Learning Crisp Edge Detector Using Logical Refinement Network

TL;DR

This paper introduces a logical refinement network that improves edge detection accuracy and boundary clarity in both 2D and 3D images, outperforming existing methods without post-processing.

Contribution

The paper presents a novel network architecture that leverages logical relationships between segmentation and edges for precise, crisp edge detection in 2D and 3D images.

Findings

Outperforms state-of-the-art methods on 2D nuclei images.

Achieves superior boundary clarity in 3D microscopy images.

Produces thinner, more accurate edge maps without post-processing.

Abstract

Edge detection is a fundamental problem in different computer vision tasks. Recently, edge detection algorithms achieve satisfying improvement built upon deep learning. Although most of them report favorable evaluation scores, they often fail to accurately localize edges and give thick and blurry boundaries. In addition, most of them focus on 2D images and the challenging 3D edge detection is still under-explored. In this work, we propose a novel logical refinement network for crisp edge detection, which is motivated by the logical relationship between segmentation and edge maps and can be applied to both 2D and 3D images. The network consists of a joint object and edge detection network and a crisp edge refinement network, which predicts more accurate, clearer and thinner high quality binary edge maps without any post-processing. Extensive experiments are conducted on the 2D nuclei images from Kaggle 2018 Data Science Bowl and a private 3D microscopy images of a monkey brain, which show outstanding performance compared with state-of-the-art methods.