A Visual Residual Perception Optimized Network for Blind Image Quality Assessment

Lihuo He,Yanzhe Zhong,Wen Lu,Xinbo Gao

doi:10.1109/access.2019.2957292

Lihuo He, Yanzhe Zhong + Show 2 more

Open Access

https://doi.org/10.1109/access.2019.2957292

Copy DOI

Abstract

Blind image quality assessment (BIQA) is a fundamental yet challenging problem in the image processing system. Existing BIQA models have the following problems: 1) Due to the lack of available quality-label images, most of the methods have poor generalization ability in different distortion categories; 2) The impact of human visual characteristics on the content of images is not taken into account. In this paper, we proposed a visual residual perception optimized network (VRPON) that can effectively solve these problems. The proposed method separates the training of BIQA into two stages: 1) a distortion degree identification network and 2) an image quality prediction network. In the first stage, the spatial and temporal features of image sequences are extracted by CNN and LSTM respectively, which are used to evaluate the degree of image distortion. And then the proposed model is learned to predict image patches' scores in the second stage with the outputs of the first stage. Finally, a pooling strategy that follows the human visual saliency is designed to evaluate the quality score of the whole image. Experimental results show that the proposed VRPON not only has better performance than state-of-the-art methods on synthetic distorted images (LIVE, TID2013, CSIQ), but also has better robustness for different authentic distortions (LIVE challenge).

Highlights

Images are omnipresent in daily life and become a necessary carrier of information
The results demonstrate that the proposed method is superior to existing blind IQA (BIQA) techniques and has good robustness on different distortion types, both synthetic and authentic images
3) EVALUATION CRITERIA To evaluate the performances of the image quality assessment (IQA) algorithms, three standard measures is used in experiments, i.e., Spearman’s rank-order correlation coefficient (SROCC), Pearson linear correlation coefficient (PLCC) and Perceptually Weighted Rank Correlation (PWRC) [37]

Summary

Introduction

Images are likely to be inevitably degraded and introduce a variety of distortions in the process of acquisition, compression, transmission, and storage. These distorted images can affect people’s visual experience and cause unnecessary trouble. Designing assessment method to automatically evaluate the image quality has become a significant task in the field of image processing, computer vision, and pattern recognition. The subjective assessment method is designed to evaluate the image quality, which is the most reliable method. It is very expensive and infeasible for big image data.

Methods

Results

Conclusion