Show simple item record

dc.contributor.authorReinders A
dc.contributor.authorVisser T
dc.contributor.authorRoos D
dc.contributor.authorVink PJF de
dc.contributor.authorLuinge HJ
dc.date.accessioned2012-12-12T15:41:13Z
dc.date.available2012-12-12T15:41:13Z
dc.date.issued1991-09-30
dc.identifier421504001
dc.identifier.urihttp://hdl.handle.net/10029/257504
dc.description.abstractAbstract niet beschikbaar
dc.description.abstractThis document describes how neural networks can be trained to classify and recognize infrared spectra. Backpropagation was used as the neural network type. The effect of noise on the recognition capabilities of a network has been investigated by generating 150 spectra with various noise levels out of 3 standard spectra. The trained network appeared to be capable of recognizing spectra correctly up to a noise level of 70%. Recognition appears to be correct up to a noise lebel of 70%. The classifying capabilities of backpropagation of spectra have been studied by training a network with 30 spectra, equally divided over three classes. Fourteen other spectra were used as a control set. Only one spectrum was found to be incorrectly classified. The preliminary conclusion is that neural networks are a useful addition to standard pattern matching techniques, especially for recognizing visual aspects.
dc.description.sponsorshipRIVM
dc.format.extent27 p
dc.language.isonl
dc.relation.ispartofRIVM Rapport 421504001
dc.relation.urlhttp://www.rivm.nl/bibliotheek/rapporten/421504001.html
dc.subject20nl
dc.subject91-3nl
dc.subjectinfraroodspectranl
dc.subjectneurale netwerkennl
dc.subjectbackpropagationnl
dc.subjectpatroonherkenningnl
dc.subjectinfrared spectranl
dc.subjectneural networksnl
dc.subjectpattern recognitionnl
dc.titleHerkennen van visuele aspecten van infraroodspectra met neurale netwerkennl
dc.title.alternativeRecognizing visual aspects of infrared spectra with neural networksen
dc.typeReport
dc.date.updated2012-12-12T15:41:14Z
html.description.abstractAbstract niet beschikbaar
html.description.abstractThis document describes how neural networks can be trained to classify and recognize infrared spectra. Backpropagation was used as the neural network type. The effect of noise on the recognition capabilities of a network has been investigated by generating 150 spectra with various noise levels out of 3 standard spectra. The trained network appeared to be capable of recognizing spectra correctly up to a noise level of 70%. Recognition appears to be correct up to a noise lebel of 70%. The classifying capabilities of backpropagation of spectra have been studied by training a network with 30 spectra, equally divided over three classes. Fourteen other spectra were used as a control set. Only one spectrum was found to be incorrectly classified. The preliminary conclusion is that neural networks are a useful addition to standard pattern matching techniques, especially for recognizing visual aspects.


This item appears in the following Collection(s)

Show simple item record