Multi-attribute and semantic-integrated modeling for interpretable sonar image quality assessment

2026-08-10

PRE

Qiang Tu

Boqin Cai

Weifan Lu

Shuwu Chen

Weiling Chen *

DOI：10.1016/j.patcog.2026.114611

Original

Original request for help

Save

Abstract

En 中文

• Interpretable sonar image quality assessment. • Multi-branch learning for disentangled quality attribute representation. • Graph-based modeling for scene-aware quality prediction. • Transfer learning for attribute-aligned representation under limited data. • Superior correlation with human perception on a public sonar IQA dataset.

Keywords:

Sonar image quality assessment

Perceptual quality attributes

Scene semantics

Graph-based modeling

Transfer learning

Multi-attribute and semantic-integrated modeling for interpretable sonar image quality assessment

Abstract

Journal

Organization

Cited Papers

Citing Papers