×

You are using an outdated browser that does not fully support the intranda viewer.
As a result, some pages may not be displayed correctly.

We recommend you use one of the following browsers:

Full text

Title
The 3rd ISPRS Workshop on Dynamic and Multi-Dimensional GIS & the 10th Annual Conference of CPGIS on Geoinformatics
Author
Chen, Jun

ISPRS, Vol.34, Part 2W2, “Dynamic and Multi-Dimensional GIS", Bangkok, May 23-25, 2001
134
Q(x) = a 0 +a i x + ... + a m _
Then
The correlation between t and several channels’ reflectivity (Ri)
or radiant lightness can express as follows:
S = A + Y j B,R, +c,r 2 ,
/=1
A, B, Bi, Ci are constant
According to the result of a certain thickness of oil film,
analysis reflectivity and the least square method and the range
of ware length in all channels we can get a relative value of
every certain thickness of oil film in all channels.
R , = A, + B (f 1 R(A)dX)
JA. n
Landsat has seven channels, NOAA has five ones. So
&tm — ^ ^ BjR-i
/=]
Based on the sample data we can construct different oil film
feature data line, and build discriminating function:(fig. 2,fig.3,
Fig. 2: the Featured-Data-Line-- Window of heavy diesel oil
Fig.3: the Featured-Data-Line-- Window of lube
Fig.4: the Featured-Data-Line- Window of crude oil
Fig. 5: the Featured-Data-Line- Window of light diesel oil
Because reflectivity of oil film change along with thickness of it
in certain extend, this change has nothing to do with
wavelength but only decided by the thickness of oil film. The
different thickness of same kind of oil film just likes a series
locomotion locus, which only has level locomotion along Y
coordinate axis direction. So we decide to make a window in
the center of featured data line. The size of it is an alterable
parameter that can be decided by the quantity of spilled-oil and
the situation of incident. By doing this featured data line
become a range of Grey intensity.
Discriminate the Grey intensity of all pixels to make sure if the
Grey intensity are belong to this windows’, if it is information of
oil film can be extracted.
4. Software of Auto-extracting the Sea Oil Information and
result analysis
This software is exploited based on VC6.0++, every function
has three parts: input data file, process data file and output
result file. The data files include two categories that are 1 df
that belong to NOAA and data that belong to TM. We process
data file by the method of Featured-Data-Line- Window (FDL-