Introduction to Artificial Intelligence
Worksheet #8
Image Processing 1: Binary Images and Detecting Connected Components


Read the Help page on starting Image Processing (Lisp version).

  • Preliminary work: Run the two programs from "Starting Image Processing
    with Lisp" (see the link above) - arrayStarter.lsp and readFile.lsp.
    Understand how each works before moving on. Also read about image file
    formats PGM and PPM and the introduction to the concept of 4 and 8
    connectivity of components in images.
    1. A. Copy the image file: image1.pgm.
      B. Load the image using xv and magnify sufficiently so you can see the 8 by 8 image.
      (xv & then right click for the menu)
      C. Copy the starter Lisp file: /home/atlas1/ai/connect4Starter.lsp.

    2. A. Threshold the image so that all values below 100 are set to 0 and
      all values equal to or above 100 are set to 255. Write the image
      to a file named "binary1.pgm". The file should look like this.
      This file should have the following header lines: 
         P2
   # Comment line identifying this as a PGM binary image, binary.pgm
   8 8
   255
      - the image pixel values should be here, 8 rows and 8 cols.

      B. View binary1.pgm with xv to check if the objects are
      in black against a white background.

    3. A. Complete the program connect4.lsp that labels and counts all
      "4-connected objects" in the picture. Print both the binary
      image and the labeled component image on the screen.
      View a sample output.
      B. Write connect8.lsp to label and count the "8-connected" objects
      in the image. View sample 8-connected labeled image.