5 Notch Filtering

Use function notchfilter to denoise FigP4.5(a)(HeadCT_corrupted).tif. Show the

original image, an image of your specified filter, and the restored result. If you completed
Project 4.5, compare your result with the image denoised using bandreject filtering in that
project. If you use function pixval to determine the coordinates of the impulses
interactively, keep in mind that this function lists the column coordinates (v) first and the
row coordinates (u) second.

7
8

Solution steps:
1. Reading the original image.
2. Generating a frequency domain notch filter
3. Filtering operation.
4. Converting class and scale.
5. reading the image result from assignment no. 2 problem 5.
6. subtracting the result of notch filtering that of bandreject filtering.
 7-8. Extracting the noise pattern by removing the image.

Figure 1: The original Image (image corrupted by noise)

Figure 2: Notch filtered image.

Figure 3: Band reject filtered image (solved in assignment no.2)

Figure 4: Difference of Notch filtered image and band reject filtered image.
(the difference shows that the notch filtered image is more sharper).
Figure 5: The noise pattern extracted from the original image.