基于高阶相位一致性的混合失真图像质量评价  

Multiply-Distorted Image Quality Assessment Based on High-Order Phase Congruency

作　　者：侯春萍[1] 马彤彤[1] 岳广辉[1] 冯丹丹[1] 刘月[1] 

机构地区：[1]天津大学电气自动化与信息工程学院,天津300072

出　　处：《激光与光电子学进展》2017年第7期122-130,共9页Laser & Optoelectronics Progress

基　　金：国家自然科学基金面上项目(61471262);重点国际(地区)合作研究项目(61520106002);教育部博士点基金(20130032110010)

摘　　要：混合失真图像质量评价是图像质量评价(IQA)领域的重点和难点,基于高阶相位一致性,提出了一种混合失真无参考IQA算法。计算了高阶相位一致性用于捕捉图像结构信息,应用灰度共生矩阵分别提取了4阶相位一致性图像的统计特征;在分析相邻阶相位一致性的相关性及相邻阶相位一致性局部熵的相关性的基础上,分别计算了相邻阶相位一致性及其局部熵的互信息和交叉熵;利用支持向量回归机制建立回归模型并进行质量预测。在MLIVE和MDID2013数据库上的实验结果表明,该算法的评价结果与主观评价分数具有很高的一致性,其性能优于当今主流的全参考和无参考IQA算法。Image quality assessment for multiply-distorted images is the emphasis and difficulty in image quality assessment （IQA） filed. Based on the high-order phase congruency, a no-reference IQA method for multiply- distorted images is proposed. The high-order phase congruency is computed to capture the structural information of the image. The statistical features of four orders phase congruency are extracted by gray level co-occurrence matrix, respectively. And, based on the analysis of the correlation between adjacent orders of phase congruency and the correlation between adjacent orders of local entropy of phase congruency, the mutual information and cross entropy o{ that are calculated. The support vector regression is utilized to build a regression model and then it is used for quality predicting. The experimental results on MLIVE and MDID2013 databases show that the proposed method has high consistency with the subjective evaluation scores and outperforms the state-of-the-art full-reference and noreference IQA metrics.

关 键 词：图像处理 混合失真图像质量评价 高阶相位一致性 灰度共生矩阵 相关性 

分 类 号：TN919.8[电子电信—通信与信息系统]