INTELLIGENT MACHINE VISION SYSTEM FOR ROAD TRAFFIC SIGN RECOGNITION

Main Authors: Aryuanto, Aryuanto, Yamada, Koich, Limpraptono, F. Yudi
Format: Article PeerReviewed Book
Bahasa: eng
Terbitan: Universitas Islam Indonesia , 2008
Subjects:
Online Access: http://eprints.itn.ac.id/3713/2/1.%20paper%20teknoin%202008.pdf
http://eprints.itn.ac.id/3713/1/2.%20cek%20plagiasi%20teknoin%202008.pdf
http://eprints.itn.ac.id/3713/3/3.%20peer%20review%20teknoin%202008.pdf
http://eprints.itn.ac.id/3713/
Daftar Isi:
  • Abstract We proposed an intelligent machine vision system to recognize traffic signs captured from a video camera installed in a vehicle. By recognizing the traffic signs automatically, it helps the driver to recognize the signs properly when drivig, to avoid accidents caused by mis-recognized the traffic signs.The system is divided into two stages : detection stage to localize signs from a whole image, and classification stage that classifies the detected sign into one of the reference signs. A geometric fragmentation technique, a method somewhat similar to Genetic Algorithm (GA) is employed to detect circular sign. Then a ring partitioned method that divides an image into several ring-shaped areas is used to classify the signs. From the experimental results, the proposed techniques are able to recognize traffic sign images under the problems of illumination changes, rotation, and occlusion efficiently. Keywords : Machine vision, traffic sign recognition, geometric fragmentation, ring partitioned matching.