Dalarna University's logo and link to the university's website

du.sePublications
Planned maintenance
A system upgrade is planned for 10/12-2024, at 12:00-13:00. During this time DiVA will be unavailable.
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • chicago-author-date
  • chicago-note-bibliography
  • 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
Adaptive Colour Segmentation for Traffic Sign Recognition using SOM
Dalarna University, School of Technology and Business Studies, Computer Engineering.
2010 (English)Independent thesis Advanced level (degree of Master (Two Years))Student thesis
Abstract [en]

This report proposed a new adaptive approach for segmentation of traffic signs in different conditions and countries for the Intelligent Transportation Systems (ITS) using Kohonen’s Self-Organizing Maps (SOM). HSV colour space, which is considered as stable, is used for the training of the SOM network. The main concentration of the report is red, blue and yellow colours. Search boundary values for these colours are obtained after research on the traffic sign images in different conditions. The dynamic segmentation is done by BMU’s obtained with SOM for the test image close to the desired colours in the search boundary in traffic sign images and discarding the unwanted areas. The image segmented by Adaptive Segmentation Algorithm is passed to the Region Growing Algorithm (RGA) which removes additional unwanted areas from the segmented region. Multistage Median Filter is then applied to remove the salt and pepper noise from the segmented image to give the final segmented image. The Adaptive algorithm designed is giving promising results for the segmentation of these colours in the traffic signs for different environmental conditions such as fog, highlight, bad lighting, rainfall, snowfall and other conditions such as faded and blurred images. A good improvement of 73% is observed in the faded signs as compared to 53.3 % of the Shadow and Highlight Invariant Algorithm. The adaptability of the system is evident from the segmentation of the traffic sign images from various countries where the result is 100 % for 6 out of 9 countries. The algorithm is giving very good results for the blue and yellow colours of traffic signs as well reflecting its powerful nature.

Place, publisher, year, edition, pages
Borlänge, 2010.
Keywords [en]
Self Organizing Maps, Intelligent Transportation System, Traffic Signs, Colour Segmentation, Best Matching Units
Identifiers
URN: urn:nbn:se:du-4801OAI: oai:dalea.du.se:4801DiVA, id: diva2:518878
Uppsok
Technology
Supervisors
Available from: 2010-06-09 Created: 2010-06-09 Last updated: 2012-04-24Bibliographically approved

Open Access in DiVA

No full text in DiVA

By organisation
Computer Engineering

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 527 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • chicago-author-date
  • chicago-note-bibliography
  • 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