Informal Complexity Analysis of Image Quality Based on Human Visual System

摘要

In this study, we propose NRDPF-IQA (No Reference Distortion Patch Features Image Quality Assessment) Model, aims to use to measure the image quality assessment for JPEG2000. The proposed method takes advantage over the other existing image quality assessment metrics of the contrast changes in the image quality. The proposed quality metric was tested by using LIVE image database and CSIQ image database. The experimental results show that the new index performance compared with the other NR-IQA Models that require training on LIVE databases, CSIQ database and TID database.