Speckle Filter

Multi-Temporal Speckle Filter Operator

SAR images have inherent salt and pepper like texturing called speckles which degrade the quality of the image and make interpretation of features more difficult. The speckles are caused by random constructive and destructive interference of the de-phased but coherent return waves scattered by the elementary scatters within each resolution cell. Multi-temporal filtering is one of the commonly used speckle noise reduction techniques.

Multi-Temporal Speckle Filtering

For a sequence of N registered multitemporal images, with intensity at position (x, y) in the kth image denoted by Ik(x, y), the temporal filtered images are given by:




   for k = 1, ..., N, where E[I] is the local mean value of pixels in a window centered at (x, y) in image I.

Pre-Processing Steps

   The operator has the following two pre-processing steps:
  1. The first step is calibration in which σ0 is derived from the digital number at each pixel. This ensures that values of from different times and in different parts of the image are comparable.
  2. The second step is registration of the multitemporal images.
   Here it is assumed that pre-processing has been performed before applying this operator. The input to the operator is assumed to be a product with multiple calibrated and co-registered bands.

Parameters Used

   The following parameters are used by the operator:
  1. Source Band: All bands (real or virtual) of the source product. User can select one or more bands for producing the filtered image. If no bands are selected, then by default all bands will be selected.
  2. Window Size: Dimension of the sliding window that is used in computing spatial average in each image of the temporal sequence. The supported window sizes are 3x3, 5x5, 7x7, 9x9 and 11x11.



Reference: S. Quegan, T. L. Toan, J. J. Yu, F. Ribbes and N. Floury, “Multitemporal ERS SAR Analysis Applied to Forest Mapping”, IEEE Transactions on Geoscience and Remote Sensing, vol. 38, no. 2, March 2000.