ECE $5256$ Project $2$

Intensity Transformations

You can use any images for the following, but the project works best if the images are of poor contrast $($too dark, too bright, mostly gray...$).$

Compare two approaches to Histogram Equalization

Part $1$

a$)$ display an image

b$)$ plot its histogram

c$)$ display the histogram$-$equalized image

d$)$ plot the histogram$-$equalized image histogram

Part $2$

Repeat the steps in Part $1$ but with a different approach to histogram equalization.

Equalize the histogram of each $77$ region of an image separately. For images with sizes that are not a multiple of seven, simply extend the image with the same value as the last row or column.

Part $3_{?}$

Compare the results of the two approaches. Which method produces a more accurate histogram equalization $($flat histogram$)?$

$3.$ Problems

Please do the following problems from the text:

$3\mathrm{.}2($b$),($c$),3\mathrm{.}7$

$3\mathrm{.}2$ Exponentials of the form $e^{-r^2}$ with $$ a positive constant, are useful for constructing smooth intensity transformation functions. Start with this basic function and construct transformation functions having the general shapes shown in the following three figures. The constants shown are input parameters, and must be included in your answer.

$3\mathrm{.}7$ Obtain the unnormalized and the normalized histograms of the following $8-$bit, $MN$ image. Give your histogram either in a table or a graph, labeling clearly the value and location of each histogram component in terms of $M$ and $N.$ Doublecheck your answer by making sure that the histogram components add to the correct value.

Turn in:

$1$ Project title, course number, date due.

$2$ Brief description of what you have done.

$3$ Explanation and$/$or discussion of results for parts $1-3.$

$4$ Appendix: program listing.

$5$ Problem solutions.