Image Processing Reference
FIGURE 3 Color matrix for Contrast image filter.
The main advantages of using “color matrix” optimization technique are the significant in-
crease in execution time (8 times for Contrast filter) and the fact that the contrast of each com-
ponent, e.g., Red can be increased in a different manner than another, e.g., Green.
The drawback of “color matrix” optimization techniques is that the one and the same set of
transformations must be applied to all pixels in the image.
This optimization technique should be used when the image filter performs many compu-
tations per pixel such as Contrast.
2.2 New Image Filter: Smart Contrast
Smart Contrast performs two similar transformations depending on the value of the pixel
component. For less than 127 (255/2) values, the formula is illustrated by
Equation (4) shows the formula for values of the pixel components greater than 127.
Value 127 is the threshold for Smart Contrast filter and the value of the threshold is chosen
to be the median value in the range from 0 to 255 (i.e., 0 is lowest value and 255 the highest
value of color intensity).