| 无参考图像质量评价方法的设计原则 |
| |
| 引用本文： | 安雪晶,田媛.无参考图像质量评价方法的设计原则[J].中国光学与应用光学,2009,2(2):140-144. |
| |
| 作者姓名： | 安雪晶 田媛 |
| |
| 作者单位： | 1. 辽宁机电职业技术学院,辽宁,丹东,118002
2. 中国科学院长春光学精密机械与物理研究所,吉林,长春,130033;中国科学院,研究生院,北京,100039 |
| |
| 摘 要： | 对于无法获得参考图像的图像处理系统，无参考（NR）图像质量评价具有重要意义。讨论了NR图像质量评价的基本问题，介绍了NR图像质量评价方法的基本原理，重点描述了基于块效应、图像模糊和交互神经进化的NR图像质量评价方法的内容。根据现有的NR图像质量方法都存在针对特定的类型而设计的缺陷，应用模糊测度和模糊积分，结合人类视觉系统（HVS）特性，并考虑盲失真度量遵循量化质量定律，提出了一个全面的NR图像质量评价方法的设计原则。
|
| 关 键 词： | 图像质量 无参考(NR)像质评价 人类视觉系统(HVS) |
| 收稿时间： | 2009-01-11 |
| 修稿时间： | 2009-03-13 |
Design principle of no reference image quality evaluation |
| |
| Authors: | AN Xue-jing TIAN Yuan |
| |
| Institution: | 1.Liaoning Technological College of Machinery and Electricity,Dandong 118002,China; 2.Changchun Institute of Optics,Fine Mechanics and Physics, Chinese Academy of Sciences,Changchun 130033,China); 3.Graduate University of Chinese Academy of Sciences,Beijing 100039,China |
| |
| Abstract: | The no reference image quality evaluation is an important method in image processing technologies when the original image can not be obtained. In this paper, the basic problems in no reference image quality evaluation methods are discussed and several methods based on block effects, image blurs and interactive neural evolution used in no reference image quality evaluation are explained and analyzed. An overall no reference image quality evaluation principle is presented in accordance with the shortnesses above mentioned, and it suggests that the evaluation principle should be a synthetical index applying fuzzy measures and fuzzy integrals, incorporating Human Vision System(HVS) characteristics and adhering to the philosophy of quantifying quality by blind distortion measurements. |
| |
| Keywords: | image quality No reference(NR) image quality evaluation Human Vision System |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《中国光学与应用光学》浏览原始摘要信息 |
| 点击此处可从《中国光学与应用光学》下载免费的PDF全文 |
|
