Human Activity Recognition on Smartphones Using a Bidirectional LSTM Network

107

In recent years, human activity recognition has become a hot topic inside the scientific community. The reason to be under the spotlight is its direct application in multiple domains, like healthcare or fitness. Additionally, the current worldwide use of smartphones makes it particularly easy to get this kind of data from people in a non-intrusive and cheaper way, without the need for other wearables. In this paper, we introduce our orientation-independent, placement-independent and subject-independent human activity recognition dataset. The information in this dataset is the measurements from the accelerometer, gyroscope, magnetometer, and GPS of the smartphone. Additionally, each measure is associated with one of the four possible registered activities: inactive, active, walking and driving. This work also proposes asupport vector machine (SVM) model to perform some preliminary experiments on the dataset. Considering that this dataset was taken from smartphones in their actual use, unlike other datasets, the development of a good model on such data is an open problem and a challenge for researchers. By doing so, we would be able to close the gap between the model and a real-life application.

Section: Related Workmentioning

confidence: 75%

“…For example, in [20,21] we can see this kind of models and with particularly good results. In fact, in [20] they implemented a modification of LSTMs which are called Bi-LSTMs (bidirectional LSTMs). What makes this modification special is that these models can also learn from the future, throwing accuracies of around 95%.…”

Section: Related Workmentioning

confidence: 75%

A Public Domain Dataset for Real-Life Human Activity Recognition Using Smartphone Sensors

García-González

Rivero

Fernández-Blanco

et al. 2020

107

“…The categorized accuracy and other advanced metrics were considered to evaluate the sensor-based HAR of five LSTM-based DL architectures. A publicly available dataset of previous studies [ 25 , 31 , 36 , 45 , 54 ] was applied to compare the generality of DL algorithms with the 10-fold cross-validation technique.…”

Section: Discussion Of Resultsmentioning

confidence: 99%

“…The Vanilla LSTM is proposed to overcome the HAR problem [ 43 ]. Later, different architecture-based LSTM networks were proposed to solve the HAR problem such as deep LSTM networks called stacked-LSTMs [ 29 ], hybrid LSTM networks called CNN-LSTM; combining the CNN with the LSTM [ 31 ], mixed LSTM networks called ConvLSTM [ 44 ], and bidirectional LSTM networks called Bidir-LSTM [ 28 , 45 ].…”

Section: Theoretical Backgroundmentioning

confidence: 99%

LSTM Networks Using Smartphone Data for Sensor-Based Human Activity Recognition in Smart Homes

Mekruksavanich

Jitpattanakul

2021

198

Human Activity Recognition (HAR) employing inertial motion data has gained considerable momentum in recent years, both in research and industrial applications. From the abstract perspective, this has been driven by an acceleration in the building of intelligent and smart environments and systems that cover all aspects of human life including healthcare, sports, manufacturing, commerce, etc. Such environments and systems necessitate and subsume activity recognition, aimed at recognizing the actions, characteristics, and goals of one or more individuals from a temporal series of observations streamed from one or more sensors. Due to the reliance of conventional Machine Learning (ML) techniques on handcrafted features in the extraction process, current research suggests that deep-learning approaches are more applicable to automated feature extraction from raw sensor data. In this work, the generic HAR framework for smartphone sensor data is proposed, based on Long Short-Term Memory (LSTM) networks for time-series domains. Four baseline LSTM networks are comparatively studied to analyze the impact of using different kinds of smartphone sensor data. In addition, a hybrid LSTM network called 4-layer CNN-LSTM is proposed to improve recognition performance. The HAR method is evaluated on a public smartphone-based dataset of UCI-HAR through various combinations of sample generation processes (OW and NOW) and validation protocols (10-fold and LOSO cross validation). Moreover, Bayesian optimization techniques are used in this study since they are advantageous for tuning the hyperparameters of each LSTM network. The experimental results indicate that the proposed 4-layer CNN-LSTM network performs well in activity recognition, enhancing the average accuracy by up to 2.24% compared to prior state-of-the-art approaches.

“…Dynamic time warping [23] 89.00 Handcrafted features + SVM [24] 89.00 Convolutional neural network [25] 90.89 Hidden Markov models [26] 91.76 PCA + SVM [27] 91.82 Stacked autoencoders + SVM [28] 92.16 Hierarchical continuous HMM [28] 93.18 Bidir-LSTM network [29] 93.79 A multi-layer parallel LSTM network [30] 94.34 Convolutional neural network [31] 94.79 Convolutional neural network [16] 95.31 Fully convolutional network [32] 96.32 Genetic algorithm to optimize feature vector [33] 96.38 Bidirectional LSTM network [34] 92.67 CNN [35] 94.00 Hierarchical deep learning model [36] 97.95 Our method (LDA + SVM)…”

Section: Recognition Accuracy (%)mentioning

confidence: 99%

An Activity-Aware Sampling Scheme for Mobile Phones in Activity Recognition

Chen

Huang

2020

In recent years, sensors in smartphones have been widely used in applications, e.g., human activity recognition (HAR). However, the power of smartphone constrains the applications of HAR due to the computations. To combat it, energy efficiency should be considered in the applications of HAR with smartphones. In this paper, we improve energy efficiency for smartphones by adaptively controlling the sampling rate of the sensors during HAR. We collect the sensor samples, depending on the activity changing, based on the magnitude of acceleration. Besides that, we use linear discriminant analysis (LDA) to select the feature and machine learning methods for activity classification. Our method is verified on the UCI (University of California, Irvine) dataset; and it achieves an overall 56.39% of energy saving and the recognition accuracy of 99.58% during the HAR applications with smartphone.