log transformations

asked 2013-04-03 06:42:39 -0500

defunktlemon gravatar image

updated 2013-04-03 06:48:14 -0500

Hi. I'm trying to understand the image analysis process. Could somebody help me to understand when considering log transformations it is said that transformation is used to expand values of dark pixels, while compressing higher-level values, please?

S = c log (1 + r)

I think I am confused by expanding and compressing. I thought the transformation would be like a threshold where grey pixels above a threshold would become white and below black, effectively making a black and white image.

Thanks

edit retag flag offensive close merge delete

Comments

Hi. Please let me know if I am right - I was wrong in my above assumptions, I believe.

The transformation takes a narrow range of grey-level pixels in the input image and maps them into a wider range in the output image.

So, in a Fourier spectrum, for example, the dominant white pixels will become more grey and the unseen black, or near black, will become visible. So, it lessens large variations in pixel values.

Why is this useful. Is it used for things like making image negatives? So, for example, if I were to perform the inverse log transform, would this make a negative image which would be useful in things like mammograms?

defunktlemon gravatar imagedefunktlemon ( 2013-04-03 07:14:33 -0500 )edit