Homogeneity of spatial variability [1]. This function divides the
  data pack (DATA) in spatial regions of WLines lines and WColumns 
  columns, in these regions are calculated activity indicators selected 
  with the variable Type, over these values are calculate the homogeneities.
  Known an activity indicator value A(i,j) in the window (i,j), the homogeneity 
  value H(i,j) is calculated as

  $Z=\{A(i,j-1), A(i-1,j), A(i,j), A(i+1,j), A(i,j+1)\}$

  $H(i,j) = \frac{StandardDeviation\{Z\}}{Mean\{Z\}}$

  * Is used the populational case of standard deviation.

  References:
  [1]  Cardoso, R.R. ; Braga, R.A. ; Rabal, H.J. Alternative protocols on 
       dynamic speckle laser analysis. SPIE 8413, V International Conference 
       on Speckle Metrology. 2012

  [2]  BRAGA, R.A. et al. Evaluation of activity through dynamic laser speckle 
       using the absolute value of the differences, Optics Communications, v. 284, 
       n. 2, p. 646-650, 2011.

  [3] R. Nothdurft and G. Yao, 'Imaging obscured subsurface inhomogeneity using 
      laser speckle,' Opt. Express  13, 10034-10039 (2005). 

  [4]  ARIZAGA, R. et al. Speckle time evolution characterization by the 
       co-occurrence matrix analysis. Optics and Laser Technology, Amsterdam, 
       v. 31, n. 2, p. 163-169, 1999.

  [5]  BRAGA R.A. CARDOSO, R.R. Enhancement of the robustness on dynamic speckle 
       laser numerical analysis. Optics and Lasers in Engineering, 
       63(Complete):19-24, 2014.

  After starting the main routine just type the following command at the
  prompt:
  [Y X] = homogeneity(DATA,WLines,WColumns,Type);
  
  Input:
  DATA     is the speckle datapack. Where DATA is a 3D matrix created grouping NTIMES 
           intensity matrices with NLIN lines and NCOL columns. When N=size(DATA), then
           N(1,1) represents NLIN and
           N(1,2) represents NCOL and
           N(1,3) represents NTIMES.
  WLines   is the number of lines in the analysed window.
  WColumns is the number of columns in the analysed window.
  Type     If Type is 1, it is used as activity indicator the AVD [2] technique.
           If Type is 2, it is used as activity indicator the Temporal S. Std. Deviation [3].
           In other case it is used as activity indicator the inertia moment [4] technique.
           In the cases of AVD and/or inertia moment indicators, it is used the Cardoso[5] normalization
           over co-occurrence matrix. In all cases, the activity indicators
           were calculated as a mean over all points of each window, not only over a line.

  Output:
  Y        is the homogeneity percentages in the analysed windows [1].
           The homogeneity value H(i,j) is represented as a window (matrix) 
           with WLines x WColumns pixels inside Y. 

  X        is the activity indicator value in the analysed windows. The activity 
           indicator value A(i,j) is represented as a window (matrix) with 
           WLines x WColumns  pixels inside X.


  For help, bug reports and feature suggestions, please visit:
  http://www.nongnu.org/bsltl

Package: bsltl