Swimming Style Classification Based on Ensemble Learning and Adaptive Feature Value by Using Inertial Measurement Unit

Abstract

We have been constructing a swimming ability improvement support system. One of the issues to be addressed is the automatic classification of swimming styles (backstroke, breaststroke, butterfly, and front crawl). The mainstream swimming style classification technique of conventional researches is based on non-ensemble learning; in their classification, breaststroke and butterfly are mixed up with each other. To improve its generalization performance, we need to use better classifiers and more adaptive feature values than previously considered. Therefore, this research has introduced (1) random forest technique, one of ensemble learning techniques, and (2) feature values specific to breaststroke and butterfly to construct a four-swimming-style classifier that has resolved this issue. From subjects with 7 to 20 years history of swimming races, we have obtained their sensor data during swimming and have divided the data into learning data and test data. We have also converted them into feature values that represent their body motions. We have selected from those body-motion-representing feature values the important data to classify four swimming styles and feature values specific to breaststroke and butterfly. We have used the learning data to construct a swimming style classifier, and the test data to evaluate its classification accuracy. The evaluation results show that (1’) the introduction of ensemble learning has improved the mean value of F-measure for breaststroke and butterfly by 0.053, and (2’) the introduction of feature values specific to breaststroke and butterfly has improved the mean value of F-measure for breaststroke and butterfly by 0.121 as compared with (1’). The proposed swimming style classifier has performed a mean F-measure of 0.981 for the four swimming styles as well as good classification accuracies for front crawl and backstroke. Therefore, we have concluded that the swimming style classifier we have constructed has resolved the problem of mixing up breaststroke and butterfly, as well as can properly classify all different swimming styles.

Automatic Swimming Activity Recognition and Lap Time Assessment Based on a Single IMU: A Deep Learning Approach

“…Furthermore, Ref. [ 58 ] showed that combining several consecutive predictions led to 100% good predictions. More recently, Ref.…”

A Novel Macro-Micro Approach for Swimming Analysis in Main Swimming Techniques Using IMU Sensors

“…Moreover, inter-cyclic variation is another important factor (Barbosa et al, 2005;Figueiredo et al, 2012) which might cause error in technique identification, which was not examined in this study. Our method, however, presents a higher accuracy in comparison to reported result in the literature based on sacrum sensor (Davey et al, 2008;Omae et al, 2017). Some studies use a network of IMUs (Wang et al, 2016) or a smartphone (Pan et al, 2016) for swimming technique identification while we focused on each sensor location separately.…”

Feature Selection Algorithm Considering Trial and Individual Differences for Machine Learning of Human Activity Recognition

“…To evaluate how our proposed method is an improved solution, we have solved the same problem using existing feature space evaluation indices such as BW-Ratio [7], OOB [8], ReliefF [9], and Minimum Reference Set (MRS) [10]. Among the existing feature space evaluation indices, BW-Ratio, ReliefF, and OOB are widely used as general methods for selecting features, and are believed to [11,12], OOB, in References [13,14], and Reli-efF, in References [15,16]). In contrast, MRS selects features using a small amount of training data, which is one of the aims of this paper.…”