• QUT logo

Robotic Vision: Processing Images

How do computers process images? Learn about image processing and image features, and how robots can be programmed to see.

4,276 enrolled on this course

Robotic Vision: Processing Images
  • Duration4 weeks
  • Weekly study3 hours

Learn how images are processed and transformed to make a robot see.

Programming a robot to see requires knowing the principles of vision, mathematical knowledge and programming skills. We explore how computers process images, learning the operations required to process an image using MATLAB.

Robotic vision relies on identifying and extracting the features of an image to recognize a shape. You’ll identify the unique features of a shape and its perimeter. We look at some advanced image processing techniques: morphology; and transformations such as scaling and warping. We finish the course with a look at colour and light, and how robots see colour.

What topics will you cover?

  • How computers process images
  • Pixels, edges and regions
  • How images are processed – histograms, monadic operations such as gamma correction and thresholding, and diadic operations
  • Spatial operators such as kernels, Gaussian smoothing and edge detection
  • How to extract image features – binary blobs and blob hierarchy
  • Advanced image processing – morphology, scaling and image warping
  • How to find dominant lines
  • Properties of colour and light

When would you like to start?

  • Date to be announced

Add to Wishlist to be emailed when new dates are announced

What will you achieve?

By the end of the course, you‘ll be able to...

  • Describe how images are represented in a computer
  • Describe the purposes of different types of image processing and feature extraction operations
  • Demonstrate basic coding in MATLAB for image processing relevant to robotics

Who is the course for?

This course assumes that you are familiar with concepts from advanced high-school mathematics or undergraduate engineering.

The course also assumes knowledge of programming – this will help you understand some of the demonstrations in the videos and is required for the third course in the robotic vision program. This course uses the MATLAB programming language and environment, but your knowledge of programming in other languages can be easily transferred to MATLAB. You can familiarize yourself with MATLAB by enrolling in the MATLAB Onramp tutorial.

Please note that this course includes video content and other visual teaching methods. Blind and visually impaired students may need a helper.

What software or tools do you need?

This course contains many practice activities in MATLAB. However, you won’t need to download the MATLAB software to complete the course (although if you already have it you are welcome to use it). The MATLAB exercises are embedded within the course, or you can use a version of MATLAB Online for which you’ll be provided a log-in and instructions for importing the Machine Vision Toolbox for MATLAB. You don’t need to own a robot, or have a robot kit, to take this course.

Who will you learn with?

Professor of Robotic Vision at QUT and Director of the Australian Centre for Robotic Vision (ACRV). Peter is also an IEEE fellow and on the editorial board of several robotics research journals.

PhD Candidate with the Australian Centre for Robotic Vision researching towards robust visual object recognition to facilitate useful robotic tasks.

Dr Pepperell completed his PhD in robotic vision at QUT in 2016, with a research focus in vision-based place recognition.

Who developed the course?

Queensland University of Technology

QUT is a leading Australian university ranked in the top 1% of universities worldwide by the 2019 Times Higher Education World University Rankings. Located in Brisbane, it attracts over 50,000 students.

  • Established1989
  • LocationBrisbane, Australia
  • World rankingTop 180Source: Times Higher Education World University Rankings 2019