Region-of-Interest (ROI) Database

Lena mean ROI

Introduction

Within images there are usually regions which are of more interest to a human observer as compared to others. These regions are widely known as regions-of-interest (ROI) and the remainder of the image is referred to as background. Many image processing and image coding applications are based on ROI schemes in order to improve their performance, such as image segmentation, unequal error protection, and saliency based image coding.

In order to obtain ROI for a particular set of images one can choose either an objective method (automatic ROI detection) or subjective method of identification. In this case, the subjective method may serve as a ground truth for the objective methods to be developed on, similar to MOS and objective metrics in quality assessment. We conducted a subjective test in which we presented the seven reference images of the Wireless Imaging Quality (WIQ) database (see http://www.bth.se/tek/rcg.nsf/pages/wiq-db) to a number of human observers and asked them to select a region which is of particular interest to them. The outcomes we have used to improve the quality prediction performance of existing image quality metrics by incorporating information about the relative saliency of regions in the images [1]. We make the ROI selections freely available in the ROI database.

Chief investigators

  • Ulrich Engelke, Blekinge Institute of Technology, Sweden (uen@bth.se)
  • Hans-Jürgen Zepernick, Blekinge Institute of Technology, Sweden (hjz@bth.se)

Database description

The ROI database is described in detail in [1] (a copy of the paper can be obtained on request). The files that are included in the ROI database are the following:

  • The spatial coordinates of the ROI selections of all 30 participants
  • A 'readme' file containing all information necessary to use the ROI database

Download information

The ROI database can be downloaded here:

The ZIP files are password protected, please send an email to Ulrich Engelke  (ulrichengelke@gmail.com) to obtain the password. A few lines about your background and the purpose of use of the ROI database would be highly appreciated. If you use the ROI database for your research, we kindly ask you to refer to our article [1] and to this website [2] (see ‘Reference Information’ below).

Reference information

If you use the ROI database for your research, please refer to our article [1] and to this website [2] as follows:

[1] U. Engelke,  H.-J. Zepernick, “A Framework for Optimal Region-of-Interest Based Quality Assessment in Wireless Imaging,” Journal of Electronic Imaging, Special Section on Image Quality, vol. 19, no. 1, ID 011005, 2010.

[2] U. Engelke and H.-J. Zepernick, "Region of Interest Database," http://www.bth.se/tek/rcg.nsf/pages/roi-db, 2010.

Other related publications:

[3] U. Engelke, H.-J. Zepernick, A. Maeder, “Visual Attention Modeling: Regions-of-Interest Versus Fixation Patterns,” Picture Coding Symposium (PCS), 2009.

[4] U. Engelke, H.-J. Zepernick, “Optimal Region-Of-Interest Based Visual Quality Assessment,” IS&T/SPIE Human Vision and Electronic Imaging XIV, 2009.

[5] U. Engelke, X. N. Vuong, H.-J. Zepernick, “Regional Attention to Structural Degradations for Perceptual Image Quality Metric Design,” IEEE Int. Conf. on Acoustics, Speech, and Signal Processing (ICASSP), 2008.

Copyright notice

Permission is hereby granted, without written agreement and without license or royalty fees, to use, copy, modify, and distribute the data provided and its documentation for research purpose only. The data provided may not be commercially distributed. In no event shall the Blekinge Institute of Technology be liable to any party for direct, indirect, special, incidental, or consequential damages arising out of the use of the data and its documentation. The Blekinge Institute of Technology specifically disclaims any warranties. The data provided is on an "as is" basis and the Blekinge Institute of Technology has no obligation to provide maintenance, support, updates, enhancements, or modifications.

Edit
Share Share