SEGMENTATION AND SAMPLING OF MOVING OBJECT TRAJECTORIES BASED ON REPRESENTATIVENESS

SEGMENTATION AND SAMPLING OF MOVING OBJECT TRAJECTORIES BASED ON REPRESENTATIVENESS

Moving Object Databases (MOD), although ubiquitous, still call for methods that will be able to understand, search, analyze, and browse their spatiotemporal content. In this paper, we propose a method for trajectory segmentation and sampling based on the representativeness of the (sub)trajectories in the MOD. In order to find the most representative subtrajectories, the following methodology is proposed. First, a novel global voting algorithm is performed, based on local density and trajectory similarity information. This method is applied for each segment of the trajectory, forming a local trajectory descriptor that represents line segment representativeness. The sequence of this descriptor over a trajectory gives the voting signal of the trajectory, where high values correspond to the most representative parts. Then, a novel segmentation algorithm is applied on this signal that automatically estimates the number of partitions and the partition borders, identifying homogenous partitions concerning their representativeness. Finally, a sampling method over the resulting segments yields the most representative subtrajectories in the MOD. Our experimental results in synthetic and real MOD verify the effectiveness of the proposed scheme, also in comparison with other sampling techniques.

Existing System:

This is a very challenging problem where very limited work has been carried out so far. An insightful solution to the problem would be an analyst to be able to supervise the sampling procedure, not only regarding the volume of the sampled data set, but also the properties of the data set that reveal the underlying movement patterns of the MOD.
We argue that this problem can be effectively tackled if interconnected to the previous two discussed problems. In other words, we propose an automatic method for sub trajectory sampling based on the “representativeness” of the sub trajectories.
In this approach, an analyst may request the top-k representative subtrajectories that best describe the MOD in an optimized way, where optimization is with respect to the “representativeness.”

Proposed System:

We propose an index-based global voting method that allows us to represent the representativeness of a trajectory in a MOD as a smooth continuous descriptor.
We introduce an algorithm for the automatic segmentation of trajectories into “homogenous” sub trajectories according to their “representativeness” in the MOD.
We define the problem of sub trajectory sampling in a MOD as an optimization problem and we propose a novel solution to tackle the problem.
Finally, we conduct a comprehensive set of experiments over synthetic and real trajectory data sets, in order to thoroughly evaluate our approach.

Software Requirements:
.Net
Front End – ASP.Net
Language – C#.Net
Back End – SQL Server
Windows XP
Hardware Requirements:
RAM : 512 Mb
Hard Disk : 80 Gb
Processor : Pentium IV

FUTURE ENHANCEMENT:

We plan to investigate the applicability of the proposed method for (sub)trajectory clustering. The idea is that MOD clustering can be provided concurrently with MOD sampling. It holds that each subtrajectory of the sampling set has been voted by different subtrajectories of the MOD (cluster), under the minimization of the objective function proposed in the current work. Therefore, each subtrajectory of the sampling set can be considered as a cluster representative (i.e., a seed around which a cluster is formatted). This is a different tactic as the one followed the same context, outliers can be discriminated from low values in voting subtrajectory descriptor

Comments are closed.