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
