Segmentation for Image Indexing and Retrieval on Discrete Cosines Domain
Abstract: This paper used region
growing segmentation technique to segment the Discrete Cosines (DC) image. The
classic problem of content Based image retrieval (CBIR) is the lack of accuracy
in matching between image query and image in the database. By using region
growing technique on DC image,it reduced the number of image regions indexed.
The proposed of recursive region growing is not new technique but its
application on DC images to build
indexing keys is quite new and not yet presented by many authors. The experimental results show that
the proposed methods on segmented images present good precision which are
higher than 0.60 on all classes. So, it could be concluded that region growing
segmented based CBIR more efficient
compared to DC images in term of their precision 0.59 and 0.75,
respectively. Moreover, DC based CBIR can save time and simplify algorithm
compared to DCT images. The most significant finding from this work is instead
of using 64 DCT coefficients this research only used 1/64 coefficients which is
DC coefficient.
Author: Suhendro Y Irianto
Journal Code: jptkomputergg130030