No-reference image quality assessment based on local and global using attention features

Ran Chang

doi:10.1117/12.2673113

14 April 2023 No-reference image quality assessment based on local and global using attention features

Ran Chang

Proceedings Volume 12612, International Conference on Artificial Intelligence and Industrial Design (AIID 2022); 1261211 (2023) https://doi.org/10.1117/12.2673113
Event: International Conference on Artificial Intelligence and Industrial Design (AIID 2022), 2022, Zhuhai, China

Abstract

No-reference image quality assessment is the prediction of image quality without the original image and has a wide range of applications in fields such as image recognition. The wide content of images and the diversity of distortions acquired by mobile devices, make it a challenge to evaluate the quality of such truly distorted images captured by mobile devices. Therefore, in this paper, we propose to use global and local features of images to deal with the diversity of image distortions, using an attention module to extract global features to improve the characterization ability of global features. The global semantic features are used as the input of an adaptive weight and bias generation network, which can generate the weights and biases of the final quality prediction network based on the image content. In this paper, experiments are conducted on three real distortion databases, KonIQ-10k, BID, and LIVEC, and the results show that the proposed method outperforms existing no-reference image quality assessment methods. Also, experiments are conducted on two synthetic distortion databases LIVE and TID2013, and the results show that the method also achieves competitive results in evaluating synthetic distorted images.

Citation Download Citation

Ran Chang "No-reference image quality assessment based on local and global using attention features", Proc. SPIE 12612, International Conference on Artificial Intelligence and Industrial Design (AIID 2022), 1261211 (14 April 2023); https://doi.org/10.1117/12.2673113

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
9 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Image quality

Databases

Feature extraction

Semantics

Content addressable memory

Image processing

Image enhancement

Show All Keywords

Keywords/Phrases

Search In:

Publication Years