Unsupervised Image Region of Interest Detection using Discrete Cosine Transform

Publication Name

2021 4th International Conference on Information Communication and Signal Processing, ICICSP 2021

Abstract

Spatial domain analysis of images has been recognised as an important task in image retrieval and computer vision, however, its application in the determination of image Region of Interests often require the inclusion of supervised training and similarity computations, both of which can be difficult to implement during the modelling of the images in a large number of images. however, unsupervised image Region of Interest determination via Frequency domain analysis can be implemented without training or similarity computation therefore it is more suitable for the modelling of images in a large set. This paper demonstrates image frequency domain analysis via Discrete Cosine Transform and presents its application in the unsupervised determination of Region of Interests for the elimination of the spatial incoherency often associated with the Bag-of-Visual Word Modelling.

Open Access Status

This publication is not available as open access

First Page

348

Last Page

351

Share

COinS
 

Link to publisher version (DOI)

http://dx.doi.org/10.1109/ICICSP54369.2021.9611849