快速注意力选择计算及其在图像质量评价中的应用

A Fast Computation of Attention Selection Map and Its Application in Image Quality Assessment

针对现有的模拟视觉注意力选择模型计算复杂度高、参数选取困难的问题,提出一种快速的频率域算法.将各频点上的幅度设为同一常数,保留其相位信息;对反变换到空间域的图像滤波后,得到视觉注意力显著图,其效果与现有的模型相似;并将显著图信息作为权重加入到几种主要的图像质量评价标准中.实验结果表明:带有注意力选择的图像质量评价标准更接近于人的主观评价,便于实时应用.

Existing computation methods of Human visual attention are rather impractical due to either the involved high computational complexity or the difficulties of selecting suitable parameters. In this paper, we propose a fast method to simulate attention selection in frequency domain. We set the amplitude spectrum to a constant and preserve phase spectrum, then transform them back into the spatial domain. After filtering, the resultant saliency map is quite similar to those of existing methods. In addition, we treat the salient information as weights to adjust several main existing criteria of image quality assessment. Experimental results show that our image quality assessment with attention selection is more consistent with human＇s subjective evaluation, besides its computational efficiency.