Abstract:
Road intersections are more prone to traffic congestion, which leads to traffic accidents. It is important to
monitor the traffic congestion at crossings for regulating the
driver behaviour and preventing the accidents. As real time
tracking systems rely on the accuracy of the system, an approach
has been proposed for vehicle tracking. This paper describes a
real time tracking approach for non-linear systems. The occluded
vehicle is extracted from the image sequences by subtracting the
image from the modelled background. Vehicles are tracked using
modified fractional order Kalman filter with better accuracy.
The non-linearity of the system is linearised using Jacobian.
The impact of behaviour of vehicle on error covariance has
been reduced using modified transition matrix. The fractional
states are calculated using GL fractional derivative definition.
The proposed method is tested for various motion models and is
evaluated using root mean square error with different data sets.
It has been shown that the root mean square error has reduced
using the proposed method.
