Multisensory systems, machine vision

Lecturer: Podržaj Primož

Syllabus outline:

• Sensor overview
• Sensor fusion, its challenges and advantages
• Digital image acquisition
• Basic point and neighbourhood processing
• Image processing software overview
• Most common image processing applications

Objectives and competences:

The course is divided into two parts. In the multisensory system part the students will first get an overview of various sensors and their capabilities. Then the benefits of sensor fusion will be discussed. As a result, the students will be able to couple various sensors and extract optimal performance of such a combination. The second part of the course is focused on machine vision. In this part, the students will get a basic understanding of a digital image and its acquisition. Image processing will then be discussed from a mathematical point of view. Consequently, the students will get the capability of designing algorithms for various machine vision tasks. After an overview of image processing software will be given, and some most common applications presented, the students will start working on a project. As a result, they will get the capability of designing a real-life machine vision application and also be able to assess all the potential risk involved in such a project. This will make the competent to execute such projects in future without too much difficulties.

Intended learning outcomes:

• to get an overview of existing sensors, their capabilites, advantages and weaknesses
• to get an understanding for the benefits of sensor fusion
• to get the basic understanding of digital image acquisition
• to develop necessary skilly for successful and efficient image processing application development
• to get an overview of possible image processing packages in various programming languages
• to assess the time needed for accomplishing the above-mentioned task and execute it in a real-life project