Theory of Image Processing (P00809)

? Credit Points : 10 ? SCQF Level : 11 ? Acronym : GEO-P-RSTIP

A theory based course on Image Processing techniques concentrating on the mathematical and physical models underlying the processing operations. Digital image representation and sampling aspects are followed by various processing techniques including point by point operations, noise models, filtering and de-convolution techniques, edge and line detection, stereo imaging, target tracking and elementary pattern recognition.

Entry Requirements

? Pre-requisites : Available only to postgraduate students Maths at degree or equivalent level for non-RSIP Normally taken with Fundamentals of Remote Sensing for GIS students

Subject Areas

Home subject area

Postgraduate Courses (School of GeoSciences), (School of GeoSciences, Schedule N)

Delivery Information

? Normal year taken : Postgraduate

? Delivery Period : Semester 1 (Blocks 1-2)

? Contact Teaching Time : 2 hour(s) per week for 10 weeks

First Class Information

Date	Start	End	Room	Area	Additional Information
18/09/2007	12:10	13:00	Room 301, Crew Building	KB

All of the following classes

Type	Day	Start	End	Area
Lecture	Tuesday	12:10	13:00	KB
Lecture	Friday	12:10	13:00	KB

Summary of Intended Learning Outcomes

The course will concentrate on the underlying physics and mathematics of processing techniquies, we aim to cover in the common material:

1: linear image formation and its underlying assumptions,

2: digital representation of image, the discrete Fourier Transform, its properties and implementation, Shannon sampling theorem, interpolation to zeroth and first order,

3: first order image statistics, point-by-point processing and histogram manipulation,

4: fixed pattern noise and random noise including underlying physics of Gaussian additive noise and methods of it estimation,

5: linear filtering in real and Fourier space, non-linear filters including shrink and expand, average threshold and median,

6: image restoration by inverse and Weiner filter, outline of CLEAN and maximum entropy restoration,

7: tomographic system and reconstruction by Fourier inversion and filtered back projection, outline of fan-beam system.

8: edge and line detection by first and second order differential edge detection, Hough Transform and its applications,

9: stereo imaging in parallel and converging geometry, outline of automated depth extraction techniques,

10: tracking by correlation, basic of statistical pattern recognition, examples of simple classifiers.