Performance Analysis of Smartphone-Sensor Behavior for Human Activity Recognition

Chen, Yufei; Shen, Chao

doi:10.1109/access.2017.2676168

Cited by 254 publications

(188 citation statements)

References 33 publications

Supporting

Mentioning

166

Contrasting

Unclassified

Order By: Relevance

“…In order to the recognition of going up stairs, going down stairs, walking, jogging, and jumping activities, the authors of [29] used the KNN, Random Forests and SVM methods with accelerometer data to identify accurately the activities. The features extracted are mean, standard deviation, maximum, minimum, correlation, interquartile range, Dynamic time warping distance (DTW), FFT coefficients and wavelet energy, reporting an accuracy around 95% [29].…”

Section: Related Workmentioning

confidence: 99%

“…The features implemented are sum of all magnitude of the vectors, sum of all magnitude of the vectors excluding the gravity, maximum and minimum value of acceleration in gravity vector direction, mean of absolute deviation of acceleration in gravity vector direction, and gravity vector changing angle, reporting that the results have a sensitivity of 96.67% and specificity of 95% [10].In order to the recognition of going up stairs, going down stairs, walking, jogging, and jumping activities, the authors of [29] used the KNN, Random Forests and SVM methods with accelerometer data to identify accurately the activities. The features extracted are mean, standard deviation, maximum, minimum, correlation, interquartile range, Dynamic time warping distance (DTW), FFT coefficients and wavelet energy, reporting an accuracy around 95% [29].In [30], the authors proposed a solution that uses SVM, J48 decision tree and Random Forest methods with accelerometer data for the recognition of Sitting, Standing, Walking, and running. The solution extracts several features, such as average of peak values, average of peak rising time, average of peak fall time, average time per sample, and average time between peaks, reporting an accuracy of 98.8283% [30].Other features are used by other authors in [31], including the mean, the median, the standard deviation, the skewness, the kurtosis, the minimum, the maximum, and the slope for each axis and for the absolute value of accelerometer.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Pattern recognition techniques for the identification of Activities of Daily Living using mobile device accelerometer

Pires¹,

García²,

Pombo³

et al. 2019

Preprint

View full text Add to dashboard Cite

This paper focuses on the recognition of Activities of Daily Living (ADL) applying pattern recognition techniques to the data acquired by the accelerometer available in the mobile devices. The recognition of ADL is composed by several stages, including data acquisition, data processing, and artificial intelligence methods. The artificial intelligence methods used are related to pattern recognition, and this study focuses on the use of Artificial Neural Networks (ANN). The data processing includes data cleaning, and the feature extraction techniques to define the inputs for the ANN. Due to the low processing power and memory of the mobile devices, they should be mainly used to acquire the data, applying an ANN previously trained for the identification of the ADL. The main purpose of this paper is to present a new method implemented with ANN for the identification of a defined set of ADL with a reliable accuracy. This paper also presents a comparison of different types of ANN in order to choose the type for the implementation of the final method. Results of this research probes that the best accuracies are achieved with Deep Learning techniques with an accuracy higher than 80%. Following the previous research studies [4][5][6], the recognition of the ADL is composed by several steps, such as the data acquisition, the data processing, composed by data cleaning, data imputation, and feature extraction, the data fusion, and the artificial intelligence methods for the concrete identification of the ADL. However, this study only uses the accelerometer sensor, removing some steps of the proposed architecture. Based on the assumption that the sensor was always acquiring the data, the final steps used are the data acquisition, the data cleaning, and the application of the artificial intelligence methods.During the last years, the recognition of ADL has been studies by several authors [7][8][9][10][11][12], verifying that the Artificial Neural Networks (ANN) are widely used. This paper proposes the creation of a method for the recognition of ADL using accelerometer, comparing three types of ANN, such as Multilayer Perception (MLP) with Backpropagation, Feedforward neural network with Backpropagation, and Deep Learning, in order to verify the method that achieves the best accuracy in the recognition of running, walking, going upstairs, going downstairs, and standing. The datasets are composed with raw accelerometer data acquired by individual with ages ranged between 16 and 60 years old and different lifestyles, with a mobile device in the pocket. For the implementation of these types of ANN are used several datasets with different sets of features, identifying the best features to increase the accuracy of the recognition, and three Java libraries are used for the implementation of the different methods, such as Neuroph [13], Encog [14], and DeepLearning4j [15], achieving the best accuracy with Deep Learning methods.The remaining sections of this paper are organized as follows: Section 2 presents a brief literature review re...

show abstract

Section: Related Workmentioning

confidence: 99%

mentioning

confidence: 99%

Pattern recognition techniques for the identification of Activities of Daily Living using mobile device accelerometer

Pires¹,

García²,

Pombo³

et al. 2019

Preprint

View full text Add to dashboard Cite

show abstract

“…Although there are some RGB-based studies in the literature, the applications suffer in environments that are totally dark or where illumination changes are present, despite the use of a multi-camera system consisting of eight cameras installed to view a room from every possible angle and to overcome an issue with subject occlusion, for example [33]. In [38], the authors invented a system which is based on systematic performance analysis of motion-sensor-captured behavior for human activity recognition via smart phones. In sensor-based studies, authors have proposed the use of multiple accelerometers, wearable sensors, and other types of sensors.…”

Section: Using Harmentioning

confidence: 99%

“…In sensor-based studies, authors have proposed the use of multiple accelerometers, wearable sensors, and other types of sensors. In [38], the authors invented a system which is based on systematic performance analysis of motion-sensor-captured behavior for human activity recognition via smart phones. Sensory data sequences using smart phones were collected while participants in the experiment performed typical and daily activities.…”

Section: Using Harmentioning

confidence: 99%

Human activity recognition with analysis of angles between skeletal joints using a RGB‐depth sensor

et al. 2019

View full text Add to dashboard Cite

Human activity recognition (HAR) has become effective as a computer vision tool for video surveillance systems. In this paper, a novel biometric system that can detect human activities in 3D space is proposed. In order to implement HAR, joint angles obtained using an RGB-depth sensor are used as features. Because HAR is operated in the time domain, angle information is stored using the sliding kernel method.Haar-wavelet transform (HWT) is applied to preserve the information of the features before reducing the data dimension. Dimension reduction using an averaging algorithm is also applied to decrease the computational cost, which provides faster performance while maintaining high accuracy. Before the classification, a proposed thresholding method with inverse HWT is conducted to extract the final feature set.Finally, the K-nearest neighbor (k-NN) algorithm is used to recognize the activity with respect to the given data. The method compares favorably with the results using other machine learning algorithms. K E Y W O R D Sactivity recognition, dimension reduction, Haar-wavelet transform, K-nearest neighbour, RGB-D sensor | 89 İNCE Et al.

show abstract

“…It features mentioned above. It is necessary for mean to be generated first before other statistical metric features can be extracted based on computed mean (Chen and Shen, 2017). The entire TD features were later transformed to FD using the FFT method.…”

Section: Feature Extractionmentioning

confidence: 99%

A systematic literature review on activity recognition with context-awareness techniques for mitigation of disasters

Sadiq

Selamat²,

Ibrahim

2018

IJDET

View full text Add to dashboard Cite

This paper presents a systematic literature review on context-awareness using activity recognition for mitigation of disasters. Disaster preparedness and the method of mitigating it or reduction of causalities are very important in order to save human lives. The review process went through four phases which are: planning, collation, filtration of dataset to get the most relevant materials, exploration and the description based on findings was carried out. The problem with the existing studies are: technology used lack the capacity to give feedback to potential victims, inability to move 178 F.I. Sadiq et al.the technology around, high false negative alarm, low sensor quality for crowd behaviour monitoring. This study used smartphones which is handy and has potential of being carried about. The analysis of the result was presented under three major headings which are publication by year, techniques, results and findings. The result shows the research gap in the area of disaster mitigation and the further study suggests the need to extend context-aware framework for stampede prediction based on activity recognition accuracy used in the previous study.

show abstract

Performance Analysis of Smartphone-Sensor Behavior for Human Activity Recognition

Cited by 254 publications

References 33 publications

Pattern recognition techniques for the identification of Activities of Daily Living using mobile device accelerometer

Pattern recognition techniques for the identification of Activities of Daily Living using mobile device accelerometer

Human activity recognition with analysis of angles between skeletal joints using a RGB‐depth sensor

A systematic literature review on activity recognition with context-awareness techniques for mitigation of disasters

Contact Info

Product

Resources

About