基于视觉感知特性的红外图像质量评价方法  

Infrared image quality evaluation based on human visual perception characteristics

作　　者：王明超[1] 李盈[1] 杨舒羽 

机构地区：[1]华北光电技术研究所,北京100015

出　　处：《激光与红外》2017年第7期921-924,共4页Laser & Infrared

摘　　要：根据红外成像特点,设计了一种基于视觉感知特性的红外图像质量评价算法。该算法结合人眼视觉和红外图像的结构信息对图像的失真程度进行描述,通过提取图像的边缘特征、对比度特征,然后利用视觉显著模型对特征进行差异融合,从而实现对红外失真图像的质量评测。实现结果表明,本文方法可对失真红外图像进行有效评价,与传统方法相较,此评价指标与人眼主观感知更一致。An infrared image quality evaluation method based on visual perception characteristics is proposed according to infrared imaging features. The visual perception characteristics and structural information of infrared image are utilized to describe the distortion degree of infrared images, and edge features and contrast features of infrared images are extracted. Then the features were fused differently by visual saliency to achieve quality evaluation of distorted in- frared images. Experimental results show this method can effectively evaluate distorted infrared images. Compared to traditional methods, the proposed method is in better consistency with human subjective perception.

关 键 词：人眼视觉 红外图像评价 视觉显著性 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]