### Abstract

A diversity of tracking problems exists in which cohorts of densely packed particles move in an organized fashion, however the stability of individual particles within the cohort is low. Moreover, the flows of cohorts can regionally overlap. Together, these conditions yield a complex tracking scenario that cannot be addressed by optical flow techniques that assume piecewise coherent flows, or by multi-particle tracking techniques that suffer from the local ambiguity in particle assignment. Here, we propose a graph-based assignment of particles in three consecutive frames to recover from image sequences the instantaneous organized motion of groups of particles, i.e. flows. The algorithm makes no a priori assumptions on the fraction of particles participating in organized movement, as this number continuously alters with the evolution of the flow fields in time. Graph-based assignment methods generally maximize the number of acceptable particles assignments between consecutive frames and only then minimize the association cost. In dense and unstable particle flow fields this approach produces many false positives. The here proposed approach avoids this via solution of a multi-objective optimization problem in which the number of assignments is maximized while their total association cost is minimized. The method is validated on standard benchmark data for particle tracking. In addition, we demonstrate its application to live cell microscopy where several large molecular populations with different behaviors are tracked.

Original language | English (US) |
---|---|

Pages (from-to) | 531-540 |

Number of pages | 10 |

Journal | Computer Vision and Image Understanding |

Volume | 115 |

Issue number | 4 |

DOIs | |

State | Published - Apr 2011 |

### Fingerprint

### Keywords

- Graph algorithms
- Multi-directional flows
- Multi-objective optimization
- Particle tracking
- Vector field

### ASJC Scopus subject areas

- Software
- Computer Vision and Pattern Recognition
- Signal Processing

### Cite this

*Computer Vision and Image Understanding*,

*115*(4), 531-540. https://doi.org/10.1016/j.cviu.2011.01.001

**Optimal-flow minimum-cost correspondence assignment in particle flow tracking.** / Matov, Alexandre; Edvall, Marcus M.; Yang, Ge; Danuser, Gaudenz.

Research output: Contribution to journal › Article

*Computer Vision and Image Understanding*, vol. 115, no. 4, pp. 531-540. https://doi.org/10.1016/j.cviu.2011.01.001

}

TY - JOUR

T1 - Optimal-flow minimum-cost correspondence assignment in particle flow tracking

AU - Matov, Alexandre

AU - Edvall, Marcus M.

AU - Yang, Ge

AU - Danuser, Gaudenz

PY - 2011/4

Y1 - 2011/4

N2 - A diversity of tracking problems exists in which cohorts of densely packed particles move in an organized fashion, however the stability of individual particles within the cohort is low. Moreover, the flows of cohorts can regionally overlap. Together, these conditions yield a complex tracking scenario that cannot be addressed by optical flow techniques that assume piecewise coherent flows, or by multi-particle tracking techniques that suffer from the local ambiguity in particle assignment. Here, we propose a graph-based assignment of particles in three consecutive frames to recover from image sequences the instantaneous organized motion of groups of particles, i.e. flows. The algorithm makes no a priori assumptions on the fraction of particles participating in organized movement, as this number continuously alters with the evolution of the flow fields in time. Graph-based assignment methods generally maximize the number of acceptable particles assignments between consecutive frames and only then minimize the association cost. In dense and unstable particle flow fields this approach produces many false positives. The here proposed approach avoids this via solution of a multi-objective optimization problem in which the number of assignments is maximized while their total association cost is minimized. The method is validated on standard benchmark data for particle tracking. In addition, we demonstrate its application to live cell microscopy where several large molecular populations with different behaviors are tracked.

AB - A diversity of tracking problems exists in which cohorts of densely packed particles move in an organized fashion, however the stability of individual particles within the cohort is low. Moreover, the flows of cohorts can regionally overlap. Together, these conditions yield a complex tracking scenario that cannot be addressed by optical flow techniques that assume piecewise coherent flows, or by multi-particle tracking techniques that suffer from the local ambiguity in particle assignment. Here, we propose a graph-based assignment of particles in three consecutive frames to recover from image sequences the instantaneous organized motion of groups of particles, i.e. flows. The algorithm makes no a priori assumptions on the fraction of particles participating in organized movement, as this number continuously alters with the evolution of the flow fields in time. Graph-based assignment methods generally maximize the number of acceptable particles assignments between consecutive frames and only then minimize the association cost. In dense and unstable particle flow fields this approach produces many false positives. The here proposed approach avoids this via solution of a multi-objective optimization problem in which the number of assignments is maximized while their total association cost is minimized. The method is validated on standard benchmark data for particle tracking. In addition, we demonstrate its application to live cell microscopy where several large molecular populations with different behaviors are tracked.

KW - Graph algorithms

KW - Multi-directional flows

KW - Multi-objective optimization

KW - Particle tracking

KW - Vector field

UR - http://www.scopus.com/inward/record.url?scp=79851511153&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79851511153&partnerID=8YFLogxK

U2 - 10.1016/j.cviu.2011.01.001

DO - 10.1016/j.cviu.2011.01.001

M3 - Article

C2 - 21720496

AN - SCOPUS:79851511153

VL - 115

SP - 531

EP - 540

JO - Computer Vision and Image Understanding

JF - Computer Vision and Image Understanding

SN - 1077-3142

IS - 4

ER -