基于Gabor滤波器的全参考图像质量评价方法  

Full reference image quality assessment based on Gabor filter

作　　者：王翔[1,2] 丁勇[1] 

机构地区：[1]浙江大学超大规模集成电路设计研究所,浙江杭州310027 [2]杭州电子科技大学射频电路与系统教育部重点实验室,浙江杭州310018

出　　处：《浙江大学学报（工学版）》2013年第3期422-430,共9页Journal of Zhejiang University：Engineering Science

基　　金：国家"863"高技术研究发展计划资助项目(2009AA011706);浙江大学基本科研业务费专项资金资助项目(KYJD09012)

摘　　要：针对现有图像质量评价方法无法像人眼一样准确地评价各种图像失真,提出一种基于Gabor滤波器的全参考型图像质量评价方法.Gabor滤波器采用复数波形去调制高斯函数,得到一个在时间和频域上联合分辨率达到理论极限.通过优化Gabor滤波器的参数,可以有效地模拟人眼视觉特性.使用Gabor滤波器对图像的特征信息进行提取和对比之后,可以得到较好的图像质量评价效果.实验结果表明:该方法在对各种典型的失真图像的质量评价中,客观评分的各项指标均与人的主观感受较为一致.To resolve the problem that existing image quality assessment methods cannot assess the images as accurate as human visual systems,a full reference image quality assessment method based on Gabor filter was proposed.Gabor filter uses complex waveforms to modulate Gaussian function.It can simultaneously reach the theoretically limited lower bound in time and frequency on the product of its bandwidth and duration.The parameters of Gabor filter are optimized to simulate the human visual system effectively.After extracting the features from the Gabor filter and comparing them,better effects of image quality assessment are achieved.Experimental results show that the proposed method can correlate well with the human subjectivity in all kinds of distortions listed in the reference database.

关 键 词：GABOR滤波器 全参考 图像质量评价 人眼视觉系统 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]