Selecting Orientation-Insensitive Features for Activity Recognition from Accelerometers

Yasser MOHAMMAD  Kazunori MATSUMOTO  Keiichiro HOASHI  

IEICE TRANSACTIONS on Information and Systems   Vol.E102-D   No.1   pp.104-115
Publication Date: 2019/01/01
Online ISSN: 1745-1361
DOI: 10.1587/transinf.2018EDP7092
Type of Manuscript: PAPER
Category: Information Network
activity recognition,  machine learning,  feature selection,  

Full Text: PDF(923.1KB)
>>Buy this Article

Activity recognition from sensors is a classification problem over time-series data. Some research in the area utilize time and frequency domain handcrafted features that differ between datasets. Another categorically different approach is to use deep learning methods for feature learning. This paper explores a middle ground in which an off-the-shelf feature extractor is used to generate a large number of candidate time-domain features followed by a feature selector that was designed to reduce the bias toward specific classification techniques. Moreover, this paper advocates the use of features that are mostly insensitive to sensor orientation and show their applicability to the activity recognition problem. The proposed approach is evaluated using six different publicly available datasets collected under various conditions using different experimental protocols and shows comparable or higher accuracy than state-of-the-art methods on most datasets but usually using an order of magnitude fewer features.