ECE 563 Assignment 4

Images and m-files for this assignment may be downloaded from impro4.zip and MacbethColor.zip.

Individual Work

  1. Use the co-occurance histogram method in cohist.m from the handout for the pairs (RG, RB, and GB) for your scanned color photograph.

Group Work

  1. Use morphological operations to count the teeth in the image gear_tooth.jpg from the second assignment.

  2. Using a binary image of a collection of objects from previous assignments, find the centroid of each object (using regionprops), and create an overlay image with a yellow crosshair (+ or x shaped) superimposed on each object at the centroid. Use the matlab plot function to create these crosshairs.

  3. For the three images in MacbethColor.zip, find the mean RGB values for each of the 24 squares.

  4. Find the mean and rms of the grayscale squares (19-24) in the bottom row of the three Macbeth color images (from above). Plot the mean value vs. Y using values from MacbethGray.xls. Find the best-fitting power-law exponent for each image.

  5. Compare the results obtained from regionprops, the code find_ellipse.m, and the code diameter.m on the image ellipse.tif. Use ellipse.m from impro3.zip to generate a variety (three or four) of ellipses and repeat the tests on those images.

  6. Consider the family of triangles starting from an equilateral triangle, show below, for which the top point is allowed to rotate around the circle. Note there will be a degenerate case where two of the points of the triangle coincide. Construct a series of binary triangle images representing members of this family.

    1. Find an equation for the perimeter of the triangles in your series of images. Measure the perimeter length using bwperim. Compare the results to the theoretical values for the perimeter.
    2. Find an equation for the circularity (ratio of mean radial distance to the standard deviation of the radial distance) for the triangle images. Calculate the circularity using the matlab code given in class and compare the results to the theoretical values.

  7. Identify (count) and measure the diameter of the isolated dowels in the image dowels.tif from assignment 2. Generate an image with those dowels missing. Plot the signatures of the remaining objects.

  8. You are given three "mystery" aerial images, obviously of the same location, but with differing brightness variations. These images might be three spectral bands, or the same image processed by some algorithm. Act on the assumption that the images are three spectral bands, decide which image should be red, green, and blue and produce a composite 24-bit image.


Maintained by John Loomis, last updated 31 Jan 2005