Copy move forgery detection based on LBP

ULUTAŞ G., ULUTAŞ M., NABIYEV V.

21st Signal Processing and Communications Applications Conference (SIU), CYPRUS, 24 - 26 Nisan 2013 identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası:
  • Doi Numarası: 10.1109/siu.2013.6531569
  • Basıldığı Ülke: CYPRUS
  • Karadeniz Teknik Üniversitesi Adresli: Evet

Özet

Increase on the availability of the image editing tools makes it easy to create digital forgeries. More popular forgery in image manipulation is the copy move forgery where part of an image is copied into another location to hide an object. The methods that are used to detect the copy move forgery in images have a drawback. Length of the feature vector that represents an image block affects the time complexity of the detection algorithm. In this work, we use the local binary patterns (LBP) to represent an image block. LBP is used to reduce the dimension of the features. Experimental results also show that the method can detect the duplicated regions even when an image was distorted by JPEG compression or Gaussian blurring.