Borrar filtros
Borrar filtros

How to eliminate intensity bias from an image?

3 visualizaciones (últimos 30 días)
Jyothis Jose
Jyothis Jose el 4 de Mayo de 2014
Comentada: Image Analyst el 4 de Mayo de 2014
I need to eliminate intensity bias fro a large image. Please help me with its matlab code.
  2 comentarios
Image Analyst
Image Analyst el 4 de Mayo de 2014
What does that mean? Why would your images have a bias? Do you mean just the average intensity? Or do you mean that the intensity has a bias higher or lower than the "true" intensity by some amount? Does what you call bias vary from pixel to pixel or is it the same for every pixel in the image?
Jyothis Jose
Jyothis Jose el 4 de Mayo de 2014
Retinal images are plagued by imaging artifacts. The most common among them is the introduction of illumination bias in CFI due to the spherical nature of the retina. This is also known as bias field.Since the green channel of CFI provides maximum contrast for distinguishing between structures and lesions, the bias field estimation and removal is done in the green channel.
This is said about bias field in one of my reference papers. I will attach the reference paper

Iniciar sesión para comentar.

Respuestas (1)

Image Analyst
Image Analyst el 4 de Mayo de 2014
It looks like what is normally called "background correction". For most image modalities, that is accomplished by getting a smooth background image and dividing the test image by it. For some situations (radiography, fluorescence) you do background subtraction instead of background division. To get the background, it looks like they just blurred the green channel enough to get rid of vessels. You can do this with conv2() or imfilter().
  2 comentarios
Jyothis Jose
Jyothis Jose el 4 de Mayo de 2014
imfilter(A, H).What is H here? what is the basis of assigning values for H while using this function?
Image Analyst
Image Analyst el 4 de Mayo de 2014
H is the window that slides along your A image, multiplies element by element H by your image A.

Iniciar sesión para comentar.

Categorías

Más información sobre 3-D Volumetric Image Processing en Help Center y File Exchange.

Community Treasure Hunt

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

Start Hunting!

Translated by