how to create 2D polynomial regression?

hi I have a 22x34 image. how to create a function just like ax^2+by^2+cxy+dx+ey+f to fit this image

 Respuesta aceptada

dpb
dpb el 30 de Dic. de 2015
If have CurveFitting Toolbox,
sf = fit([x, y], z, 'poly22'); % doc fit for details...
If not, write the specific model matrix and use \ to solve.
doc mldivide % illustrates
Oh, or if have Statistics Toolbox, there's the regress linear regression method.

Más respuestas (1)

Image Analyst
Image Analyst el 30 de Dic. de 2015

1 voto

You can use John D'Errico's polyfitn in the File Exchange: http://www.mathworks.com/matlabcentral/fileexchange/34765-polyfitn.
Attached is an example of how I used it to correct (flatten) a background.

Categorías

Más información sobre Linear and Nonlinear Regression en Centro de ayuda y File Exchange.

Etiquetas

Preguntada:

el 30 de Dic. de 2015

Comentada:

el 5 de En. de 2016

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by