A Dark and Bright Channel Prior Guided Deep Network for Retinal Image Quality Assessment

TL;DR

This paper introduces GuidedNet, a deep learning model that incorporates dark and bright channel priors to enhance retinal image quality assessment, validated on public and newly annotated datasets.

Contribution

It proposes a novel deep network architecture embedding image priors at the initial layer for improved quality classification of retinal images.

Findings

GuidedNet outperforms existing models on Eye-Quality dataset.

The re-annotated RIQA-RFMiD dataset improves validation robustness.

Embedding priors enhances discriminative ability of deep features.

Abstract

Retinal image quality assessment is an essential task in the diagnosis of retinal diseases. Recently, there are emerging deep models to grade quality of retinal images. Current state-of-the-arts either directly transfer classification networks originally designed for natural images to quality classification of retinal images or introduce extra image quality priors via multiple CNN branches or independent CNNs. This paper proposes a dark and bright channel prior guided deep network for retinal image quality assessment called GuidedNet. Specifically, the dark and bright channel priors are embedded into the start layer of network to improve the discriminate ability of deep features. In addition, we re-annotate a new retinal image quality dataset called RIQA-RFMiD for further validation. Experimental results on a public retinal image quality dataset Eye-Quality and our re-annotated dataset RIQA-RFMiD demonstrate the effectiveness of the proposed GuidedNet.