Contrast Image Quality Assessment Algorithm Based on Probability Density Functions Features

Abstract

Recently, the existing image quality Assessment algorithms (IQAs) works focusing on distorted images by compression, noise and blurring. Reduced-reference Image Quality Metric for Contrast-changed images (RIQMC) and No Reference-Image Quality Assessment (NR-IQA) for Contrast-Distorted Images (NR-IQA-CDI) have been created for CDI. For each of the five global feature that used in NR-IQA-CDI, The statistical model or the Probability Density Function (PDF) was determined using a Sun2012 database which containing a wide variety of natural scene images. NR-IQA-CDI showed poor performance in two out of three image databases, where the Pearson Correlation Coefficient (PLCC) were only 0.5739 and 0.7623 in TID2013 and CSIQ database, respectively. For this reason, we present the NR-IQA-CDI based on Monotonic Probability Density Functions (PDFs) (NR-IQA-CDI-MPCF) to address the problem of the existing bell-curve-like PDF of contrast features that cannot reflect the monotonic relation between contrast feature values and perceptual image quality. The findings indicate that the NR-IQA-CDI-MPCF outperforms the current NR-IQA-CDI, especially in the TID2013 database.