122
Fig. 9. Schematic diagram of
the evolution of the St.
Lawrence Island polynya from
February 21 to 24 to 27,1992.
The solid lines are the edges of
polynya and the closed contours
are the matched ice floes for ice
tracking
6.3.3
Tracking of Ice Floes
a
50
km
A. K. LIU AND C. Y. PENG
100
An algorithm for tracking ice floes in the interior of ocean basins has been developed
by Kwok et al. (1994) by matching distributions of pixel intensity in a fixed-size window scanning a sequence of SAR images. It has been used by ASF at the University of
Alaska to provide operational maps of ice flow vectors in the Bering and Chukchi Seas.
In the MIZ and the polynyas, because the processes are so dynamic, ice motion tracking at 3-day intervals on a grid-to-grid basis may become very challenging. Although
most ice floes in these areas may change beyond recognition in a 3-day interval, some
large ice floes may still be identifiable and can be tracked. In this study, an algorithm
for ice-floe tracking in these areas has been developed for demonstration. It is based
on template matching of the ice floe shapes approximated by closed contours of ice floes
obtained from the wavelet transform. Details of the procedures are described in the following case study.
The procedure for ice floe tracking is similar to the tracking of ice edge. We first compute the wavelet transform of SAR images of St. Lawrence Island on days 21,24, and 27
with a scale a = 16, corresponding to a typical ice floe size in this area. Then, we plot
the contours of the wavelet transform with a contour level setting to 5% of the maximum. With a finite contour level instead of a zero value, edge elements corresponding
to noisy features are effectively filtered out, and closed contours are generated except
near the edge of the image. Each closed contour corresponds to the approximate
boundary of a large floe or an area with the same texture, as shown in Fig. loa, b for
Fig. 9. Schematic diagram of
the evolution of the St.
Lawrence Island polynya from
February 21 to 24 to 27,1992.
The solid lines are the edges of
polynya and the closed contours
are the matched ice floes for ice
tracking
6.3.3
Tracking of Ice Floes
a
50
km
A. K. LIU AND C. Y. PENG
100
An algorithm for tracking ice floes in the interior of ocean basins has been developed
by Kwok et al. (1994) by matching distributions of pixel intensity in a fixed-size window scanning a sequence of SAR images. It has been used by ASF at the University of
Alaska to provide operational maps of ice flow vectors in the Bering and Chukchi Seas.
In the MIZ and the polynyas, because the processes are so dynamic, ice motion tracking at 3-day intervals on a grid-to-grid basis may become very challenging. Although
most ice floes in these areas may change beyond recognition in a 3-day interval, some
large ice floes may still be identifiable and can be tracked. In this study, an algorithm
for ice-floe tracking in these areas has been developed for demonstration. It is based
on template matching of the ice floe shapes approximated by closed contours of ice floes
obtained from the wavelet transform. Details of the procedures are described in the following case study.
The procedure for ice floe tracking is similar to the tracking of ice edge. We first compute the wavelet transform of SAR images of St. Lawrence Island on days 21,24, and 27
with a scale a = 16, corresponding to a typical ice floe size in this area. Then, we plot
the contours of the wavelet transform with a contour level setting to 5% of the maximum. With a finite contour level instead of a zero value, edge elements corresponding
to noisy features are effectively filtered out, and closed contours are generated except
near the edge of the image. Each closed contour corresponds to the approximate
boundary of a large floe or an area with the same texture, as shown in Fig. loa, b for