11-05-2012, 01:14 PM
Mining Group Movement Patterns for Tracking Moving Objects Efficiently
Abstract:
The object tracking with similar movement patterns as find as using a single object or group of object. In local mining phase used to finds movement patterns based on the local trajectories. This is derived on the movement patterns and moving object with similar single object or group of object. To address the energy conservation issue in resource-constrained from transmits local grouping results. And the mining results to track moving the object efficiently, at the same time data mining algorithm achieves to reduce the energy consumption by reducing the amount of data to be transmitted from one local to another local group mining. This information is important in biological research domains, clustering data search in network credentials, and wild life migration, etc.,
Existing System:
• These applications generate large amounts of location data, and many approaches focus on compiling the collected data to identify the repeating movement patterns of objects of interest.
• We find that discovering their movement patterns of a group of objects is more than difficult than finding the patterns of a single object because we need to identify a group of object before or after discovering their movement patterns.
• The object next location we can be predicated based on its preceding locations.
• But now we used to conditional probability distribution, so we get over all of the object location in sequence dataset.
• A smaller group of relationship data’s we control the movement of the range, and
A group of objects by using linear distance between the starting points to furthest point to reached.
Proposed System:
A distributed mining algorithm identifies a group of objects with similar movements patterns.
• It’s used comprises a local mining phase and cluster ensembling phase
• Data collect from locally and generates the group of information with GMPMine algorithm.
• CE algorithm comprised of three steps., first collect the similarly coefficient pair of objects. That means presently moving object is present in partial clusters and absent from others.
• Second the coefficient object are same group or different group, is that simple match that coefficient underestimates the object’s correlations
• Final step find the normalized mutual information to select the ensembling result from the group of objects.
• In network data aggregation in improves the scalability and reduces the long-distance of communication demands and thus saves energy.
Some of the main advantages of the proposed system are as follows
Reduces the energy,
Eliminate redundant update traffic and ,
Limit the flooding attacks.
SOFTWARE REQUIREMENT: -
Front End/GUI Tool : Microsoft Visual studio 2010
Operating System : Windows Family
Language : C#.NET
Application : Windows Application
Database : Sql server
HARDWARE REQUIREMENT: -
Processor : Pentium series
RAM : 1 GB
Hard Disk Drive : 80 GB