On this page, you should find all the information about exercises of the Image Analysis course.

Timetable

Exercises

Thursday
10:45 - 12:15, EB405
12:30 - 14:00, EB405

Evaluation - how can you earn a credit

During the exercises, we will cover some topic of digital image processing. Your task is complete and handover them for final grading. We'll use the OpenCV library in our implementation.

Other sources

A simple tutorial describing how to program with the OpenCV is provided in the following link: Introduction to programming with OpenCV

Linux Image for the Project

I've created a VirtualBox Linux Image (1.4 GB) that you can use to develop our project. The only thing you have to do is to install VirtualBox on your system and install VirtualBox Tools inside your image to be able to use a better display resolution.

Login creditials are as following:
username: lab
password: lab

Exercise 1

Color image of a valve (already available in the provided project) that we can use for our experiments.

You can use a project for the CodeBlocks in the Linux environment to code the exercise or you can also use a project for the Visual Studio 2015 in the Windows environment to code the exercise.

We'll implement x and y derivatives of image and visualize edge magnitudes of the valve image:

Exercise 2

We'll implement a Laplace operator of the valve image:

A colored version of the previous image:

Exercise 3

Edge simplification & double thresholding on the valve image:

Exercise 4

We'll implement thresholding of objects.

We'll use the image below for our experiments:
Train Image

Text describing the content of the exercise: features_01.

Exercise 5

We'll implement object extraction, indexing, and simple features computation.

Text describing the content of the exercise: features_01.

Exercise 6

We'll implement features computation: features_02

Exercise 7

We'll implement saving the ethalons features and use this features to classify objects in a new image. A description is provided in this document.

Exercise 8

We'll implemente k-means algorithm for simple clustering of extracted features.

Exercise 9

Train image Test image

Exercise 10

We'll use a back propagation neural network as a classifier. See OpenCV docs for neural network API.

Exercise 11

We'll look at the motion detection based on background modeling usign gaussion mixture model. The text that describes the method is available here: mog.pdf (in Czech, author: Tomáš Fabián)

A video that we can use for our experiments is also available: dt_passat.mpg