Module imagingbook_plugins_book
Package Ch08_Binary_Regions

Class Flusser_Moments_Covariance_Matrix

java.lang.Object
Ch08_Binary_Regions.Flusser_Moments_Covariance_Matrix
All Implemented Interfaces:
PlugIn, JavaDocHelp
public class Flusser_Moments_Covariance_Matrix extends Object implements PlugIn, JavaDocHelp

This ImageJ plugin calculates the covariance matrix for the 11-element Flusser moment vectors over a collection of binary images. The plugin assumes binary images (with 0 background and non-zero foreground) and calculates the 11 scale and rotation invariant Flusser moments (sec. G.3 of [1] for details) for the largest contained region contained in each image of the data set . The resulting covariance matrix is output in float and bit-encoded long format to avoid precision loss. The latter can be converted to double using Matrix.fromLongBits(long[][]). The covariance matrix is used as a parameter to the Mahalanobis distance for matching moment vectors (see Flusser_Moments_Matching_Demo).

[1] W. Burger, M.J. Burge, Digital Image Processing – An Algorithmic Introduction, 3rd ed, Springer (2022).

Version:
2022/12/28
See Also: