TY - JOUR
T1 - Image Quality Assessment Based on Contour and Region
AU - Huang , Chen
AU - Jiang , Ming
AU - Jiang , Tingting
JO - Journal of Computational Mathematics
VL - 6
SP - 705
EP - 722
PY - 2016
DA - 2016/12
SN - 34
DO - http://doi.org/10.4208/jcm.1611-m2016-0534
UR - https://global-sci.org/intro/article_detail/jcm/9821.html
KW - Image quality assessment, Contour detection, Image segmentation.
AB - <p style="text-align: justify;">Image Quality Assessment (IQA) is a fundamental problem in image processing. It is a common principle that human vision is hierarchical: we first perceive global structural information such as contours then focus on local regional details if necessary. Following this principle, we propose a novel framework for IQA by quantifying the degenerations of structural information and region content separately, and mapping both to obtain the objective score. The structural information can be obtained as contours by contour detection techniques. Experiments are conducted to demonstrate its performance in comparison with multiple state-of-the-art methods on two large scale datasets.</p>