单幅自然场景深度恢复

离焦测距算法是一种用于恢复场景深度信息的常用算法。传统的离焦测距算法通常需要采集多幅离焦图像，实际应用中具有很大的制约性。文中基于局部模糊估计提出单幅离焦图像深度恢复算法。基于局部模糊一致性的假设，本文采用简单而有效的两步法恢复输入图像的深度信息：1)通过求取输入离焦图和利用已知高斯核再次模糊图之间的梯度比得到边缘处稀疏模糊图 2)将边缘位置模糊值扩离至全部图像，完整的相对深度信息即可恢复。为了获得准确的场景深度信息，本文加入几何条件约束、天空区域提取策略来消除颜色、纹理以及焦点平面歧义性带来的影响，文中对各种类型的图片进行对比实验，结果表明该算法能在恢复深度信息的同时有效抑制图像中的歧义性。

Recovering depth from a single natural defocused image

Depth from defocus is a common algorithms for recovery scene depth information.In traditional,its need to collect multiple defocus images. However, it is difficult in practical application.In this paper, we address a challenging problem of recovering the depth from a single photograph taken with an uncalibrated conventional camera. In order to achieve this, we present a simple yet effective two-step approach to estimate the blur map of input image . i). A sparse blur map is obtained by estimating the amount of defocus blur at edge locations. ii). Based on image segmentation and sparse blur map to get global blur map,complete depth information can be restoration.In order to obtain accurate scene depth information, this paper join geometry constraints, the sky area extraction and segmentation strategy to eliminate the impact of color, texture and focal plane ambiguity.Experimental results on a variety of images show that our method can acquire a reliable estimation of the depth of a scene.