|
<< Click to Display Table of Contents >> Basic Module - Intensity Processing - Filtering - Single image Filtering |
  
|
|
<< Click to Display Table of Contents >> Basic Module - Intensity Processing - Filtering - Single image Filtering |
|
Purpose
Images obtained from coherent sensors such as SAR (or Laser) system are characterized by speckle. This is a spatially random multiplicative noise due to coherent superposition of multiple backscatter sources within a SAR resolution element. In other words, speckle is a statistical fluctuation associated with the radar reflectivity of each pixel of a scene. A first step to reduce the speckle - at the expense of spatial resolution - is usually performed during the multi-looking, where range and/or azimuth resolution cells are averaged.
Technical Note
For fully developed speckle it is well know that a multiplicative fading random process F is an appropriate model:
I = R . F
where I is the observed intensity (speckled measured radiance), R is the random radar reflectivity process (unspeckled radiance), F is a stationary fading random process statistically independent of R, with unit mean <F> = 1 and whose variance is inversely proportional to the effective number of looks L. The mean intensity <I> = R is proportional to the backscattering coefficient of the pixel.
Speckle Specific Filters
The most well known adaptive linear filters are based on the multiplicative model and the use of the local statistics. The Frost filter is an adaptive Wiener filter, and convolves the pixel values within a fixed size window with an adaptive exponential impulse response. The Lee filters perform a linear combination of the observed intensity and of the local average intensity value within the fixed window. They are all adaptive as a function of the local coefficient of variation and can be enhanced by fixing a minimum value for better speckle smoothing and an upper limit texture or point target preservation. The coefficient of variation is a good indicator of the presence of some heterogeneity within the window; it is well adapted when only isotropic texture is present and it can be assisted by ratio operators for anisotropic oriented textural features.
Input Files
Input file list
Input file names (e.g. _pwr, _rsp, _geo). This file(s) is mandatory.
Parameters - Principal Parameters
Depending on the chosen filtering method, one or more of the following fields will be activated.
Filter method:
Mean, Median, Mode. Active parameters:
-Azimuth window size
-Range window size
Edge Preserving Smoothing. Active parameters:
-Azimuth window size
-Range window size
-Iterations Number
-Directionality Number
Frost, Lee, Refined Lee. Active parameters:
-Azimuth window size
-Range window size
-Equivalent Number of Looks (ENL)
Equivalent number of looks (ENL)
The Equivalent Number of Looks is equivalent to the number of independent Intensity values averaged per pixel during the multi-looking process. This parameter can be easily estimated over a homogeneous (stationary) sample in the input Intensity data according to:
ENL = mean2 / standard deviation2
In case that ENL is not set, the software tries to retrieve it automatically; if it fails it takes the Number of Looks (NL) used during the multi-looking process is considered.
Note that, to tune the strength of speckle filtering and the level of preservation of scene details, it is preferable to adjust the value of the ENL, rather than to change the size of the processing window:
•To reduce the strength of speckle filtering, with the aim to preserve the thinnest details of the scene, enter a ENL value slightly higher than the calculated one;
•Inversely, to improve the filtering of the speckle (possibly at the cost of the thinnest details of the scene), enter a ENL value slightly lower than the calculated one.
Azimuth window size
Size – in pixel units – of the moving window in azimuth.
Range window size
Size – in pixel units – of the moving window in range.
Iteration number
Iteration times.
Directionality number
Depending upon the window size, different directions – in degree unit – can be considered during the filtering. An increase in the number of directions corresponds to a better preservation of the structures.
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 list
Output file names. This file(s) is mandatory.
_fil
Filtered Intensity image and associated header files (.sml, .hdr).
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 Aspert F., M. Bach Cuadra, J.P. Thiran, A. Cantone, and F. Holecz: "Time-varying segmentation for mapping of land cover changes". Proceeding of ESA Symposium, Montreux, 2007. Frost V.S., J. Stiles, K. Shanmugan and J. Holtzman: "A model for radar images and its application to adaptive digital filtering of multiplicative noise". Transactions on Pattern Analysis and Machine Intelligence, Vol. 4, No. 2, 1982. Lee J.S.: "Speckle suppression and analysis for SAR images". Optical Engineering, Vol. 25, No. 5, 1986. Nagao M. and Matsuyama: "Edge Preserving Smoothing". Computer Graphics and Image Processing, Vol. 9, 1979.