Abstract:
In this paper, a motion detection module
is proposed for real time dynamic video frames by
comparing the three major classes of methods for
motion detection namely Background Subtraction,
Temporal differencing and Optical Flow method .A
hierarchical background model is proposed based on
segmenting the background images. The region model
is extracted from the histogram of a specific region
which is similar to the kind of a Gaussian mixture
model. The pixel model is described by histograms of
oriented gradients of pixels in each region based on the
co-occurrence of image variations. Silhouette
detection algorithm is proposed. The experimental
results are carried out with a video database to
demonstrate the effectiveness, which is applied to both
static and dynamic scenes by comparing it with some
well-known motion detection methods namely
Temporal differencing and Optical Flow method and
based on the results a motion detection module for
dynamic video frames can be developed which is cost
effective, shows high rate of accuracy, low rate of
complexity, and well adapt to different kinds of shadow
distribution
