基于边缘梯度信息的图像质量评价方法  

Image Quality Assessment Method Based on Edge and Gradient Information

作　　者：梁敏瑜[1] 孙权森[1] 

机构地区：[1]南京理工大学计算机科学与技术学院,南京210094

出　　处：《计算机科学与探索》2012年第11期1019-1025,共7页Journal of Frontiers of Computer Science and Technology

基　　金：国家自然科学基金(60773172);高等学校博士学科点专项科研基金(200802880017);南京理工大学自主科研专项计划资助项目(2011ZDJH26)~~

摘　　要：利用人眼视觉感知特性评价图像的质量一直是图像处理领域的研究热点,但是目前很多客观评价方法未能充分考虑视觉感知特性。针对当前工作的不足,提出了基于边缘梯度信息的图像质量评价方法。采用基于小波变换模极大值的方法提取图像的边缘梯度信息,并利用高斯函数对图像区域进行区域加权,提出基于边缘梯度信息的评价方法。实验结果表明,该方法能够准确和有效地度量不同失真类型图像的质量,与主观评价值的一致性较好,并且该算法的性能也有较大的提高。The human visual characteristics to evaluate image quality has been a hot research field of image processing. However lots of metrics do not fully consider the characteristics of human visual system （HVS）. For the deficiencies of current work, this paper proposes a metric based on edge and gradient information. Firstly, wavelet transform modulus maxima is used to detect edge and gradient. Then, different image regions are weighted with Gaussian function. Finally, the assessment method is proposed. The experimental results show that this method can measure the quality of different distorted images accurately and effectively, and its consistency with subjective evaluation value is better than other metrics. The experimental results also show that the performance of this method has been improved greatly.

关 键 词：图像质量评价 人眼视觉系统(HVS) 结构相似度 边缘信息 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]