自动化学报2017，Vol.43Issue(3)：376-389,14.DOI:10.16383/j.aas.2017.c160039
基于结构化预测的细胞跟踪方法
Cell Tracking Using Structured Prediction
摘要
Abstract
In this work we propose a new joint detection and tracking method for cell tracking. First we develop a new procedure for generating an over complete set of detection hypothesis via ellipse fitting methods. Then we define several local events and corresponding labeling variables to account for the biological behavior of cells and the imperfection in segmentation, and formulate the task of cell tracing as an integer programming problem with constraints. In addition, instead of learning local classifiers, we exploit a recently proposed block-coordinate Frank-Wolfe algorithm to automatically learn optimal parameters of our model. We also present the kernelized version of the learning algorithm which can boost the tracking performance even further. We conduct extensive experiments on public datasets, showing that our method consistently outperforms traditional countetparts.关键词
细胞跟踪/结构化预测/结构化学习/Block-coordinate/Frank-Wolfe算法Key words
Cell tracking/structured prediction/structured learning/Block-coordinate Frank-Wolfe algorithm引用本文复制引用
陈旭,万九卿..基于结构化预测的细胞跟踪方法[J].自动化学报,2017,43(3):376-389,14.基金项目
国家自然科学基金(61174020)资助Supported by National Natural Science Foundation of China(61174020) （61174020）
