Modeling Herds and Their Evolvements from Trajectory Data

Huang, Yan; Chen, Cai; Dong, Pinliang

doi:10.1007/978-3-540-87473-7_6

Cited by 38 publications

(24 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Existing theoretical approaches for comparing and grouping trajectories are explored in [23,7,20,19,12,11]. Real-world applications of trajectory analysis, such as tracking animal herds and vehicle traffic, are discussed in [15,16,2,10,18,1]. Tractography, the science of using trajectories to model human brain white matter from diffusion magnetic resonance imaging (dMRI) data, is described in [4,3,21,8].…”

Section: Previous Workmentioning

confidence: 99%

A reeb graph approach to tractography

Sun

Cieslak

Grafton

et al. 2015

Proceedings of the 23rd SIGSPATIAL International Conference on Advances in Geographic Information Systems

View full text Add to dashboard Cite

We propose an efficient algorithm for discovering the highlevel topological structure of a collection of 3-dimensional trajectories. Our algorithm computes a sparse graph representing the latent "bundling" and "unbundling" structure of the trajectory data. This graph can serve both as a compact signature of the trajectory data set as well as a tool for efficient comparison among different data sets. Our problem formulation and the algorithms are broadly applicable and general-purpose but we focus on a particular neuroscience application to highlight the key features. In particular, our motivation stems from the emerging area of brain tractography, which aims to construct the connectome of human brain white matter fibers. These fibers can be inferred noninvasively using magnetic resonance imaging (MRI) diffusion scans of the brain interior and modeled abstractly as a set of time-independent geometric trajectories in a threedimensional brain space. Real neuronal fiber pathways exhibit complex but natural bundling structures, which elude existing MRI reconstruction techniques, but are easily captured by our algorithm. We validate our algorithms both theoretically (uniqueness of the graph representation and provably efficient algorithms) and empirically (using both synthetic and real scanned brain data sets).

show abstract

Section: Previous Workmentioning

confidence: 99%

A reeb graph approach to tractography

Sun

Cieslak

Grafton

et al. 2015

Proceedings of the 23rd SIGSPATIAL International Conference on Advances in Geographic Information Systems

View full text Add to dashboard Cite

show abstract

“…In a similar way, other types of patterns are defined. Thus, in a 'herd' [12], the entities in each time instant must form a dense spatial cluster, and so on.…”

Section: 1mentioning

confidence: 99%

“…In data mining and spatial computing, numerous algorithms have been developed for finding specific types of collective movement patterns in movement data [20]: 'flock', 'leadership', 'convergence', 'encounter' [21] [22], 'trend setting' (movements of some individual are repeated by others after a time lag) [17], 'moving cluster' [23], 'herd' [12], etc. ; Laube [3] and Gudmundsson et al [24] present reviews of the existing methods.…”

Section: Detection Of Collective Movement Patternsmentioning

confidence: 99%

Space Transformation for Understanding Group Movement

Andrienko

Barrett

Dostie

et al. 2013

IEEE Trans. Visual. Comput. Graphics

View full text Add to dashboard Cite

Abstract-We suggest a methodology for analyzing movement behaviors of individuals moving in a group. Group movement is analyzed at two levels of granularity: the group as a whole and the individuals it comprises. For analyzing the relative positions and movements of the individuals with respect to the rest of the group, we apply space transformation, in which the trajectories of the individuals are converted from geographical space to an abstract "group space". The group space reference system is defined by both the position of the group center, which is taken as the coordinate origin, and the direction of the group's movement. Based on the individuals' positions mapped onto the group space, we can compare the behaviors of different individuals, determine their roles and/or ranks within the groups, and, possibly, understand how group movement is organized. The utility of the methodology has been evaluated by applying it to a set of real data concerning movements of wild social animals and discussing the results with experts in animal ethology.

show abstract

“…• Flocks, group, herd detection (Benkert et al 2008;Buchin et al 2015;Huang et al 2008): Finding a special type of pattern determined by spatial proximity of a subset of the entities over a period of time. It is related to clustering, but in clustering we generally consider the whole trajectory when making clusters, whereas in grouping a single entity can be in different groups at different times, or even at the same time.…”

mentioning

confidence: 99%

On Measures for Groups of Trajectories

Wiratma

Kreveld

Löffler

2017

Lecture Notes in Geoinformation and Cartography

View full text Add to dashboard Cite

We present a list of measures for a single trajectory, including measures that require the presence of other trajectories, such as the centrality of a trajectory amidst other trajectories. Then, we introduce three different views in order to extend measures of a single trajectory to a group, namely the representative view, the complete view and the area view. Furthermore, we give measures that exist only for a group of trajectories, like density and formation stability. We also show that it may be possible to define new measures by combining trajectory data with data from other sources, such as the environment where the entities move. Finally, we discuss several tasks: settlement selection, visualization and segmentation, where measures on groups of trajectories are necessary.

show abstract

Modeling Herds and Their Evolvements from Trajectory Data

Cited by 38 publications

References 19 publications

A reeb graph approach to tractography

A reeb graph approach to tractography

Space Transformation for Understanding Group Movement

On Measures for Groups of Trajectories

Contact Info

Product

Resources

About