|
<< Click to Display Table of Contents >> General Tools - Mosaicking - Gradient Mosaicking |
![]() ![]()
|
Purpose
Terrain geocoded, radiometrically calibrated and normalized backscattering coefficient data acquired over different satellite orbits are usually mosaiced, making it possible to cover large areas.
It is not possible to use this functionality to mosaic data which are not partially overlapping. For this purpose the Conventional Mosaic>Last Pixel method is suggested.
Technical Note
Gradient mosaicing is implemented by creating cut lines within the overlapping areas of the images. These lines are automatically identified (using edge detection techniques) along image discontinuities, in order to minimize (by calibrating the image histograms) the visibility of the mosaic junction. It has to be pointed out that this method includes an automatic image calibration algorithm which is essentially articulated in three steps:
-The intensity ratio is estimated between neighbouring images in the overlapping areas.
-The correction factors are obtained by means of a global optimization.
-Outlier images are treated in advance, hence allowing more degree of freedom in the calibration step.
The adopted model for the images equalization is performed as follows:
-For standard images the applied calibration factor is
Ical (r,a) = Iuncal (r,a) . (k1 + k2 . r)
-For images at the borders the applied calibration factor is
Ical (r,a) = Iuncal (r,a) . (k2 . r)
-For temporal outlier images the applied calibration factor is
Ical (r,a) = Iuncal (r,a) . (k1 + k2 . r + k3 . a)
where r is the range coordinate, a the azimuth one.
The calibration process will result in the value of coefficients k1 , k2 , k3 for each single image. They are estimated as solution of a global minimization problem, in order to ensure the global continuity of the data radiometry. The relationship to be fulfilled for each possible image pair is
R . (k11 + k12 . r + k13 . a) / (k21 + k22 . r + k23 . a) = 1
where R is defined as the ratio between I1 and I2.
The calibration process is performed first with an "Absolute Calibration", which is intended for removing constant and linear trends in all input data; these trends are calculated by means of a certain number of points (Preferences>Mosaic-Filtering>Absolute Calibration) collected in the overlapping areas. The calibration is the locally refined by using a Krigging interpolation on a new grid of points (Preferences>Mosaic-Filtering>Local Calibration), which are located along each cut line where each image will be mosaicked to the adjacent one.
Input Files
Input file list
Input file list of the geocoded data (_geo). This file list is mandatory.
Parameters - Principal Parameters
Absolute Calibration
An absolute and linear calibration is carried out considering the pixel values in the overlapping area.
Local Calibration
The previous absolute calibration is locally refined around the cut line within the overlapping area.
Use Existing Shape
If a previously generated shape (possibly edited) exists, it will be used as mosaic cut line.
Parameters - Global
It brings to the general section of the Preferences parameters. Any modified value will be used and stored for further processing sessions.
Parameters - Other Parameters
It brings to the general section of the Preferences parameters. Any modified value will be used and stored for further processing sessions.
Output Files
Output file
Output file name of the mosaiced data. This file is mandatory.
_msc
Mosaic and associated header files (.sml, .hdr).
_shape.shp
Shape file containing the mosaic cut line.
Details specific to the Units of Measure and Nomenclature of the output products can be found in the Data Format section.
General Functions
Exec
The processing step is executed.
Store Batch
The processing step is stored in the batch list. The Batch Browser button allows to load the batch processing list.
Close
The window will be closed.
Help
Specific help document section.
Specific Function(s)
None.
Task, SARscapeBatch object, SARscapeBatch script example
References None.