Robust multi-scale ship tracking via multiple compressed features fusion

作者：

Highlights：

• Extract two complementary good features to track the target ship.

• Construct random measurement matrices according to spatio-temporal structure constraints.

• Perform efficient feature evaluation through cumulative classification performances.

摘要

•Extract two complementary good features to track the target ship.•Construct random measurement matrices according to spatio-temporal structure constraints.•Perform efficient feature evaluation through cumulative classification performances.

论文关键词：Compressive sensing theory,Random measurement matrices,Naive Bayes classifier,Coarse-to-fine strategy

论文评审过程：Received 15 June 2014, Revised 12 December 2014, Accepted 15 December 2014, Available online 24 December 2014.

论文官网地址：https://doi.org/10.1016/j.image.2014.12.006