Background Models for Tracking Objects on Water Surface
Loading...
Files
Date
2016-07-14T11:38:28Z
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Object tracking is the process of segmenting an object of interest from a video scene. The
main goal of object tracking is to locate a moving object over time using camera. It is under
the domain Image Processing. This project presents a Novel Background Analysis (NBA)
technique to enable robust tracking of objects in water based scenarios. In water based
scenarios, waves caused by wind or by moving vessels (wakes) form highly correlated
moving patterns that confuse traditional background analysis models.
In this work we introduce Detecting Contiguous Outliers in Low Rank Representation
(DECOLOR) and Adaptive Dynamic Group Sparsity (AdaDGS) methods that explicitly
models this type of background variation. Using DECOLOR method, the background is
registered and then subtracted by comparing frames, resultant of which is a frame that
contains only object, that is sent to AdaDGS method, where noise is eliminated by grouping
the similar vector and removing the dissimilar ones. Then by using BLOB analysis method,
the region of the object is masked. Therefore, we obtain an image where high amount of
noise is eliminated.
These methods can converge well to the current background dynamics more rapidly and
accurately and also works well in highly textured environments. Therefore, effectively
removes the noise and thereby increases the quality of the image.
Description
Keywords
Background Models for Tracking Objects on Water Surface, Pratheeksha J, Shreeya P, Susmetha K C, Nandini M