A Survey on Activity Detection and Classification Using Wearable Sensors

Cornacchia, Maria; Özcan, Koray; Zheng, Yu; Velipasalar, Senem

doi:10.1109/jsen.2016.2628346

Cited by 358 publications

(186 citation statements)

References 193 publications

Supporting

Mentioning

183

Contrasting

Unclassified

Order By: Relevance

“…Table 1 provides an example of some recent algorithms to segment macro gait. (A more comprehensive presentation of gait recognition algorithms can be found elsewhere [41].) Once macro recognition is achieved, the segmented signals can be examined for micro gait characteristics.…”

Section: Activity Recognition: Macro and Micro Gaitmentioning

confidence: 99%

Wearables for independent living in older adults: Gait and falls

Godfrey

2017

Maturitas

View full text Add to dashboard Cite

Solutions are needed to satisfy care demands of older adults to live independently. Wearable technology (wearables) are one approach to offer a viable means for ubiquitous, sustainable and scalable monitoring in habitual free-living environments. Gait has been presented as a relevant (bio) marker in ageing and pathological studies, with objective assessment achievable by inertial-based wearables. Commercial wearables have struggled to provide accurate analytics and have been limited by non-clinically oriented gait outcomes. Moreover, some research grade wearables also fail to provide transparent functionality due to limitations in proprietary software. Innovation within this field is often sporadic with large heterogeneity of wearable types and algorithms for gait outcomes leading to a lack of pragmatic use. This review provides a summary of recent wearable gait assessment literature, focusing on the need for an algorithm fusion approach to measurement cumulating in the ability to better detect and classify falls. A brief presentation of wearables in one pathological group is presented, identifying appropriate work for researchers in other cohorts to utilise. Opportunities for how this domain needs to progress are also summarised. Highlights• Wearables can meet older adults' needs for independent living.• Gait assessment is a (bio)marker within ageing and different pathologies.• Measuring gait with wearables has been innovative but fraught with inconsistencies.

show abstract

Section: Activity Recognition: Macro and Micro Gaitmentioning

confidence: 99%

Wearables for independent living in older adults: Gait and falls

Godfrey

2017

Maturitas

View full text Add to dashboard Cite

show abstract

“…This was highlighted by a recent survey conducted in [9], which showed an even bigger trend in the use of IMUs with accelerometers, gyroscopes and magnetometers. The increased usage and interest of wrist-worn devices is not surprising, given the global acceptance of these devices in our daily lives.…”

Section: Related Workmentioning

confidence: 99%

Location-Enhanced Activity Recognition in Indoor Environments Using Off the Shelf Smart Watch Technology and BLE Beacons

Filippoupolitis

Oliff

Takand

et al. 2017

Sensors

View full text Add to dashboard Cite

Activity recognition in indoor spaces benefits context awareness and improves the efficiency of applications related to personalised health monitoring, building energy management, security and safety. The majority of activity recognition frameworks, however, employ a network of specialised building sensors or a network of body-worn sensors. As this approach suffers with respect to practicality, we propose the use of commercial off-the-shelf devices. In this work, we design and evaluate an activity recognition system composed of a smart watch, which is enhanced with location information coming from Bluetooth Low Energy (BLE) beacons. We evaluate the performance of this approach for a variety of activities performed in an indoor laboratory environment, using four supervised machine learning algorithms. Our experimental results indicate that our location-enhanced activity recognition system is able to reach a classification accuracy ranging from 92% to 100%, while without location information classification accuracy it can drop to as low as 50% in some cases, depending on the window size chosen for data segmentation.

show abstract

“…Activities recognizable using images include fine-grained activities (making tea vs coffee), social interactions [12] and context. [8] includes an extensive review of wearable camera-based activity recognition works. Other than wearable cameras, visual HAR based on stationary cameras has been extensively studied by the computer vision community and we refer readers to [39] for a survey in this area.…”

Section: Are Imagers the Answer?mentioning

confidence: 99%

“…Image data contains substantially more information than acceleration traces. Visual details captured in images may be high-level scene features quantifying the motion or contextualizing the scene, or they may be fine-grained details which specify an object that a person is interacting with [8]. The rich information provided by images not only enables more granular definitions of activity classes, it also provides abundant context for open-ended study of the subjects' behaviour.…”

Section: Are Imagers the Answer?mentioning

confidence: 99%

Are Accelerometers for Activity Recognition a Dead-end?

Tong

Tailor

Lane

2020

Proceedings of the 21st International Workshop on Mobile Computing Systems and Applications

View full text Add to dashboard Cite

Accelerometer-based (and by extension other inertial sensors) research for Human Activity Recognition (HAR) is a dead-end. This sensor does not offer enough information for us to progress in the core domain of HAR-to recognize everyday activities from sensor data. Despite continued and prolonged efforts in improving feature engineering and machine learning models, the activities that we can recognize reliably have only expanded slightly and many of the same flaws of early models are still present today. Instead of relying on acceleration data, we should instead consider modalities with much richer information-a logical choice are images. With the rapid advance in image sensing hardware and modelling techniques, we believe that a widespread adoption of image sensors will open many opportunities for accurate and robust inference across a wide spectrum of human activities.In this paper, we make the case for imagers in place of accelerometers as the default sensor for human activity recognition. Our review of past works has led to the observation that progress in HAR had stalled, caused by our reliance on accelerometers. We further argue for the suitability of images for activity recognition by illustrating their richness of information and the marked progress in computer vision. Through a feasibility analysis, we find that deploying imagers and CNNs on device poses no substantial burden on modern mobile hardware. Overall, our work highlights the need to move away from accelerometers and calls for further exploration of using imagers for activity recognition. 1

show abstract

A Survey on Activity Detection and Classification Using Wearable Sensors

Cited by 358 publications

References 193 publications

Wearables for independent living in older adults: Gait and falls

Wearables for independent living in older adults: Gait and falls

Location-Enhanced Activity Recognition in Indoor Environments Using Off the Shelf Smart Watch Technology and BLE Beacons

Are Accelerometers for Activity Recognition a Dead-end?

Contact Info

Product

Resources

About