Image Processing Lab 1
Ted Gould
Bryan Evenson
For the first few images we used a graphics package called  The GIMP  to manipulate the files.  For the last objective (fourier transform) we used Matlab.
  1. Grab a 512x512 image

  2. Explore Spacial Digitization
  3. 256x256 - Looks about the same as the 512x512 image...

    128x128 - We are starting to notice some blurring in the image

    64x64 - The lady is not recognizable anymore, along with most objects in the picture.


  4. Explore Quantization
  5. 128 grays - No noticeable difference

    64 grays - We still don't notice much of a difference, this may be because it is such a dark image already.

    32 gray - Starting to notice a little bit of false contouring on her face.

    16 gray - The false contouring is now more obvious.

    8 gray - Objects are starting to lose their definition.  Noticeable on the lady's hand, and the phone.

    4 gray - Now the image looks like it is from the 1980s of computing :)

    2 gray - It is hard to figure out what is in the image without previous knowledge of what it is.  It is interesting that the credit card name and number are still readable.


  6. Look at FFT

  7. Matlab Code:
    [X, MAP] = tiffread('mot.512x512.tiff');
    imshow(32 * log10(1 + abs(fftshift(fft2(X)))), MAP);