局部和全局特征融合的红外图像显著性检测

为了更好地凸显复杂环境的红外目标特征,提出 一种融合局部和全局特征的红外图像 显著性检测方法。在获取图像超像素的基础上,提取每个区域空间距离加权的邻域对比度特 征,并考虑区域大小和位置的影响,构建局部显著图；然后提取每个区域空间距离加权的全 局灰度特征,构建全局显著图；最后融合局部和全局显著图,实现图像显著性检测。实验结 果 表明,本文方法的显著图结果目标区域一致高亮且边缘清晰,同时背景杂波抑制效果好。无 论 主观评价还是客观指标,本文方法都优于当前流行的图像显著性检测方法。

Saliency detection of infrared images based on local feature and global feature

In order to better represent infrared target feature under different environm ents,a novel saliency detection method based on local feature and global feature is proposed .Firstly,based on the image superpixels,the contrast feature of every region is extracted by weigh ting its spatial distance in the neighbor,and at the same time,considering the influence of size and position of every region,the local saliency map is constructed.Then the global feature of every region is extracted by weighting its spatial distance in the whole image and the global sa liency map is obtained.Finally,the saliency detection result of infrared image is obtained b y fusing the local and global saliency maps.The experimental results show that the saliency map of the proposed method has complete target feature,obvious edges and good background suppressio n,and the novel method is also better than state of the art methods both qualitatively and quantitatively.