Intensity Transformations (Chapter 3). CS474/674 – Prof. Bebis. Spatial Domain Methods. f(x,y). g(x,y). Point Processing. g(x,y). f(x,y). Area/Mask Processing. Point Processing Transformations. Convert a given pixel value to a new pixel value based on some predefined function. .
CS474/674 – Prof. Bebis
we desire more information
(slope > 1).
are of little interest
(0 < slope < 1).
Same as double
X: # of heads
non-decreasingProbability distribution function
The intensity levels can be viewed as a random variable in [0,1]
For PGM images:
k=0,1,2, …, L-1 (possible graylevels)
rk=k/(L-1) (normalized graylevel in [0, 1])
sk x (L-1)
64 x 64 image
original images and histograms
equalized images and histograms
fZ(a)daHistogram Specification (Matching)
64 x 64 image
results of histogram equalization
results of histogram specification
Define a transformation function based on the intensity distribution in a neighborhood of every pixel in the image!
1. Define a neighborhood and move its center from pixel to pixel.
2. At each location, the histogram of the points in the neighborhood is computed. Obtain histogram equalization or histogram specification transformation.
3. Map the intensity of the pixel centered in the neighborhood.
4. Move to the next location and repeat the procedure.
3 x 3 neighborhood
σ is useful for estimating image contrast!
Task: enhance dark
areas without changing
Idea: Find dark, low contrast
areas using local statistics.
[fmin – fmax] [ 0 – 255]