estimGQV2DH

Computes a Generalized Quadratic Variations based estimation of the Holder exponent for a 2D signal (an image)

Syntax

[H,G] = estimGQV2DH(x,gamma,delta,kv)
[H,G] = estimGQV2DH(...,'Regression type')
[H,G] = estimGQV2DH(...,'Propertyname',Propertyvalue)

Description

[H,G] = estimGQV2DH(x,gamma,delta,kv) Estimates the Holder function, H, and the scale factor, G, of the input image, x, using a least square regression. The paramaters gamma and delta are real values in (0:1) which characterize the neighborhood of each point where the exponent is computed. The vector kv gives the values of the succesive sub-samplings used for the computations of the GQV.
For each point the Holder exponent is estimated using a neighborhood of points.

[H,G] = estimGQV2DH(...,'Regression type') Estimates the Holder function, H, and the scale factor, G, using a specific type of regression. The Regression Type can be choosen from the list below :

[H,G] = estimGQV2DH(...,'Propertyname',Propertyvalue) returns the estimators H and G applying the specified property settings. The Property setting can be choosen from the list below:

Examples

See Also

estimOSC2DH

References

[1] A. Ayache, J. Lévy-Véhel, "Identification of the pointwise holder exponent of generalized multifractional brownian motion", Stochastic Processes and their Applications, Vol. 111 (2004) 119-156.

[2] O. Barrière, "Synthèse et estimation de mouvements Browniens multifractionnaires et autres processus à régularité prescrite. Définition du processus autorégulé multifractionnaire et applications", PhD Thesis (2007).