Computer vision is the science of automatically computing information and making decisions from an observed image, image set or an image sequence. It combines concepts from 'image processing' (in the spatial and frequency domains) and 'pattern recognition'. Computer vision has a wide number of potential applications, including satellite imaging, control and measurement, industrial inspection, surveillance (e.g. face recognition) and medical applications. This unit covers topics such as binary image analysis, greyscale image manipulation, linear and nonlinear filtering, feature extraction, image enhancement, image segmentation and recognition. It also covers camera calibration and projective geometry and how three-dimensional information can be reconstructed from single images, stereo pairs of images and motion sequences. In the future, it is anticipated that computer vision systems will become prevailing, and that vision technology will be more applied across a broad range of business and consumer products. This will result in a strong industry demand for computer vision engineers—for people who understand vision technology and know how to apply it in real-world problems.

Students are able to (1) explain computer vision problems in writing; (2) write MATLAB code to solve computer vision problems; (3) describe the theories and principles in computer vision; (4) conduct independent research on a chosen research topic, write a small research report, and give an oral presentation; (5) demonstrate logical thinking and problem-solving skills; (6) process images in both the spatial and frequency domains; (7) explain the technical theory behind formation of images; and (8) critique various methodologies for solving problems in computer vision and image processing.