Human Daily Activity Recognition With Sparse Representation Using Wearable Sensors

Mi, Zhang; Sawchuk, Alexander A.

doi:10.1109/jbhi.2013.2253613

Cited by 166 publications

(83 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similar results of researches on activity classification using wearable sensor have been also reported in Yang et al 9 and Zhang and Sawchuk. 15 In conclusion, our proposed technique can take advantage of the block sparsity of multi-sensors data for the best performance of joint reconstruction of nonsparse multi-sensors data as well as the lowest computational time cost. This helps to further activity classification with high quality.…”

Section: Discussionmentioning

confidence: 87%

DCS-based MBSBL joint reconstruction of multi-sensors data for energy-efficient telemonitoring of human activity

Wang

Ling

2018

International Journal of Distributed Sensor Networks

View full text Add to dashboard Cite

The joint reconstruction of nonsparse multi-sensors data with high quality is a challenging issue in human activity telemonitoring. In this study, we proposed a novel joint reconstruction algorithm combining distributed compressed sensing with multiple block sparse Bayesian learning. Its basic idea is that based on the joint sparsity model, the distributed compressed sensing technique is first applied to simultaneously compress the multi-sensors data for gaining the highcorrelation information regarding activity as well as the energy efficiency of sensors, and then, the multiple block sparse Bayesian learning technique is employed to jointly recover nonsparse multi-sensors data with high fidelity by exploiting the joint block sparsity. The multi-sensors acceleration data from an open wearable action recognition database are selected to assess the practicality of our proposed technique. The sparse representation classification model is used to classify activity patterns using the jointly reconstructed data in order to further examine the effectiveness of our proposed method. The results showed that when compression rates are selected properly, our proposed technique can gain the best joint reconstruction performance as well as energy efficiency of sensors, which greatly contributes to the best sparse representation classification-based activity classification performance. This has a great potential for energyefficient telemonitoring of human activity.

show abstract

Section: Discussionmentioning

confidence: 87%

DCS-based MBSBL joint reconstruction of multi-sensors data for energy-efficient telemonitoring of human activity

Wang

Ling

2018

International Journal of Distributed Sensor Networks

View full text Add to dashboard Cite

show abstract

“…The system demonstrates many advantages such as low-power, easy configuration, convenient carrying, and real-time reliable data. In [36], the authors use wearable sensors to monitor daily activities of humans which they perform during different activities. The correct monitoring of these complex actions is challenging.…”

Section: Energy-efficient Routingmentioning

confidence: 99%

Modeling induction and routing to monitor hospitalized patients in multi-hop mobility-aware body area sensor networks

Javaid

Ahmad

Tauqir

et al. 2016

J Wireless Com Network

View full text Add to dashboard Cite

In wireless body area sensor networks (WBASNs), energy efficiency is an area of extreme significance. At first, we present a mathematical model for a non-invasive inductive link which is used to recharge the battery of an implanted biomedical device (pacemaker). Afterwards, we propose a distance-aware relaying energy-efficient (DARE) and mutual information-based DARE (MI-DARE) routing protocols for multihop mobility-aware body area sensor networks (MM-BASNs). Both the routing protocols and the non-invasive inductive link model are tested with the consideration of eight patients in a hospital unit under different topologies, where the vital signs of each patient are monitored through seven on-body sensors and an implanted pacemaker. To reduce energy consumption of the network, the sensors communicate with a sink via an on-body relay which is fixed on the chest of each patient. The behavior (static/mobile) and position of the sink are changed in each topology, and the impact of mobility due to postural changes of the patient(s) arms, legs, and head is also investigated. The MI-DARE protocol further prolongs the network lifetime by minimizing the number of transmissions. Simulation results show that the proposed techniques outperform contemporary schemes in terms of the selected performance metrics.

show abstract

“…Five popular features were extracted from each window: the mean value, the standard deviation, the entropy, the energy and the correlation coefficients between X and Z axis, Y and Z axis [11][12][13]. As an example, Figure 1 shows the raw accelerometer signal along with the first three IMFs for the X axis of different movements ((a) the sitting, (b) the walking).…”

Section: Traditional Features Extractionmentioning

confidence: 99%

The application of EMD in activity recognition based on a single triaxial accelerometer

Liao

Guo

Qin

et al. 2015

BME

View full text Add to dashboard Cite

Abstract. Activities recognition using a wearable device is a very popular research field. Among all wearable sensors, the accelerometer is one of the most common sensors due to its versatility and relative ease of use. This paper proposes a novel method for activity recognition based on a single accelerometer. To process the activity information from accelerometer data, two kinds of signal features are extracted. Firstly, five features including the mean, the standard deviation, the entropy, the energy and the correlation are calculated. Then a method called empirical mode decomposition (EMD) is used for the feature extraction since accelerometer data are non-linear and non-stationary. Several time series named intrinsic mode functions (IMFs) can be obtained after the EMD. Additional features will be added by computing the mean and standard deviation of first three IMFs. A classifier called Adaboost is adopted for the final activities recognition. In the experiments, a single sensor is separately positioned in the waist, left thigh, right ankle and right arm. Results show that the classification accuracy is 94.69%, 86.53%, 91.84% and 92.65%, respectively. These relatively high performances demonstrate that activities can be detected irrespective of the position by reducing problems such as the movement constrain and discomfort.

show abstract

Human Daily Activity Recognition With Sparse Representation Using Wearable Sensors

Cited by 166 publications

References 20 publications

DCS-based MBSBL joint reconstruction of multi-sensors data for energy-efficient telemonitoring of human activity

DCS-based MBSBL joint reconstruction of multi-sensors data for energy-efficient telemonitoring of human activity

Modeling induction and routing to monitor hospitalized patients in multi-hop mobility-aware body area sensor networks

The application of EMD in activity recognition based on a single triaxial accelerometer

Contact Info

Product

Resources

About