Stochastic shadow detection using a hypergraph partitioning approach

作者：

Highlights：

• A novel shadow detection method is proposed for video sequences or still images.

• Color, texture and hypergraph methods are used to classify shadow/non-shadow segments.

• A voting scheme finds consensus among the classifiers to detect the shadow segments.

• The proposed method tends to perform better than comparable state-of-the-art methods.

摘要

Highlights•A novel shadow detection method is proposed for video sequences or still images.•Color, texture and hypergraph methods are used to classify shadow/non-shadow segments.•A voting scheme finds consensus among the classifiers to detect the shadow segments.•The proposed method tends to perform better than comparable state-of-the-art methods.

论文关键词：Shadow detection,Chromaticity,Gradient correlation,Hypergraph segmentation

论文评审过程：Received 9 July 2015, Revised 14 September 2016, Accepted 15 September 2016, Available online 20 September 2016, Version of Record 28 September 2016.

论文官网地址：https://doi.org/10.1016/j.patcog.2016.09.008