A new cast shadow detection method for traffic surveillance video analysis using color and statistical modeling

作者：

Highlights：

• A new foreground detection method is introduced.

• A set of new chromatic criteria is proposed for shadow pixels differentiation.

• A shadow region detection method is used to detect the continuous shadow regions.

• A statistical shadow model is built to model and classify the shadow pixels.

• Real world traffic videos are used for experiments.

摘要

•A new foreground detection method is introduced.•A set of new chromatic criteria is proposed for shadow pixels differentiation.•A shadow region detection method is used to detect the continuous shadow regions.•A statistical shadow model is built to model and classify the shadow pixels.•Real world traffic videos are used for experiments.

论文关键词：Traffic video analysis,Shadow detection,Global Foreground Modeling (GFM),New chromatic criteria,Shadow region detection method,Statistical shadow modeling

论文评审过程：Received 8 April 2019, Revised 26 September 2019, Accepted 5 December 2019, Available online 13 December 2019, Version of Record 26 December 2019.

论文官网地址：https://doi.org/10.1016/j.imavis.2019.103863