A No-Reference Image Quality Assessment Metric by Multiple Characteristics of Light Field Images

Liang Shan,Chunli Meng,Chao Yang,Ping An,Xinpeng Huang,Liquan Shen

doi:10.1109/access.2019.2940093

Liang Shan, Chunli Meng + Show 4 more

Open Access

https://doi.org/10.1109/access.2019.2940093

Copy DOI

Abstract

Evaluation of light field image (LFI), especially micro-lens camera light field (LF), is a new and challenging work. The development of image quality assessment (IQA) metric of LFIs relies on the subjective quality assessment database. In this paper, we establish a perceptual quality assessment dataset consisting of 240 distorted images from 8 source images with five distortion types. Furthermore, a no-reference IQA metric is proposed by combining 2D and 3D characteristics of LFI with the Support Vector Regression (SVR) model. The performance of the proposed metric is demonstrated by comparing with some classical full reference IQA metrics both on the presented dataset and a third-party dataset. The experiment results show that our method has a better performance than others.

Highlights

With the development of Computational Photography [1], the capability of a camera can be greatly improved
Various types of post-processing of light field image (LFI) based on microlens array cameras have developed rapidly in recent years, but the lack of Light Field (LF) image quality assessment (IQA) standards to evaluate the quality of LFIs has gradually become a vital issue limiting the development of LF
This paper deeply studies the work of predecessors and gives a complete research plan for IQA of LFIs

Summary

INTRODUCTION

With the development of Computational Photography [1], the capability of a camera can be greatly improved. In the field of LFI, whether it is the verification of post-processing algorithms such as compression coding and super-resolution, or the optimization of LF camera imaging, IQA has an important guiding role as a standard. It is the main work of this paper to reestablish an LF IQA dataset containing more abundant distortion factors and use this as the basis for proposing an objective IQA method.

RELATED WORK

EXPERIMENTS AND RESULTS

Objective

Findings

CONCLUSION