Classification of Hand-Based and Non-Hand-Based Physical Activities Using Wearable Sensors

Daş, Duygu Bağcı; Birant, Derya

doi:10.1109/ismsit50672.2020.9255171

Cited by 2 publications

(3 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Classification and comments [25] 6 (standing, sitting, lying, walking, and transitions: sit-to-stand and stand-to-sit) The study also includes free-living activities Detecting human movements through wearables is achieved mainly by accelerometers [30][31][32], inertial units, that also include gyroscopes or magnetometers [16,27], or the use of smartphones or smartwatches [36,38], which integrate all these elements. Other studies have also analyzed the use of a set of pressure sensors placed in a shoe insole [19] or integrated into an armband, to detect upper limb movements [22].…”

Section: Ref #Activities Devices and Placement Metricsmentioning

confidence: 99%

“…Traditionally, various methods from the field of signal processing have been leveraged to distill collected sensor data. These have included k-NN [14,30,33,35], random forest (RF), decision tree (DT) [20,38], gaussian mixture model (GMM) and hidden Markov models (HMM) [16,24] or even models based exclusively on thresholds [29] ], all of which requires domain-specific expert knowledge to process raw data. Feature engineering is required to fit a model and this is expensive and not scalable.…”

Section: Ref #Activities Devices and Placement Metricsmentioning

confidence: 99%

See 1 more Smart Citation

Towards Human Stress and Activity Recognition: A Review and a First Approach Based on Low-Cost Wearables

et al. 2022

View full text Add to dashboard Cite

Detecting stress when performing physical activities is an interesting field that has received relatively little research interest to date. In this paper, we took a first step towards redressing this, through a comprehensive review and the design of a low-cost body area network (BAN) made of a set of wearables that allow physiological signals and human movements to be captured simultaneously. We used four different wearables: OpenBCI and three other open-hardware custom-made designs that communicate via bluetooth low energy (BLE) to an external computer—following the edge-computingconcept—hosting applications for data synchronization and storage. We obtained a large number of physiological signals (electroencephalography (EEG), electrocardiography (ECG), breathing rate (BR), electrodermal activity (EDA), and skin temperature (ST)) with which we analyzed internal states in general, but with a focus on stress. The findings show the reliability and feasibility of the proposed body area network (BAN) according to battery lifetime (greater than 15 h), packet loss rate (0% for our custom-made designs), and signal quality (signal-noise ratio (SNR) of 9.8 dB for the ECG circuit, and 61.6 dB for the EDA). Moreover, we conducted a preliminary experiment to gauge the main ECG features for stress detection during rest.

show abstract

Section: Ref #Activities Devices and Placement Metricsmentioning

confidence: 99%

Section: Ref #Activities Devices and Placement Metricsmentioning

confidence: 99%

Towards Human Stress and Activity Recognition: A Review and a First Approach Based on Low-Cost Wearables

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Most HAR studies in the literature (Gao et al, 2019; Tian et al, 2019) have attempted the identification of basic human activities such as sitting, downstairs, walking, and upstairs. Nonetheless, some recent HAR studies focused on identifying various types of activities, such as military‐based activities (Mukherjee et al, 2017) (i.e., run, jump, jump‐rope), transportation‐based activities (Carpineti et al, 2018) (i.e., riding a car, train, or bus), hand‐oriented/non‐hand‐oriented activities (Das & Birant, 2020; Weiss et al, 2019) (i.e., typing, clapping, eating, drinking or kicking, jogging), and sports activities (Zokas & Lukosevicius, 2018) (i.e., push‐ups, sit‐ups, and squats). In addition to such activities, some studies mainly considered transitional human activities (sit‐to‐stand or stand‐to‐sit, etc.)…”

Section: Related Workmentioning

confidence: 99%

Human activity recognition based on multi‐instance learning

Daş

Birant

2023

Expert Systems

Self Cite

View full text Add to dashboard Cite

Human activity recognition (HAR) is the process of classifying a person's actions, and it is an essential task for many human-centered applications. Multi-instance learning (MIL) is a special case of machine learning where the training examples are bags containing many instances, and a single class label is assigned for an entire bag of instances. In this study, we integrated these two concepts by introducing a novel approach: "human activity recognition based on multi-instance learning", called HAR-MIL. Unlike previous studies, the proposed HAR-MIL method represents human activities differently: as a bag of various wearable sensors (gyroscope, magnetometer, accelerometer, and linear acceleration). HAR-MIL presents an applicable and flexible model by providing multiinstance representation and eliminating the restrictions of traditional single-instance representation. Therefore, the adverse effects of missing data, defective sensors, and biased measurement on activity classification performance were minimized. This study is the first to investigate the performance of two MIL algorithms (SimpleMI and MIWrapper) on HAR. In this study, we explored the effect of four main factors (sensor positions, sensor types, base learners, and single or multiple participants) on the multiinstance representation of relevant human daily activities. The effectiveness of the proposed HAR-MIL method was demonstrated on 50 participant-based and sensorposition-based activity recognition datasets. The experimental results showed that HAR-MIL is effective for wearable sensor-based HAR with high classification accuracy (99.32%). Furthermore, the results showed that the proposed method outperformed the state-of-the-art methods by 10% on average on the same dataset.

show abstract

Classification of Hand-Based and Non-Hand-Based Physical Activities Using Wearable Sensors

Cited by 2 publications

References 14 publications

Towards Human Stress and Activity Recognition: A Review and a First Approach Based on Low-Cost Wearables

Towards Human Stress and Activity Recognition: A Review and a First Approach Based on Low-Cost Wearables

Human activity recognition based on multi‐instance learning

Contact Info

Product

Resources

About