Neural Object Classification Using Ultrasonic Spectrum Analysis
Klahold J, Jürgens H, Rückert U (2003)
In: Proceedings of the 2nd International Symposium on Autonomous Minirobots for Research and Edutainment (AMiRE). Brisbane, Australia: 219-228.
Konferenzbeitrag
| Veröffentlicht | Englisch
Download
Es wurden keine Dateien hochgeladen. Nur Publikationsnachweis!
Autor*in
Klahold, Jürgen;
Jürgens, Hartmut;
Rückert, UlrichUniBi
Abstract / Bemerkung
This paper describes how reflected broadband sound signals are marked by interference
phenomena if the surface of the ensonified object is structured. For an efficient
extraction of features of the signal that relate to the surface structure of the object, the calculation
of the cepstrum is introduced. A cylindrical test object is presented, which shows
an angle independent and an angle dependent structure. This allows to specify the accuracy
of the discrimination of structure sizes, that is based on a selected part of the cepstrum. In
addition, the object can be used as a landmark for ultrasonic sensing. Classification of the
cepstra is done by the statistical ‘One Nearest Neighbour’ (1NN) method as well as by a
‘Kohonen Self-Organising Feature Map’ (SOM). The results show, that changes in structure
size down to 0.1mm are detectable.
Erscheinungsjahr
2003
Titel des Konferenzbandes
Proceedings of the 2nd International Symposium on Autonomous Minirobots for Research and Edutainment (AMiRE)
Seite(n)
219-228
Page URI
https://pub.uni-bielefeld.de/record/2288625
Zitieren
Klahold J, Jürgens H, Rückert U. Neural Object Classification Using Ultrasonic Spectrum Analysis. In: Proceedings of the 2nd International Symposium on Autonomous Minirobots for Research and Edutainment (AMiRE). Brisbane, Australia; 2003: 219-228.
Klahold, J., Jürgens, H., & Rückert, U. (2003). Neural Object Classification Using Ultrasonic Spectrum Analysis. Proceedings of the 2nd International Symposium on Autonomous Minirobots for Research and Edutainment (AMiRE), 219-228. Brisbane, Australia.
Klahold, Jürgen, Jürgens, Hartmut, and Rückert, Ulrich. 2003. “Neural Object Classification Using Ultrasonic Spectrum Analysis”. In Proceedings of the 2nd International Symposium on Autonomous Minirobots for Research and Edutainment (AMiRE), 219-228. Brisbane, Australia.
Klahold, J., Jürgens, H., and Rückert, U. (2003). “Neural Object Classification Using Ultrasonic Spectrum Analysis” in Proceedings of the 2nd International Symposium on Autonomous Minirobots for Research and Edutainment (AMiRE) (Brisbane, Australia), 219-228.
Klahold, J., Jürgens, H., & Rückert, U., 2003. Neural Object Classification Using Ultrasonic Spectrum Analysis. In Proceedings of the 2nd International Symposium on Autonomous Minirobots for Research and Edutainment (AMiRE). Brisbane, Australia, pp. 219-228.
J. Klahold, H. Jürgens, and U. Rückert, “Neural Object Classification Using Ultrasonic Spectrum Analysis”, Proceedings of the 2nd International Symposium on Autonomous Minirobots for Research and Edutainment (AMiRE), Brisbane, Australia: 2003, pp.219-228.
Klahold, J., Jürgens, H., Rückert, U.: Neural Object Classification Using Ultrasonic Spectrum Analysis. Proceedings of the 2nd International Symposium on Autonomous Minirobots for Research and Edutainment (AMiRE). p. 219-228. Brisbane, Australia (2003).
Klahold, Jürgen, Jürgens, Hartmut, and Rückert, Ulrich. “Neural Object Classification Using Ultrasonic Spectrum Analysis”. Proceedings of the 2nd International Symposium on Autonomous Minirobots for Research and Edutainment (AMiRE). Brisbane, Australia, 2003. 219-228.
Google Scholar