A New Approach of Image Quality Index

Sheikh Md. Rabiul Islam, Md. Tariqul Islam, Xu Huang

Research output: A Conference proceeding or a Chapter in BookConference contributionpeer-review

4 Citations (Scopus)

Abstract

Image quality measurement procedures the quality of images automatically constructed on subjective evaluation. The smallest element of an image is pixel, can be converted to structure based evaluation using structure similarity index. We are proposed a novel modified feature similarity index (MFSIM) for image quality assessment (IQA) in this paper. This index used three features of an image namely Phase Congruency (PC), Gradient Magnitude (GM) and Angle (theta). The primary feature of MFSIM is PC which is contrast in variant system and measure the local structure of images. GM is the secondary feature which plays a complementary role. The angle of an image uses as tertiary feature of MFSIM. For this index need weighting function for evaluating of features and PC act as weighting function. All the task performs using five databases. The result of this proposed method shows higher rank-order correlation, low MSE than any other method like human visual system (HVS).

Original languageEnglish
Title of host publicationProceedings of the 2017 4th International Conference on Advances in Electrical Engineering (ICAEE)
EditorsMustafa Habib Chowdhury
Place of PublicationUnited States
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages223-228
Number of pages6
ISBN (Electronic)9781538608692
ISBN (Print)9781538608708
DOIs
Publication statusPublished - 1 Jul 2017
Event4th International Conference on Advances in Electrical Engineering (ICAEE 2017) - Dhaka, Dhaka, Bangladesh
Duration: 28 Sept 201730 Sept 2017

Publication series

Name4th International Conference on Advances in Electrical Engineering, ICAEE 2017
Volume2018-January

Conference

Conference4th International Conference on Advances in Electrical Engineering (ICAEE 2017)
Abbreviated titleICAEE 2017
Country/TerritoryBangladesh
CityDhaka
Period28/09/1730/09/17

Cite this