Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Improving image classification robustness using predictive data augmentation
KTH, School of Industrial Engineering and Management (ITM), Machine Design (Dept.), Mechatronics. Scania CV AB, Södertälje, 15187, Sweden.
KTH, School of Industrial Engineering and Management (ITM), Machine Design (Dept.), Mechatronics. Scania CV AB, Södertälje, 15187, Sweden.
2018 (English)In: Workshops: ASSURE, DECSoS, SASSUR, STRIVE, and WAISE 2018 co-located with 37th International Conference on Computer Safety, Reliability and Security, SAFECOMP 2018, Springer, 2018, p. 548-561Conference paper, Published paper (Refereed)
Abstract [en]

Safer autonomous navigation might be challenging if there is a failure in sensing system. Robust classifier algorithm irrespective of camera position, view angles, and environmental condition of an autonomous vehicle including different size & type (Car, Bus, Truck, etc.) can safely regulate the vehicle control. As training data play a crucial role in robust classification of traffic signs, an effective augmentation technique enriching the model capacity to withstand variations in urban environment is required. In this paper, a framework to identify model weakness and targeted augmentation methodology is presented. Based on off-line behavior identification, exact limitation of a Convolutional Neural Network (CNN) model is estimated to augment only those challenge levels necessary for improved classifier robustness. Predictive Augmentation (PA) and Predictive Multiple Augmentation (PMA) methods are proposed to adapt the model based on acquired challenges with a high numerical value of confidence. We validated our framework on two different training datasets and with 5 generated test groups containing varying levels of challenge (simple to extreme). The results show impressive improvement by$$\approx $$ 5–20% in overall classification accuracy thereby keeping their high confidence.

Place, publisher, year, edition, pages
Springer, 2018. p. 548-561
Series
Lecture Notes in Computer Science, ISSN 0302-9743, E-ISSN 1611-3349 ; 11094
Keywords [en]
Convolutional neural network, Predictive augmentation, Real-time challenges, Safety-risk assessment, Traffic sign classification
National Category
Transport Systems and Logistics
Identifiers
URN: urn:nbn:se:kth:diva-238414DOI: 10.1007/978-3-319-99229-7_49ISI: 000458807000049Scopus ID: 2-s2.0-85053899561ISBN: 9783319992280 (print)OAI: oai:DiVA.org:kth-238414DiVA, id: diva2:1261020
Conference
Workshops: ASSURE, DECSoS, SASSUR, STRIVE, and WAISE 2018 co-located with 37th International Conference on Computer Safety, Reliability and Security, SAFECOMP 2018, Västerås, Sweden, 18 September 2018 through 21 September 2018
Note

QC 20181106. QC 20191105

Available from: 2018-11-06 Created: 2018-11-06 Last updated: 2019-11-05Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textScopus

Authority records BETA

Gustavsson, Joakim

Search in DiVA

By author/editor
Nyberg, MattiasGustavsson, Joakim
By organisation
Mechatronics
Transport Systems and Logistics

Search outside of DiVA

GoogleGoogle Scholar

doi
isbn
urn-nbn

Altmetric score

doi
isbn
urn-nbn
Total: 61 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf