Table 1 Summary of the previous studies which used feature extraction and coordinate transformation methods
Method | Sensor types | Sensor positions | Classifier | References |
|---|---|---|---|---|
Feature extraction | Accelerometer and gyroscope | Coat pocket, hand, trouser pocket and bag | SVM | [24] |
Feature extraction | Accelerometer and gyroscope | Right upper arm, right hand, right jacket pocket, right trousers pocket and waist | KNN, RF, SVM | [25] |
Feature extraction | Accelerometer, gyroscope and magnetometer | Multiple positions from 5 data sets | Bayesian decision making (BDM), KNN, SVM, Artificial Neural Network (ANN) | [26] |
Coordination transformation | Accelerometer and gyroscope | Left upper arm, the shirt pocket, the trousers front pocket, and the behind trouser pocket | Motif-based classification system | [31] |
Coordination transformation and feature extraction | Accelerometer | Pants’ pocket, shirt’s pocket and backpack | Online SVM | [32] |
Coordination transformation and feature extraction | Accelerometer, gyroscope and magnetic field | Trouser pockets | KNN | [33] |