Performance Analysis Of A Spatial Domain Based Model For Blind Image Quality Assessment

Mohamad Nor Azam, Nurin Insyirah Amanina (2018) Performance Analysis Of A Spatial Domain Based Model For Blind Image Quality Assessment. Project Report. Universiti Teknikal Malaysia Melaka, Melaka,Malaysia. (Submitted)

Text (24 Pages) Performance Analysis Of A Spatial Domain Based Model For Blind Image Quality Assessment.pdf - Submitted Version Download (2MB)

Abstract

This project focuses on image quality assessment (IQA), especially in the problems of estimating the quality of an image blindly or without the presence of reference information.Previous blind IQA (BIQA) models often utilize transform-based quality predictive features to perform their quality prediction.This approach,however, can be computationaly expensive due to the need of image transformation process.This project attempts to alleviate this by developing a transform free BIQA model char operates on the image spatial domain.The model use generalized Gaussian distribution (GGD) and asymmetric generalized Gaussian distribution (AGGD)statistical parameters as quality predictive features.The project involves extracting relevant quality predictive features from the image's spatial domain,developing a quality prediction model through support vector regression (SVR) utilizing the extracted features and analyse the model 's performance through comparison with several available BIQA models in terms of prediction accuracy, generalization capability as well as computational requirement.

Actions (login required)

View Item