Course Catalogue

View Specification Document | Reading List

This course introduces fundamental concepts in digital image processing. It recapitulates the mechanics of the human visual system and typical general-purpose image processing system. It discusses several "classical" image processing techniques for intensity transformation and spatial filtering. It helps students develop a solid understanding of how the transform domain can be used for image filtering without becoming a signal processing expert. It also introduces techniques used in image enhancement/restoration/compression/segmentation, object detection and recognition. This course contains laboratory work or open project designs in which participants work in small groups solving real-life problems.

Course will be delivered continuously, 4 hours per week over 12 weeks.

Mandatory Entry Requirements

Knowledge of linear algebra and probability basics

Recommended Entry Requirements

Knowledge of high-level programming language such as C / Python / Matlab

None

None

Final exam (2 hours), 40%

Projects, 60%

This course aims to provide an overview of fundamental theories, concepts, methods, and digital image processing techniques. It will help students build their image processing toolbox with some essential methods such as vector quantization, histogram equalization, different image transforms, noise reduction, edge detection, morphological operations. It also helps students develop the ability to solve real-life problems.

By the end of this course students will be able to:

• Describe how visual information is processed before and after it reaches the end-user

• Evaluate and apply "classical" image processing techniques for image enhancing, restoration, segmentation or compression in both spatial and transform domains

• Design appropriate image processing algorithm to solve practical image processing or computer vision problems

• Work in teams and communicate effectively on complex engineering matters