Publikation:

Fast and Efficient Image Novelty Detection Based on Mean-Shifts

Lade...
Vorschaubild

Dateien

Hermann_2-1q5sij66uivsn1.pdf
Hermann_2-1q5sij66uivsn1.pdfGröße: 16.82 MBDownloads: 60

Datum

2022

Herausgeber:innen

Kontakt

ISSN der Zeitschrift

Electronic ISSN

ISBN

Bibliografische Daten

Verlag

Schriftenreihe

Auflagebezeichnung

DOI (zitierfähiger Link)
ArXiv-ID

Internationale Patentnummer

Link zur Lizenz

Angaben zur Forschungsförderung

Projekt

Open Access-Veröffentlichung
Open Access Gold
Core Facility der Universität Konstanz

Gesperrt bis

Titel in einer weiteren Sprache

Publikationstyp
Zeitschriftenartikel
Publikationsstatus
Published

Erschienen in

Sensors. MDPI. 2022, 22(19), 7674. eISSN 1424-8220. Available under: doi: 10.3390/s22197674

Zusammenfassung

Image novelty detection is a repeating task in computer vision and describes the detection of anomalous images based on a training dataset consisting solely of normal reference data. It has been found that, in particular, neural networks are well-suited for the task. Our approach first transforms the training and test images into ensembles of patches, which enables the assessment of mean-shifts between normal data and outliers. As mean-shifts are only detectable when the outlier ensemble and inlier distribution are spatially separate from each other, a rich feature space, such as a pre-trained neural network, needs to be chosen to represent the extracted patches. For mean-shift estimation, the Hotelling T2 test is used. The size of the patches turned out to be a crucial hyperparameter that needs additional domain knowledge about the spatial size of the expected anomalies (local vs. global). This also affects model selection and the chosen feature space, as commonly used Convolutional Neural Networks or Vision Image Transformers have very different receptive field sizes. To showcase the state-of-the-art capabilities of our approach, we compare results with classical and deep learning methods on the popular dataset CIFAR-10, and demonstrate its real-world applicability in a large-scale industrial inspection scenario using the MVTec dataset. Because of the inexpensive design, our method can be implemented by a single additional 2D-convolution and pooling layer and allows particularly fast prediction times while being very data-efficient.

Zusammenfassung in einer weiteren Sprache

Fachgebiet (DDC)
004 Informatik

Schlagwörter

Konferenz

Rezension
undefined / . - undefined, undefined

Forschungsvorhaben

Organisationseinheiten

Zeitschriftenheft

Zugehörige Datensätze in KOPS

Zitieren

ISO 690HERMANN, Matthias, Georg UMLAUF, Bastian GOLDLÜCKE, Matthias O. FRANZ, 2022. Fast and Efficient Image Novelty Detection Based on Mean-Shifts. In: Sensors. MDPI. 2022, 22(19), 7674. eISSN 1424-8220. Available under: doi: 10.3390/s22197674
BibTex
@article{Hermann2022-10-10Effic-58928,
  year={2022},
  doi={10.3390/s22197674},
  title={Fast and Efficient Image Novelty Detection Based on Mean-Shifts},
  number={19},
  volume={22},
  journal={Sensors},
  author={Hermann, Matthias and Umlauf, Georg and Goldlücke, Bastian and Franz, Matthias O.},
  note={Article Number: 7674}
}
RDF
<rdf:RDF
    xmlns:dcterms="http://purl.org/dc/terms/"
    xmlns:dc="http://purl.org/dc/elements/1.1/"
    xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
    xmlns:bibo="http://purl.org/ontology/bibo/"
    xmlns:dspace="http://digital-repositories.org/ontologies/dspace/0.1.0#"
    xmlns:foaf="http://xmlns.com/foaf/0.1/"
    xmlns:void="http://rdfs.org/ns/void#"
    xmlns:xsd="http://www.w3.org/2001/XMLSchema#" > 
  <rdf:Description rdf:about="https://kops.uni-konstanz.de/server/rdf/resource/123456789/58928">
    <dc:contributor>Goldlücke, Bastian</dc:contributor>
    <dc:creator>Goldlücke, Bastian</dc:creator>
    <dc:language>eng</dc:language>
    <dspace:hasBitstream rdf:resource="https://kops.uni-konstanz.de/bitstream/123456789/58928/1/Hermann_2-1q5sij66uivsn1.pdf"/>
    <dc:contributor>Umlauf, Georg</dc:contributor>
    <dc:creator>Franz, Matthias O.</dc:creator>
    <bibo:uri rdf:resource="https://kops.uni-konstanz.de/handle/123456789/58928"/>
    <dcterms:available rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2022-10-27T09:31:09Z</dcterms:available>
    <dc:contributor>Franz, Matthias O.</dc:contributor>
    <dcterms:title>Fast and Efficient Image Novelty Detection Based on Mean-Shifts</dcterms:title>
    <dcterms:hasPart rdf:resource="https://kops.uni-konstanz.de/bitstream/123456789/58928/1/Hermann_2-1q5sij66uivsn1.pdf"/>
    <dc:creator>Umlauf, Georg</dc:creator>
    <dcterms:issued>2022-10-10</dcterms:issued>
    <dc:contributor>Hermann, Matthias</dc:contributor>
    <foaf:homepage rdf:resource="http://localhost:8080/"/>
    <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/>
    <dcterms:abstract xml:lang="eng">Image novelty detection is a repeating task in computer vision and describes the detection of anomalous images based on a training dataset consisting solely of normal reference data. It has been found that, in particular, neural networks are well-suited for the task. Our approach first transforms the training and test images into ensembles of patches, which enables the assessment of mean-shifts between normal data and outliers. As mean-shifts are only detectable when the outlier ensemble and inlier distribution are spatially separate from each other, a rich feature space, such as a pre-trained neural network, needs to be chosen to represent the extracted patches. For mean-shift estimation, the Hotelling T&lt;sup&gt;2&lt;/sup&gt; test is used. The size of the patches turned out to be a crucial hyperparameter that needs additional domain knowledge about the spatial size of the expected anomalies (local vs. global). This also affects model selection and the chosen feature space, as commonly used Convolutional Neural Networks or Vision Image Transformers have very different receptive field sizes. To showcase the state-of-the-art capabilities of our approach, we compare results with classical and deep learning methods on the popular dataset CIFAR-10, and demonstrate its real-world applicability in a large-scale industrial inspection scenario using the MVTec dataset. Because of the inexpensive design, our method can be implemented by a single additional 2D-convolution and pooling layer and allows particularly fast prediction times while being very data-efficient.</dcterms:abstract>
    <dc:creator>Hermann, Matthias</dc:creator>
    <dcterms:isPartOf rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/36"/>
    <dc:rights>Attribution 4.0 International</dc:rights>
    <dcterms:rights rdf:resource="http://creativecommons.org/licenses/by/4.0/"/>
    <dspace:isPartOfCollection rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/36"/>
    <dc:date rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2022-10-27T09:31:09Z</dc:date>
  </rdf:Description>
</rdf:RDF>

Interner Vermerk

xmlui.Submission.submit.DescribeStep.inputForms.label.kops_note_fromSubmitter

Kontakt
URL der Originalveröffentl.

Prüfdatum der URL

Prüfungsdatum der Dissertation

Finanzierungsart

Kommentar zur Publikation

Allianzlizenz
Corresponding Authors der Uni Konstanz vorhanden
Internationale Co-Autor:innen
Universitätsbibliographie
Ja
Begutachtet
Ja
Diese Publikation teilen