视觉注意机制在图像增强中的应用研究

将视觉注意机制引入到直方图构造中,并在此基础上提出了一种新的基于灰度级信息量直方图的图像增强算法.该算法利用Itti视觉注意计算模型对图像的显著性进行分析,获得全局显著图；然后,将全局显著图划分为若干等大的子区域,求取各子区域的平均显著值,并做归一化处理,得到子区域的加权统计系数；再将各子区域的灰度级加权统计值相加,得到灰度级信息量直方图；最后,依据直方图均衡化的映射函数,调整灰度级的动态范围.实验结果表明,该算法明显优于经典的GHE算法和AHE算法,具有满意的视觉效果.

Image Enhancement Based on Visual Attention Mechanisms

A novel image enhancement algorithm based on gray-level information histogram was proposed.The image saliency analysis was performed using Itti′s visual attention model and a global salient graph was obtained.The global salient graph was divided into several sub-parts with the same size and then weighted statistical coefficients of sub-parts were achieved by normalizing the average salient values of those sub-parts.The information histogram was obtained by adding weighted statistical results of sub-parts.And the gray-level dynamic ranges were adjusted according to the mapping function of histogram equalization.The experimental results show that the proposed algorithm has better performance than the traditional GHE (Global Histogram Equalization) and AHE (Adaptive Histogram Equalization) methods.In addition,the visual effect of enhanced image was satisfied.