摘要
在交互式多媒体中,屏幕内容图像客观质量评价有广泛的应用价值。本文首先阐述了屏幕内容图像和自然图像的区别;然后分析了基于手工提取特征和基于深度学习自动提取特征两种屏幕内容图像评价算法;接着总结了常用的评价指标和屏幕内容图像数据集,并对屏幕内容图像质量评价算法的性能进行比较;最后对屏幕内容图像质量评价算法的研究进行展望。本文有助于研究人员深入理解该领域的研究现状,为推动其技术的发展和应用提供指导和支持。
In interactive multimedia,the objective quality assessment of screen content images has extensive applications.Firstly,the differences between screen content images and natural images were discussed in this paper.Then,from the perspective of feature extraction methods,the objective image quality assessment algorithms were divided into two categories:the methods based on manually extracting features and the methods based on deep learning for automatic feature extraction,besides comparing and analyzing their characteristics.Next,the standard technical metrics for measuring the performance of assessment algorithms and screen content image databases were summarized.The comparative analysis was conducted on the screen content image quality assessment methods.Finally,prospects for the research directions in screen content quality assessment algorithms were discussed.This paper helps the researchers gain a deeper understanding of the current research status in the field and provides the guidance and support for advancing the development and application of the technology.
作者
周子镱
董武
张艳
陆利坤
曾庆涛
马倩
ZHOU Zi-yi;DONG Wu;ZHANG Yan;LU Li-kun;ZENG Qing-tao;MA Qian(Beijing Key Laboratory of Signal and Information Processing for High-End Printing Equipment,Beijing Institute of Graphic Communication,Beijing 102600,China)
出处
《印刷与数字媒体技术研究》
北大核心
2025年第1期14-27,共14页
Printing and Digital Media Technology Study
基金
北京市数字教育研究重点课题(No.BDEC2022619027)
北京市高等教育学会2023年立项面上课题(No.MS2023168)
北京印刷学院校级科研项目(No.Ec202303、Ea202301、E6202405)
北京印刷学院学科建设和研究生教育专项(No.21090224002、21090323009、21090124013)
北京印刷学院出版学新兴交叉学科平台建设项目(No.04190123001/003)
北京邮电大学网络与交换技术全国重点实验室开放课题资助项目(No.SKLNST-2023-1-12)。
关键词
屏幕内容图像
客观质量评价
深度学习
人眼视觉特性
Screen content image
Objective quality assessment
Deep learning
Human visual system
作者简介
周子镱(2000年-),硕士研究生,主要研究方向为屏幕内容图像质量评价。;通讯作者:董武(1980年-),博士,副教授,主要研究方向为深度学习和图像质量评价。
