Deep-Learning-Based Signal Enhancement of Low-Resolution Accelerometer for Fall Detection Systems

• Published in: IEEE transactions on cognitive and developmental systems Vol. 14; no. 3; pp. 1270 - 1281
• Main Authors: Liu, Kai-Chun, Hung, Kuo-Hsuan, Hsieh, Chia-Yeh, Huang, Hsiang-Yun, Chan, Chia-Tai, Tsao, Yu
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.09.2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Accelerometer signal enhancement (ASE); Accelerometers; Batteries; Deep learning; deep learning (DL) approach; Degradation; Fall detection; Feature extraction; low-resolution fall detection (LR-FD); Microprocessors; Power demand; Sampling; Signal processing; Signal resolution; Support vector machines; wearable sensors
• ISSN: 2379-8920; 2379-8939
• DOI: 10.1109/TCDS.2021.3116228

In the last two decades, fall detection (FD) systems have been developed as a popular assistive technology. To support long-term FD services, various power-saving strategies have been implemented. Among them, a reduced sampling rate is a common approach for an energy-efficient system in the real world. However, the performance of FD systems is diminished owing to low-resolution (LR) accelerometer signals. To improve the detection accuracy with LR accelerometer signals, several technical challenges must be considered, including mismatch of effective features and the degradation effects. In this work, a deep-learning-based accelerometer signal enhancement (ASE) model is proposed as a front-end processor to help typical LR-FD systems achieve better detection performance. The proposed ASE model based on a deep denoising convolutional autoencoder architecture reconstructs high-resolution (HR) signals from the LR signals by learning the relationship between the LR and HR signals. The results show that the FD system using support vector machine (SVM) and the proposed ASE model at an extremely low sampling rate (sampling rate < 2 Hz) achieved 97.34% and 90.52% accuracies in the SisFall and FallAllD data sets, respectively, while those without ASE models only achieved 95.92% and 87.47% accuracies in the SisFall and FallAllD data sets, respectively. The results also demonstrate that the proposed ASE mode can be suitably combined with deep-learning-based FD systems.