A novel sparse representation model for pedestrian abnormal trajectory understanding

作者：

Highlights：

• A novel sparse representation method for pedestrian trajectory abnormal analysis.

• Utilizing Lp-regularization (0 < p < 1) to get sparser solutions.

• An effective solver for the proposed method with EM algorithm and entropy.

摘要

•A novel sparse representation method for pedestrian trajectory abnormal analysis.•Utilizing Lp-regularization (0 < p < 1) to get sparser solutions.•An effective solver for the proposed method with EM algorithm and entropy.

论文关键词：Pedestrian behavior,Sparse representation,Lp-norm,Information entropy,Trajectory understanding

论文评审过程：Received 17 September 2018, Revised 20 June 2019, Accepted 20 June 2019, Available online 21 June 2019, Version of Record 20 July 2019.

论文官网地址：https://doi.org/10.1016/j.eswa.2019.06.041