Image search system with feedback

GÜNDÜZ H., KOÇAL O. H.

2016 24th Signal Processing and Communication Application Conference (SIU), Zonguldak, Turkey, Türkiye, 16 - 19 Mayıs 2016 identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/siu.2016.7495913
  • Basıldığı Şehir: Zonguldak, Turkey
  • Basıldığı Ülke: Türkiye
  • Anahtar Kelimeler: image search system, relevance feedback, color descriptors, edge descriptors
  • Kocaeli Üniversitesi Adresli: Hayır

Özet

In this study, an application that can retrieve images similar to a given query image is developed. The application extracts multiple image descriptors related to color and texture from the images and use them to rank the images in its database according to their similarity to the query image. It displays the most similar images on screen, after which the user give feedback about the relevancy of the returned images to the application. In the latter case, the application returns a set of images that are more relevant to the users' query. The performance of the application is evaluated on the Simplicity dataset.