Human Mobility Monitoring in Very Low Resolution Visual Sensor Network

Bo, Nyan Bo; Deboeverie, Francis; Eldib, Mohamed; Guan, Junzhi; Xie, Xingzhe; Niño, Jorge; Haerenborgh, Dirk Van; Slembrouck, Maarten; Velde, Samuel Van de; Steendam, Heidi; Veelaert, Peter; Kleihorst, Richard; Aghajan, Hamid; Philips, Wilfried

doi:10.3390/s141120800

Cited by 18 publications

(10 citation statements)

References 18 publications

(31 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The resident has a very clear mind. In previous work, 9 users' locations and mobility statistics were obtained from a robust people tracker based on recursive maximum likelihood principles. However, this people tracker requires accurate camera calibration.…”

Section: Introductionmentioning

confidence: 99%

Behavior analysis for elderly care using a network of low-resolution visual sensors

Eldib

Deboeverie

Philips

et al. 2016

J. Electron. Imaging

Self Cite

View full text Add to dashboard Cite

Recent advancements in visual sensor technologies have made behavior analysis practical for in-home monitoring systems. The current in-home monitoring systems face several challenges: (1) visual sensor calibration is a difficult task and not practical in real-life because of the need for recalibration when the visual sensors are moved accidentally by a caregiver or the senior citizen, (2) privacy concerns, and (3) the high hardware installation cost. We propose to use a network of cheap low-resolution visual sensors (30×30 pixels) for long-term behavior analysis. The behavior analysis starts by visual feature selection based on foreground/background detection to track the motion level in each visual sensor. Then a hidden Markov model (HMM) is used to estimate the user’s locations without calibration. Finally, an activity discovery approach is proposed using spatial and temporal contexts. We performed experiments on 10 months of real-life data. We show that the HMM approach outperforms the k-nearest neighbor classifier against ground truth for 30 days. Our framework is able to discover 13 activities of daily livings (ADL parameters). More specifically, we analyze mobility patterns and some of the key ADL parameters to detect increasing or decreasing health conditions

show abstract

Section: Introductionmentioning

confidence: 99%

Behavior analysis for elderly care using a network of low-resolution visual sensors

Eldib

Deboeverie

Philips

et al. 2016

J. Electron. Imaging

Self Cite

View full text Add to dashboard Cite

show abstract

“…p m (t)) where m is the number of the visual sensors. In this paper, we opted to use the correlation method as a foreground/background subtraction algorithm for its good performance with the visual sensor of [3], as indicated in a previous study [18]. The obtained results were good enough for detecting motion patterns produced by the senior citizen as shown in Figure 3.…”

Section: Sleep Analysis Approachmentioning

confidence: 99%

“…We used the low-resolution visual sensors [3] in previous studies [6,18] for person tracking in our lab, as a first step, towards building a sensor network in a service flat for activity analysis. The sensor network deployed in the service flat is composed of 10 cameras as shown in Figure 1.…”

Section: Sleep Analysis Approachmentioning

confidence: 99%

Sleep Analysis for Elderly Care Using a Low-Resolution Visual Sensor Network

Eldib

Deboeverie

Philips

et al. 2015

Human Behavior Understanding

Self Cite

View full text Add to dashboard Cite

Abstract. Nearly half of the senior citizens report difficulty initiating and maintaining sleep. Frequent visits to the bathroom in the middle of the night is considered as one of the major reasons for sleep disorder. This leads to serious diseases such as depression and diabetes. In this paper, we propose to use a network of cheap low-resolution visual sensors (30 × 30 pixels) for long-term activity analysis of a senior citizen in a service flat. The main focus of our research is on elderly behaviour analysis to detect health deterioration. Specifically, this paper treats the analysis of sleep patterns. Firstly, motion patterns are detected. Then, a rule-based approach on the motion patterns is proposed to determine the wake up time and sleep time. The nightly bathroom visit is identified using a classification-based model. In our evaluation, we performed experiments on 10 months of real-life data. The ground truth is collected from the diaries in which the senior citizen wrote down his sleep time and wake up time. The results show accurate extraction of the sleep durations with an overall Mean Absolute Error (MAE) of 22.91 minutes and Spearman correlation coefficient of 0.69. Finally, the nightly bathroom visits analysis indicate sleep disorder in several nights.

show abstract

“…In the case of homogeneous sensor networks, feedback R c is utilised for projection modelling. Considering the homogeneous network design solution presented in the previous section and using (2), the chosen sensing range R c = 8.09m leads to 3D coverage volume 106.90m 3 .…”

Section: H Sensor's 3d Coverage Estimationmentioning

confidence: 99%

“…In a VSN, each node captures image data that can be processed locally to extract relevant information (such as visual features) and it collaborates with other nodes in the network [2]. VSNs are used in surveillance [3,4], environmental monitoring [5,6], assisted living and tele-healthcare [7,8] applications. Visual sensors within a VSN employ directional sensing to provide pixel based measurements as a 2D dataset and they require a large bandwidth to transmit image data.…”

Section: Introductionmentioning

confidence: 99%

Characterization of Field-of-View for Energy Efficient Application-Aware Visual Sensor Networks

et al. 2016

View full text Add to dashboard Cite

Energy consumption is one of the primary concerns in a resource constrained Visual Sensor Network. The existing Visual Sensor Network design solutions under particular resource constrained scenarios are application specific; whereas the degree of sensitivity of any of these resource constrains (e.g. energy etc) varies from one application to another. This limits the implementation of the existing energy efficient solutions within a Visual Sensor Network node which may be considered to be a part of a heterogeneous network. The heterogeneity of image capture and processing within a Visual Sensor Network can be adaptively reflected with a dynamic Field-of-View realisation. This is expected to allow the implementation of a generalised energy efficient solution to adapt with the heterogeneity of the network. In this paper, an energy efficient Field-of-View characterisation framework is proposed which can support a diverse range of applications. The context of adaptivity in the proposed Field-of-View characterisation framework is considered to be: a) sensing range selection; b) maximising spatial coverage; c) adaptive task classification and d) minimising the number of required nodes. Soft decision criteria is exploited and it is observed that for a given detection reliability, the proposed framework provides energy efficient solutions which can be implemented within heterogeneous networks. It is also found that the proposed design solution for heterogeneous networks leads to 49.8% energy savings compared to the trivial design solution.

show abstract

Human Mobility Monitoring in Very Low Resolution Visual Sensor Network

Cited by 18 publications

References 18 publications

Behavior analysis for elderly care using a network of low-resolution visual sensors

Behavior analysis for elderly care using a network of low-resolution visual sensors

Sleep Analysis for Elderly Care Using a Low-Resolution Visual Sensor Network

Characterization of Field-of-View for Energy Efficient Application-Aware Visual Sensor Networks

Contact Info

Product

Resources

About