Deep convolutional neural networks on multichannel time series for human activity recognition

TLDR

Human activity recognition relies on multichannel inertial sensor time‑series, yet extracting discriminative features is difficult and most prior work uses hand‑crafted features and shallow models that fail to capture distinguishing patterns. The authors propose a systematic feature‑learning approach for HAR. They employ a deep convolutional neural network that learns hierarchical representations from raw sensor signals, and through supervised training produces discriminative features that jointly improve feature extraction and classification. Experiments on the Opportunity Activity Recognition Challenge and other benchmark datasets demonstrate that the CNN outperforms existing HAR algorithms.

Abstract

This paper focuses on human activity recognition (HAR) problem, in which inputs are multichannel time series signals acquired from a set of bodyworn inertial sensors and outputs are predefined human activities. In this problem, extracting effective features for identifying activities is a critical but challenging task. Most existing work relies on heuristic hand-crafted feature design and shallow feature learning architectures, which cannot find those distinguishing features to accurately classify different activities. In this paper, we propose a systematic feature learning method for HAR problem. This method adopts a deep convolutional neural networks (CNN) to automate feature learning from the raw inputs in a systematic way. Through the deep architecture, the learned features are deemed as the higher level abstract representation of low level raw time series signals. By leveraging the labelled information via supervised learning, the learned features are endowed with more discriminative power. Unified in one model, feature learning and classification are mutually enhanced. All these unique advantages of the CNN make it outperform other HAR algorithms, as verified in the experiments on the Opportunity Activity Recognition Challenge and other benchmark datasets.