperiepinephelusitajaraibehaviorsfromanovelmultisensortag AT breannacdegroot classifyinggoliathgrouperiepinephelusitajaraibehaviorsfromanovelmultisensortag AT thomasjostendorf classifyinggoliathgrouperiepinephelusitajaraibehaviorsfromanovelmultisensortag AT hanqizhuang classifyinggoliathgrouperiepinephelusitajaraibehaviorsfromanovelmultisensortag AT laurentmcherubin classifyinggoliathgrouperiepinephelusitajaraibehaviorsfromanovelmultisensortag AT matthewjajemian classifyinggoliathgrouperiepinephelusitajaraibehaviorsfromanovelmultisensortag |