Evolving discriminative features robust to sensor displacement for activity recognition in body area sensor networks

Created by W.Langdon from gp-bibliography.bib Revision:1.3973

@InProceedings{Forster:2009:ISSNIP,
  author =       "Kilian Forster and Pascal Brem and Daniel Roggen and 
                 Gerhard Troster",
  title =        "Evolving discriminative features robust to sensor
                 displacement for activity recognition in body area
                 sensor networks",
  booktitle =    "5th International Conference on Intelligent Sensors,
                 Sensor Networks and Information Processing, ISSNIP
                 2009",
  year =         "2009",
  month =        dec,
  pages =        "43--48",
  keywords =     "genetic algorithms, genetic programming",
  DOI =          "doi:10.1109/ISSNIP.2009.5416810",
  abstract =     "Activity and gesture recognition from body-worn
                 acceleration sensors is an important application in
                 body area sensor networks. The key to any such
                 recognition task are discriminative and variation
                 tolerant features. Furthermore good features may reduce
                 the energy requirements of the sensor network as well
                 as increase the robustness of the activity recognition.
                 We propose a feature extraction method based on genetic
                 programming. We benchmark this method using two
                 datasets and compare the results to a feature selection
                 which is typically used for obtaining a set of
                 features. With one extracted feature we achieve an
                 accuracy of 73.4percent on a fitness activity dataset,
                 in contrast to 70.1percent using one selected standard
                 feature. In a gesture based HCI dataset we achieved
                 95.0percent accuracy with one extracted feature. A
                 selection of up to five standard features achieved
                 90.6percent accuracy in the same setting. On the HCI
                 dataset we also evaluated the robustness of extracted
                 features to sensor displacement which is a common
                 problem in movement based activity and gesture
                 recognition. With one extracted features we achieved an
                 accuracy of 85.0percent on a displaced sensor position.
                 With the best selection of standard features we
                 achieved 55.2percent accuracy. The results show that
                 our proposed genetic programming feature extraction
                 method is superior to a feature selection based on
                 standard features.",
  notes =        "Also known as \cite{5416810}",
}

Genetic Programming entries for Kilian Forster Pascal Brem Daniel Roggen Gerhard Troster

Citations