Human activity recognition from inertial sensor time-series using batch normalized deep LSTM recurrent networks
| Autor: | Tahmina Zebin, Matthew Sperrin, Alexander J. Casson, Niels Peek |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2018 |
| Předmět: |
Normalization (statistics)
ResearchInstitutes_Networks_Beacons/MICRA Computer science Movement Feature extraction 02 engineering and technology 01 natural sciences Machine Learning Activity recognition Discriminative model 0202 electrical engineering electronic engineering information engineering Humans Human Activities Artificial neural network business.industry 010401 analytical chemistry Pattern recognition 0104 chemical sciences Support vector machine Recurrent neural network Manchester Institute for Collaborative Research on Ageing 020201 artificial intelligence & image processing Neural Networks Computer Artificial intelligence business Algorithms |
| Zdroj: | EMBC Zebin, T, Sperrin, M, Peek, N & Casson, A 2018, Human activity recognition from inertial sensor time-series using batch normalized deep LSTM recurrent networks . in IEEE EMBC . https://doi.org/10.1109/embc.2018.8513115 |
| Popis: | In recent years machine learning methods for human activity recognition have been found very effective. These classify discriminative features generated from raw input sequences acquired from body-worn inertial sensors. However, it involves an explicit feature extraction stage from the raw data, and although human movements are encoded in a sequence of successive samples in time most state-of-theart machine learning methods do not exploit the temporalcorrelations between input data samples. In this paper we present a Long-Short Term Memory (LSTM) deep recurrent neural network for the classification of six daily life activities from accelerometer and gyroscope data. Results show that ourLSTM can processes featureless raw input signals, and achieves 92% average accuracy in a multi-class-scenario. Further, we show that this accuracy can be achieved with almost four times fewer training epochs by using a batch normalization approach. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|