Color edge detection by learning classification network with anisotropic directional derivative matrices

作者：

Highlights：

• A color edge detector combining anisotropic directional derivative matrix and neural network is proposed.

• The proposed detector is compared with several color edge detectors on three datasets.

• The proposed detector shows great performance on noiseless and Gaussian noise cases.

摘要

•A color edge detector combining anisotropic directional derivative matrix and neural network is proposed.•The proposed detector is compared with several color edge detectors on three datasets.•The proposed detector shows great performance on noiseless and Gaussian noise cases.

论文关键词：Anisotropic directional derivative matrices,Color edge detector,Learning-based detector,Ground truth edges,Edge probability map

论文评审过程：Received 4 August 2020, Revised 21 March 2021, Accepted 26 April 2021, Available online 8 May 2021, Version of Record 15 May 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.108004