Next: Detection of Fire Colored
Up: Detection Algorithm
Previous: Detection Algorithm
Moving pixels and regions in the video are determined by using a hybrid
background estimation method developed by Collins et al. [6].
Let
represent the intensity (brightness) value at pixel position
in the
image frame
. Estimated background intensity
value at the same pixel position,
, is calculated as follows:
![\begin{displaymath}
B_{n+1}[k,l] = \left\{ \begin{array}{ll}
aB_n[k,l]+(1-a)x_n[...
...[k,l] & \textrm{if $[k,l]$\ is moving,}\\
\end{array} \right.
\end{displaymath}](img8.png) |
(1) |
where
is the previous estimate of the background intensity value
at the same pixel position. The update parameter
is a positive real
number close to one. Initially,
is set to the first image
frame
.
A pixel positioned at
is assumed to be moving if the brightness
values corresponding to it in image frame
and image frame
satisfy the following inequality:
![\begin{displaymath}
\vert x_n[k,l] - x_{n-1}[k,l]\vert > T_n[k,l],
\end{displaymath}](img14.png) |
(2) |
where
is a threshold that is adaptively determined as described
in [6]. It is assumed that the regions significantly different
from the background are moving regions. Estimated background image is subtracted
from the current image to detect moving regions. In other words, all of
the pixels satisfying
![\begin{displaymath}
\vert x_n[k,l] - B_n[k,l]\vert > T_n[k,l]
\end{displaymath}](img16.png) |
(3) |
are determined. These pixels at
locations are grouped into connected
regions (blobs) and labeled by using a two-level connected component labeling
algorithm [12]. The output of the first step of the algorithm
is a binary pixel map
that indicates whether or not the pixel at location
in image
is moving.
Other more sophisticated methods, including the ones developed by Bagci et al. [2]
and Stauffer and Grimson [21], can also be used for moving pixel estimation.
In our application, accurate detection of moving regions is not as critical as in
other object tracking and estimation problems; we are mainly concerned with real-time detection of moving regions as an initial step in the fire and flame detection system. We choose to implement the method suggested by Collins et al. [6],
because of its computational efficiency.
Next: Detection of Fire Colored
Up: Detection Algorithm
Previous: Detection Algorithm
ugur toreyin
2005-11-27