Motion deblurring utilizing the variable speed egomotion of camera and comparison based on corner feature points Deǧişken Hizli Kamera Öz Hareketinden Kaynaklanan Hareket Bulanikliǧinin Giderilmesi ve Köşe Öznitelik Noktalari Temelli Karşilaştirma

GÜLTEKİN G. K., SARANLI A.

25th Signal Processing and Communications Applications Conference, SIU 2017, Antalya, Türkiye, 15 - 18 Mayıs 2017 identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/siu.2017.7960445
  • Basıldığı Şehir: Antalya
  • Basıldığı Ülke: Türkiye
  • Anahtar Kelimeler: corner detection, feature extraction, motion blur, motion deblurring
  • Orta Doğu Teknik Üniversitesi Adresli: Evet

Özet

In this study, the motion blur caused by the variable speed egomotion of camera is deblurred using multiple image frames and the obtained results are compared based on the detection performance of corner features. A linear, uniform motion blur data set is collected which is suitable for testing the multi-image deblurring method. The proposed method is tested on the dataset together with two baseline motion deblurring methods in the literature and compared based on the corner feature detection performance. The obtained results show that the proposed multi-image deblurring method can achieve a higher true positive rate than the compared methods.