A Comparison of Feature Extraction Methods for the Classification of Dynamic Activities From Accelerometer Data
Summary (2 min read)
1. Introduction
- Over the last decade there has been considerable research effort directed towards the monitoring and classification of physical activity patterns from body-fixed sensor data [1, 2].
- A mobile phone may detect when a person is driving a vehicle and automatically divert a call.
- Effective algorithms are also required to interpret the accelerometer data in the context of ~3~ different activities.
- The overall aim of this study was to extensively compare the performance of a number of previously reported and novel wavelet features with a range of time-domain and frequency domain features for the classification of different activities.
- It was felt that this work would underpin the development of an off-the-shelf activity monitor which could be used to classify activity patterns across different subjects.
2.1 Data collection
- Accelerometer data was collected using Pegasus activity monitors developed by ETB, UK.
- A sampling frequency of 64Hz was selected for this study as this is sufficiently larger than the 20Hz sampling required to assess daily activity [26].
- To secure each unit in place specialised bandage (FabriFoam®) was first positioned around each of the body segments and the activity monitors, which were backed with Velcro®, adhered to the underwrapped bandage.
- A number of studies have shown that static postures can be differentiated from dynamic activity by applying a single threshold to some measure of acceleration variability [28, 29].
- Subjects completed a total of eight different activities (level walking, walking upstairs and downstairs, jogging, running, hopping on the left and right leg and jumping).
2.2 Wavelet features
- A number of previous activity classification studies have derived time-frequency features obtained using the filter bank interpretation of the discrete wavelet transform (DWT) [22, 24].
- The first set of wavelet features was proposed by Tamura et al. [23].
- Again there are two features, the first being the total of the summations of the detail signal at levels 6 and 7.
- In contrast Wang et al. [14] used wavelet packet analysis to derive 33 features from a tri-axial accelerometer signal.
- Given the high sampling frequency used by Sekine et al. [21] (1024 Hz), they were able to calculate the fractal dimension from the variance of the detail coefficients across seven different levels.
2.3 Time and frequency-domain features
- For additional comparison, the authors also employed three sets of time-domain features and four sets of frequency-domain features (Table II).
- Within each of these seven sets, the features were derived individually for each of the three components of the tri-axial accelerometer signal.
- These two statistics were therefore used to define the third set of time-domain features.
- This has been used previously as an addition to time-domain measures in order to improve classification accuracy [35].
- The second frequency-domain feature set was chosen to be spectral energy, which is defined to be the sum of the squared ~11~ FFT coefficients [11, 43].
2.4 Activity classification
- In order to compare the discriminate ability of each of the different features sets, a k-Nearest Neighbour (kNN) classifier was implemented and its accuracy determined using leave-onesubject-out cross validation.
- This process is repeated until each subject has been used once as the testing dataset.
- Cross validation is a popular statistical resampling procedure [44] and the authors use it here to evaluate the accuracy of the kNN classifier for a given set of features.
- This test was chosen as it was not possible to guarantee that these distributions were normally distributed.
- For this three-activity classification problem, accuracy was determined for the waist-mounted accelerometer for each of the seven sets of wavelet features and for each of the seven sets of time/frequency features.
3. Results
- Table III gives the classification accuracies for the wavelet feature sets and different accelerometer placements for the three-activity classification problem.
- Table IV illustrates the same information but for the time/frequency features.
- This distribution of accuracies was significantly higher than those obtained from all other feature sets derived from a single sensor (p<0.01).
- In order to establish whether, in general, the time/frequency features outperformed the wavelet features, a number of statistical tests were performed.
- For the time/frequency features, maximal classification accuracy for a single sensor (92±7%) was again obtained when the individual FFT components were derived from the ankle-mounted unit (Table VI).
4. Discussion
- This study was designed to compare the discriminative ability of wavelet features with time/frequency features for two activity classification problems: a simple three-activity problem and an eight-activity problem.
- Analysis of this data showed that features derived from an FFT analysis outperformed those derived from wavelet coefficients.
- Their reported maximum classification accuracy of 84% using data from five sensors is similar to the maximum accuracy (90%) achieved in their study for the eight-activity problem.
- For such individuals, jerkiness of movement may lead to isolated frequency transients which maybe better characterised using wavelet features.
5. Conclusion
- This study was performed on healthy individuals.
- More specifically, the highest levels of classification accuracy were obtained from individual FFT components.
- The study also compared classification accuracies across three different sensor placements and showed a sensor mounted at the ankle to outperform the thigh and waist sensors for most feature sets.
- Further work is required to determine the most appropriate features sets for other subjects groups, such as the elderly or neurologically impaired. ~18~.
Did you find this useful? Give us your feedback
Citations
1,000 citations
Additional excerpts
...coefficients, or wavelet features [62], [63]....
[...]
926 citations
Cites methods from "A Comparison of Feature Extraction ..."
...Various feature extraction methods for the classification of dynamic activities from accelerometer data were compared based on two datasets of activities collected from 20 subjects [94]....
[...]
588 citations
Cites background or methods from "A Comparison of Feature Extraction ..."
...In a recent paper by Preece et al (2008b) the performance of a number of wavelet-based sets was compared to previously used time- and frequency-domain features for the classification of eight different activities....
[...]
...For example, median frequency (Foerster and Fahrenberg 2000) or a subset of the different FFT coefficients can be used (Preece et al 2008a, b) (Preece et al., 2008b, a)....
[...]
...A range of window sizes have been used in previous studies from 0.25 secs (Huynh and Schiele 2005) to 6.7 secs (Bao and Intille 2004), with some studies including a degree of overlap between adjacent windows (Bao and Intille 2004; Preece et al 2008b)....
[...]
...The range of different features includes signal magnitude area (the area under the high pass filtered acceleration curve) (Mathie et al 2003; Preece et al 2008c), peak-to-peak acceleration (Makikawa and Iizumi 1995), mean rectified value (Bussmann et al 1998a; Bussmann et al 1998b) and root mean…...
[...]
...More recently the kNN approach has been compared to other classification schemes (Bao and Intille 2004; Maurer et al 2006) (Table 4) and used as part of an algorithm for comparing different features for activity classification (Huynh and Schiele 2005; Preece et al 2008b)....
[...]
561 citations
Cites background from "A Comparison of Feature Extraction ..."
...selected activity classes (or a subset of them) have been investigated prominently in a large number of HAR publications [10, 11, 14, 27, 29, 41, 47]....
[...]
549 citations
Cites background from "A Comparison of Feature Extraction ..."
...For more information the reader is referred to [40,45]....
[...]
References
24,199 citations
"A Comparison of Feature Extraction ..." refers background in this paper
...The fractal dimension quantifies the variance progression of the detail coefficient over the different wavelet scales and as such gives a measure of the complexity within the original signal [41]....
[...]
20,028 citations
7,560 citations
4,064 citations
"A Comparison of Feature Extraction ..." refers background in this paper
...it is possible to identify localized temporal instances at which there is a change in frequency characteristics of the original signal [25]....
[...]
3,223 citations
"A Comparison of Feature Extraction ..." refers background or methods in this paper
...A recent study carried out by Bao and Intille [11] obtained high levels of classification accuracy using a mixed set of time and frequency-domain features....
[...]
...This choice was motivated by previous studies that have used a range of different features to characterize acceleration signals [11], [14], [16],...
[...]
...celeration signal [12], [16]–[18] and others from a frequency analysis [11], [13], [19], [20]....
[...]
...cessive sliding windows has been shown to be effective in previous studies of activity classification [11], [38]....
[...]
...The second frequency-domain feature set was chosen to be spectral energy, which is defined to be the sum of the squared FFT coefficients [11], [43]....
[...]
Related Papers (5)
Frequently Asked Questions (14)
Q2. What are the future works mentioned in the paper "A comparison of feature extraction methods for the classification of dynamic activities from accelerometer data" ?
Further work is required to determine the most appropriate features sets for other subjects groups, such as the elderly or neurologically impaired. ~18~
Q3. What is the popular method of evaluating the accuracy of the kNN classifier?
Cross validation is a popular statistical resampling procedure [44] and the authors use it here to evaluate the accuracy of the kNN classifier for a given set of features.
Q4. What is the way to collect and store accelerometer data?
With recent advances in miniaturised sensing technology, it is now possible to collect and store acceleration data from individual body segments over extended periods of time.
Q5. What was used to secure the monitors in place?
To secure each unit in place specialised bandage (FabriFoam®) was first positioned around each of the body segments and the activity monitors, which were backed with Velcro®, adhered to the underwrapped bandage.
Q6. What is the role of the accelerometer in the prediction of fall risk?
Ambulatory monitoring of various activities, including the time spent in sit-stand transitions have shown promise as predictors of fall-risk [8].
Q7. What is the role of portable systems in a ubiquitous computing scenario?
In addition to health-related applications, portable systems which can accurately identify the activity of the user have the potential to play a fundamental role in a ubiquitous computing scenario [9, 10].
Q8. How many steps did the subjects have to do to perform the first activity?
For the first of these two activities, subjects were instructed to perform a gentle jog over a 50m distance and for the second to perform a fast run over the same distance.
Q9. What was the method used for the video problem?
The video method, used~17~in this study, was selected as it was believed to be more accurate than self observation by the subject.
Q10. What is the role of self reported measures in epidemiological studies?
As self reported measures have been shown to be unreliable [4, 5], systems for activity profiling are beginning to play an important role in large-scale epidemiological studies in this area [6, 7].
Q11. What frequency bands were used to calculate the features?
Both Nyan et al. [24] and Sekine et al. [22] collected data at 256Hz, therefore as before, wavelet coefficients corresponding to appropriate frequency bands were used to calculate of each of the features.
Q12. What is the highest classification accuracy for a single sensor?
The highest classification accuracy for a single sensor was obtained for the FFT component feature set and the ankle-mounted sensor.
Q13. What is the method for calculating the classification accuracy of a single sensor?
for the three-activity problem, the highest classification accuracy for a single sensor (97±3%) was obtained using FFT components derived from the ankle-mounted unit.
Q14. What is the way to interpret the accelerometer data?
Although this technology offers the ideal platform for monitoring daily activity patterns, effective algorithms are also required to interpret the accelerometer data in the context of~3~different activities.