Frequency based classification of activities using accelerometer data

Sharma, Annapurna; Purwar, Amit; Lee, Young-Dong; Lee, Youngsook; Chung, Wan-Young

doi:10.1109/mfi.2008.4648056

Cited by 24 publications

(17 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The cost of power for computations when the device is in a continuous monitoring mode has to be kept at a minimum. Sharma et al conducted a detailed analysis of frequency-based classification of activities such as rest, walk, and run, using data from an accelerometer [7]. The paper highlighted the classification of user activities based on frequency components seen in the accelerometer readings in a wireless sensor network.…”

Section: Related Workmentioning

confidence: 99%

Real-Time Processing and Analysis for Activity Classification to Enhance Wearable Wireless ECG

Shaji

Ramesh

Menon

2015

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

Health care facilities in rural India are in a state of utter indigence. Over three-fifths of those who live in rural areas have to travel more than five km to reach a hospital and health care services are becoming out of reach for economically challenged communities in India. Since rural communities experience about 22.9 % of deaths due to heart disease [1], there is a need to improve remote ECG monitoring devices to cater to the needs of rural India. The existing wearable ECG devices experience several issues in accurately detecting the type of heart disease someone has due to the presence of motion artifacts. Hence, even though wearable devices are finding their place in today's health care systems, the above-mentioned issues discourage a doctor in depending upon it. So to enhance the existing wearable ECG device, we designed a context aware system to collect the body movement activity (BMA). In this paper, an innovative BMA classifier has been designed to classify the physical activities of user from the real-time data received from a context aware device. The test results of the BMA classifier integrated with the complete system show that the algorithm developed in this work is capable of classifying the user activity such as walking, jogging, sitting, standing, climbing upstairs, coming downstairs, and lying down, with an accuracy of 96.66 %.

show abstract

Section: Related Workmentioning

confidence: 99%

Real-Time Processing and Analysis for Activity Classification to Enhance Wearable Wireless ECG

Shaji

Ramesh

Menon

2015

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

show abstract

“…In the time domain the features were based on the following studies [6,7] and we extract the following features: mean vector, standard deviation vector, euclidean norm of mean vector, euclidean norm of the standard deviation and correlation values. The features in the frequency domain were based in [8,9] and was extracted the power spectral density thought the fast fourier transform.…”

Section: Feature Extractionmentioning

confidence: 99%

Classification of Physical Activities Using a Smartphone: Evaluation Study Using Multiple Users

Duarte

Lourenço

Abrantes

2014

Procedia Technology

View full text Add to dashboard Cite

Nowadays, smartphones play an ubiquitous role in accessing and processing information, most of them having a myriad of integrated sensors that makes them capable of generating information with high accuracy and precision. The monitoring of physical exercises presents itself as one of the new trends, made possible by the use of devices like smartphones. Motion sensors such as the accelerometer enable live motion measurement. This paper intends to study this issue and develop an application for the Android operating system, which takes advantage of sensors embedded in smartphones and web technologies, with the goal to classify multiple physical activities. The developed solution is based on client-server architecture. The client application performs data acquisition, visualization and recording of the signal obtained by the smartphone's accelerometer and the server application receives the information acquired by the client, processes it and classifies it. In order for the application to be able to classify multiple movements throughout the activity performed by the user, an extensive analysis of the acquired signals was carried out to understand their most distinctive features. We used a supervised approach with the goal of reviewing the best techniques that should be useful for achieving the classification with the lowest error. For the signals acquisition the smartphone was positioned along the waist, inside the right front pocket in an attempt to simulate conditions as naturally as possible. The study explored features extracted in both the time and frequency domain, and parametric and non-parametric classifiers. Preliminary results demonstrate that the classification of activities can be done with remarkable accuracy ( > 95%).

show abstract

“…In recent years, there is a substantial amount of literature aimed at classifying human activity through the use of sensory information such as acceleration and angular velocity [8][9][10][11][12][13][14][15]. Some of the work focuses specifically on detecting fall events [16][17][18], and others investigated fall-detection using walking aids [19][20][21].…”

Section: Introductionmentioning

confidence: 99%

Design and implementation of an instrumented cane for gait recognition

Wade

Beccani

Myszka

et al. 2015

2015 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

Abstract-Independent mobility is an important aspect of an individual's life and must sometimes be augmented by use of an assistive device such as a wheeled walker or cane following a fall, injury, or functional decline. Physical therapists perform functional gait assessments to gauge the probability of an individual experiencing a fall and often recommend use of a walker, cane, or walking stick to decrease fall risk. Our team has developed a clinical assessment tool centered on a standard walking cane embedded system that can enhance a therapist's observation-based gait assessment with use of additional objective and quantitative data. This system can be utilized to detect timing and speed of cane placement, angular acceleration of the cane, and amounts of weight borne on the cane . This system is designed to assist physical therapists at the basic level in collection of objective data during gait analysis, to facilitate appropriate assistive gait device prescription, to provide patients and therapists feedback during gait training, and to reduce wrist and shoulder injuries with cane usage. However, more importantly, using the plethora of objective data that can be obtained from this cane, automated gait analysis and gait pattern classification can be performed to understand a patient's walking performance.

show abstract

Frequency based classification of activities using accelerometer data

Cited by 24 publications

References 5 publications

Real-Time Processing and Analysis for Activity Classification to Enhance Wearable Wireless ECG

Real-Time Processing and Analysis for Activity Classification to Enhance Wearable Wireless ECG

Classification of Physical Activities Using a Smartphone: Evaluation Study Using Multiple Users

Design and implementation of an instrumented cane for gait recognition

Contact Info

Product

Resources

About