Image Processing

Gül Eda Aydemir - Contact with me! :)

www.linkedin.com/in/gul-eda-aydemir

guledaaydemir@gmail.com

CME 4412 Introduction to Image Processing

1. Spatial Resolution

Using the image Rose1024.tif obtain the subsampled versions of the image as given given in the presentation, i.e. obtain the images of 512x512, 256x256, 128x128, 64x64 and 32x32 rose images.

2. Tresholding

Perform thresholding on moon.bmp image to obtain the corresponding result in the presentation.

3. Intensity Level Resolution

Use Ctskull-256.tif image to obtain different quantization levels of the image starting from 8 bits (original skull image), 7 bits, 6 bits, 5bits, 4 bits, 3 bits, 2 bits, and single bit images.

4. Power Law Transformation

Use power law transformation with c=1 and γ = 0.6, 0.4, and 0.3 to obtain resulting images on fractured_spine.tif image

5. Piecewise Linear Transformation Functions

Write a Matlab function output = stretch(input, T1, T2) that will perform the pixel transformation shown in figure given below.

Use your stretch function to increase the contrast of the kids.tif image. Choose T1 and T2 so that the output image has a histogram which spans the full range of values from 0 to 255. Display the output image and its histogram.

6. Histogram Equalization

Write a Matlab function Y=equalize(X) that equalizes the histogram of a given image X. Apply your equalize function to different images.Compare the histograms of equalized and original images.

Compare your function with the built-in histogram equalization function of MATLAB.

7. Histogram Matching

Write a Matlab function Y=match(X,Z) that matches the histogram of a given image X to the histogram of a reference image Z. You can use your own equalize() function in your implementation.(*imhist used instead)

8. Combining Histogram Equalization and Matching with GUI

Try to embed these two functions (equalize and match) in a Matlab GUI.

Step 1. Select a function (Histogram Equalize)
Step 2. Browse an image – Show result
Step 3. Select a function (Histogram Matching)
Step 4. Browse an image
Step 5. Browse a referance image - Show Result

9. Spatial Filtering

Generate a filter mask using the following code:

radius=3; filter=ones(radius,radius)/(radius*radius);

Perform spatial filtering with the above generated template and display the output image. Change radius to 5 or 7, and use the newly generated masks to filter the original image. Compare the difference between different outputs.

Generate another filter mask using the following code:

base_filter=[0 0 0 0 0;1 1 1 1 1;0 0 0 0 0;-1 -1 -1 -1 -1;0 0 0 0 0]; angle=0; filter=imrotate(base_filter,angle,'crop');

Perform spatial filtering with the above mask and display the output image. Change angle to 45, 90 and 135. Use the newly generated masks to filter the original image. Compare the difference between different outputs.

10. Sobel Operators

Download the two_cats.jpg image and store it.

Load the image data.
Use a spatial filter to find and display the horizontal edges of the image.
Use a spatial filter to find and display the vertical edges of the image hint: read the MATLAB documentation on fspecial