Blurring-Effect-Free CNN Network of Structural Edge for Focus Stacking

Abstract

Focus stacking is a promising computational technique to extend depth of field by fusing images focused at different focal planes. However, existing focus stacking methods could not cope with the blurring-effect problem produced in structural edges, where depth values change abruptly. In this work, we firstly extract structural edges robustly by designing Des(depthmap-based extraction of structural edges)-ResNet. Then we propose a novel convolutional neural network (BEF-CNN) to restore blurring-effect-free image patches in order to enhance all-in-focus performance. To the best of our knowledge, it is the first work to utilize CNN to generate all-in-focus image directly instead of pixel-to-pixel correspondence with depthmap. Experimental results validate that our proposed algorithm has achieved best all-in-focus image while keeping the accuracy of depthmap.

Keywords

• All-in-focus
• focal stack
• Des-ResNet
• BEF-CNN
• structural edge