Active colloids segmentation and tracking

作者：

Highlights：

• A combination of the level set method and circular hough transform to segment highly cluttered colloids.

• All the colloids' trajectories are recovered simultaneously within a single graph.

• We annotate in total nine video sequences by human observers.

• The proposed method reaches average accuracy of 93.19

摘要

Highlights•A combination of the level set method and circular hough transform to segment highly cluttered colloids.•All the colloids' trajectories are recovered simultaneously within a single graph.•We annotate in total nine video sequences by human observers.•The proposed method reaches average accuracy of 93.19

论文关键词：Multiple objects tracking,Colloidal suspension,Min-cost/max flow,Data association,Image segmentation,Level set method

论文评审过程：Received 11 March 2015, Revised 11 March 2016, Accepted 25 April 2016, Available online 26 May 2016, Version of Record 4 June 2016.

论文官网地址：https://doi.org/10.1016/j.patcog.2016.04.022