基于残差细化和边界感知的显著性检测

Saliency Detection Based on Residuals Refinement and Edge Perception

针对目前显著性检测算法精度不高、目标边缘模糊等问题,本文提出一种基于残差细化和边界感知的显著性检测算法。其中残差细化充分利用高层语义特征和低层细节特征得到信息较为完整的显著图,而边界感知能关注到更多的边缘细节信息,解决目标边缘模糊的问题。实验表明,通过定性和定量比较,本文算法的精确度和准确度都得到了提高,且有效解决了目标边界模糊的问题。

The saliency detection based on residuals refinement and edge perception solves the problems of low precision and target edge.The residual refinement makes use of the high-level semantic features and low-level detail features to get a more complete picture,while the edge sensing focus on edge details to solve the problem of the target edge.The experimental results show that the accuracy of the algorithm is improved by qualitative and quantitative comparison,and the problem of the target boundary is solved.