CITS4402 Computer Vision

Week 5 Laboratory Exercises

Due: Check the deadline on cssubmit.

Finding Circles using the Hough Transform

Pupillometry is the study of the pupil diameter. It is used in psychology and medicine. A pupil can be well approximated with a circle. Moreover, its darker colour with respect to the iris gives a very strong edge response. Your task is to make a Matlab GUI that will load an image and detect any number of circles in it. Your GUI should look similar to the one shown below.

Your main aim is to detect the pupil however, your code should work for other circular objects in images as well e.g. as shown above. You will need to do edge detection first. Use a popup menu to choose an edge detection algorithm and a slidebar to adjust the thresold of the edge detector. Your edge image should automatically update as you move the slide bar. When you press the Detect Edges buttion, the edge detector should execute with default threshold i.e. no threshold given using the selected algorithm. There should be a minimum of two algorithm options. There should be another slide bar that varies the approximate radius for the Hough circle detection. As you move this slidebar, any circles detected in the image should appear on both images. 

Sample Images

Bonus Marks

There are bonus marks for anyone who additionally detects the iris circle and segments out the area between the iris and pupil into another image.

Submission Requirements

  • Your week5lab.m Matlab function file.
  • Your week5lab.fig Matlab gui figure file.
  • Your favourite images.

Your Matlab code should be commented (but not over-commented) and indented for readability (Hint: Use the Matlab Editor to help you indent your code).

School of Computer Science and Software Engineering

This Page

Updated by: Ajmal Mian