CoChat

On the Generality of Codebook Approach for Sensor-Based Human Activity Recognition

Open access
Shirahama, Kimiaki, Grzegorzek, Marcin · 2017 · Electronics

The DOI matches its Crossref record; 46 of the 47 references it cites were found in a registry or an index.

Abstract

With the recent spread of mobile devices equipped with different sensors, it is possible to continuously recognise and monitor activities in daily life. This sensor-based human activity recognition is formulated as sequence classification to categorise sequences of sensor values into appropriate activity classes. One crucial problem is how to model features that can precisely represent characteristics of each sequence and lead to accurate recognition. It is laborious and/or difficult to hand-craft such features based on prior knowledge and manual investigation about sensor data. To overcome this, we focus on a feature learning approach that extracts useful features from a large amount of data. In particular, we adopt a simple but effective one, called codebook approach, which groups numerous subsequences collected from sequences into clusters. Each cluster centre is called a codeword and represents a statistically distinctive subsequence. Then, a sequence is encoded as a feature expressing the distribution of codewords. The extensive experiments on different recognition tasks for physical, mental and eye-based activities validate the effectiveness, generality and usability of the codebook approach.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.3390/electronics6020044

Cites

46 of 46 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 An overview of our codebook-based human activity recognition method. Here, sequences of respiration rates are used to illustrate ( a ) how to construct a codebook; ( b ) how to extract the feature of a sequence based on the extracted codebook; and ( c ) how such features are utilised to train/test a classifier that distinguishes the mental activity 'anger' from the others.
Figure 1 An overview of our codebook-based human activity recognition method. Here, sequences of respiration rates are used to illustrate ( a ) how to construct a codebook; ( b ) how to extract the feature of a sequence based on the extracted codebook; and ( c ) how such features are utilised to train/test a classifier that distinguishes the mental activity 'anger' from the others.CC-BY · From the PDF
Figure 2 Examples of four types of physiological sequences in the dataset [11]. From the left, the first, second, third and last sequences represent values of Blood Volume Pressure (BVP), Galvanic Skin Response (GSR), RESpiration (RES) and ElectroMyoGram (EMG), respectively. Note that these sequences are normalised to have the mean zero and the variance one. In addition, to emphasise value changes, each type of sequence is plotted using the vertical axis with a different range.
Figure 2 Examples of four types of physiological sequences in the dataset [11]. From the left, the first, second, third and last sequences represent values of Blood Volume Pressure (BVP), Galvanic Skin Response (GSR), RESpiration (RES) and ElectroMyoGram (EMG), respectively. Note that these sequences are normalised to have the mean zero and the variance one. In addition, to emphasise value changes, each type of sequence is plotted using the vertical axis with a different range.CC-BY · From the PDF
Figure 3 ( a ) An example of an accuracy distribution based on different combinations of w and N ; ( b ) Performance comparison between hard and soft assignments where accuracy distributions are summarised into box plots. Box plots marked as BVP, GSR, RES, EMG, RES_sp and Fusion indicate accuracy distributions obtained for blood volume pressure sequences, galvanic skin response sequences, respiration sequences, electromyogram sequences, respiration spectra, and their fusion, respectively.
Figure 3 ( a ) An example of an accuracy distribution based on different combinations of w and N ; ( b ) Performance comparison between hard and soft assignments where accuracy distributions are summarised into box plots. Box plots marked as BVP, GSR, RES, EMG, RES_sp and Fusion indicate accuracy distributions obtained for blood volume pressure sequences, galvanic skin response sequences, respiration sequences, electromyogram sequences, respiration spectra, and their fusion, respectively.CC-BY · From the PDF
Figure 4 Transition of accuracies depending on fused features. The accuracies at the upper part result from fusing features expressed by the 'fusion indicator' at the bottom part. The first, second, third, forth and last rows in the fusion indicator indicate the usages of features extracted from Blood Volume Pressure (BVP) sequences, Galvanic Skin Response (GSR) sequences, RESpiration (RES), ElectroMyoGram (EMG) sequences and RESpiration spectra (RES_sp). If a feature is used, the corresponding element is filled with black, otherwise it is white.
Figure 4 Transition of accuracies depending on fused features. The accuracies at the upper part result from fusing features expressed by the 'fusion indicator' at the bottom part. The first, second, third, forth and last rows in the fusion indicator indicate the usages of features extracted from Blood Volume Pressure (BVP) sequences, Galvanic Skin Response (GSR) sequences, RESpiration (RES), ElectroMyoGram (EMG) sequences and RESpiration spectra (RES_sp). If a feature is used, the corresponding element is filled with black, otherwise it is white.CC-BY · From the PDF
Figure 5 Examples of codewords extracted by our method. Codewords marked as BVP, GSR, RES, EMG and RES_sp are extracted from Blood Volume Pressure (BVP) sequences, Galvanic Skin Response (GSR) sequences, RESpiration (RES), ElectroMyoGram (EMG) sequences and RESpiration spectra (RES_sp), respectively.
Figure 5 Examples of codewords extracted by our method. Codewords marked as BVP, GSR, RES, EMG and RES_sp are extracted from Blood Volume Pressure (BVP) sequences, Galvanic Skin Response (GSR) sequences, RESpiration (RES), ElectroMyoGram (EMG) sequences and RESpiration spectra (RES_sp), respectively.CC-BY · From the PDF
Figure 6 Examples of three-dimensional sequences from six accelerometers for the activity 'jumping in place' in [22].
Figure 6 Examples of three-dimensional sequences from six accelerometers for the activity 'jumping in place' in [22].CC-BY · From the PDF
Image 7
CC-BY · From the PDF
Figure 8 Examples of horizontal and vertical ElectroOculoGram (hEOG and vEOG) sequences in [10].
Figure 8 Examples of horizontal and vertical ElectroOculoGram (hEOG and vEOG) sequences in [10].CC-BY · From the PDF

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 46 references

Research works printed but not linked: 3

  1. Han, J., Kamber, M., and Pei, J. (2011). Data Mining: Concepts and Techniques, Morgan Kaufmann. [3rd ed.].
  2. Murphy, K.P. (2012). Machine Learning: A Probabilistic Perspective, The MIT Press.
  3. Vapnik, V.N. (1998). Statistical Learning Theory, Wiley-Interscience.

Cited by

Metadata

Authors
Shirahama, K., & Grzegorzek, M.
Affiliations
  • University of Siegen (DE)
  • University of Economics in Katowice (PL)
Corresponding author
Shirahama, Kimiaki
Venue
Electronics
Publisher
MDPI AG
Published
2017-06-01
Volume
6
Issue
2
Pages
44
ISSN
2079-9292 (ISSN-L)
OpenAlex
W2621021458
Cited by
29
References
47
Citations by year
2025: 1 · 2024: 6 · 2023: 4 · 2022: 1 · 2021: 2 · 2020: 5 · 2019: 7 · 2018: 2
FWCI
1.59
Type
journal article
Language
en
Open access
Locations
Electronics (publishedVersion, cc-by)
Funders
Bundesministerium für Bildung und Forschung (16SV7223K)
Topics
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
  • Physical Sciences › Computer Science › Signal Processing › Time Series Analysis and Forecasting
  • Physical Sciences › Engineering › Biomedical Engineering › Gait Recognition and Analysis
Keywords
human activity recognition, sensor-based activity recognition, representation learning, sequence classification, code words, mobile devices, physical activity, mental activity, activity monitoring

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.