EDLINES: REAL-TIME LINE SEGMENT DETECTION BY EDGE DRAWING (ED)

Akinlar C., TOPAL C.

18th IEEE International Conference on Image Processing (ICIP), Brussels, Belçika, 11 - 14 Eylül 2011 identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/icip.2011.6116138
  • Basıldığı Şehir: Brussels
  • Basıldığı Ülke: Belçika
  • Anahtar Kelimeler: Real-Time Line Segment Detection, Edge Drawing Algorithm, LSD, Helmholtz Principle, NFA
  • Anadolu Üniversitesi Adresli: Evet

Özet

We propose a linear time line segment detector that gives accurate results, requires no parameter tuning, and runs up to 11 times faster than the fastest known line segment detection algorithm in the literature; namely, the LSD by Gioi et al. The proposed algorithm also includes a line validation step due to the Helmholtz principle, which lets it control the number of false detections. Our detector makes use of the clean, contiguous (connected) chain of edge pixels produced by our novel edge detector, the Edge Drawing (ED) algorithm; hence the name EDLines. With its accurate results and blazing speed, EDLines will be very suitable for the next generation real-time computer vision applications.