ENEL 697 DIGITAL IMAGE PROCESSING (WINTER 2008)

 

Raj Rangayyan, PhD, PEng, FIEEE,

FEIC, FAIMBE, FSPIE, FSIIM, FCMBES, “University Professor”

Professor, Department of Electrical and Computer Engineering, Adjunct Professor of Surgery and Radiology

University of Calgary

Room ICT 440, Phone 220-6745,   email: ranga@ucalgary.ca   Website:  http://www2.enel.ucalgary.ca/People/Ranga/

 

 

 

Lectures: Mondays and Wednesdays 4:00 to 5:15 PM. Room: ICT 446.

Laboratories: on your own schedule.

Classes start Monday, 14 January, 2008, and end Wednesday, 16 April, 2008.

No classes during Reading Week: 17 – 24 February, 2008.

 

Calendar description: Image formation and visual perceptual processing. Digital image representation. Two dimensional Fourier transform analysis. Image enhancement and restoration. Selected topics from: image reconstruction from projections; image segmentation and analysis; image coding for data compression and transmission; introduction to image understanding and computer vision. Case studies from current applications and research.

 

Detailed course content:

 

Note: Actual course content subject to variation, depending upon student interest and selection of topics from the following.

 

Introduction: Image acquisition and representation. Visual perceptual processing. Image quality and information content.

 

Two-dimensional systems and transforms. 2D convolution. 2D Fourier transform. Point spread function and system transfer function. Matrix and vector representation of images, linear system operations, and transforms.

 

Removal of artifacts: Random noise. Structured artifacts. Methods to remove artifacts.

 

Image enhancement: Contrast enhancement. Histogram operations. Spatial and frequency-domain filtering for smoothing and edge enhancement. Algebraic operations with images. Local, global, and adaptive methods.

 

Detection of regions of interest: Edge detection. Segmentation. Detection of objects of known geometry.

 

Analysis of shape: Moments. Fourier descriptors. Shape factors.

 

Analysis of texture: Ordered, oriented, and random texture. Statistical and structural analysis of texture.

 

Image reconstruction from projections: Projection geometry. Backprojection, Fourier, convolution backprojection, and algebraic reconstruction techniques. Fundamentals of computed tomography.

 

Image restoration: Degradation models. Inverse filter. Wiener filter. Deblurring.

 

Image coding: Information theory. Source-coding techniques. Decorrelation techniques. Huffman, run-length, predictive, interpolative, and transform coding.

 

Image analysis and computer vision: Feature representation and pattern classification.

 

Case studies from image processing research.

 

Recommended text:

 

R.M. Rangayyan, “Biomedical Image Analysis”, CRC Press, Boca Raton, FL, 2005.

 

Additional references:

 

R.C. Gonzalez and R.E. Woods, "Digital Image Processing", 2nd Ed., Prentice Hall, Upper Saddle River, NJ, 2002.

 

A. Rosenfeld and A.C. Kak, "Digital Picture Processing", 2nd Ed., vols. 1 and 2, Academic Press, New York, NY, 1982.

 

M.D. Levine, "Vision in Man and Machine", McGraw-Hill, New York, NY, 1985.

 

K.R. Castleman, "Digital Image Processing", Prentice-Hall, Englewood Cliffs, NJ, 1996.

 

A.K. Jain, "Fundamentals of Digital Image Processing", Prentice-Hall, Englewood Cliffs, NJ, 1989.

 

W.K. Pratt, "Digital Picture Processing", 2nd Ed., Wiley, New York, NY, 1991.

 

M. Sonka, V. Hlavac, and R. Boyle, “Image Processing, Analysis, and Machine Vision”, 2nd Ed., Brooks/ Cole, Pacific Grove, CA, 1999.

 

 

Evaluation and grading: There will be two tests (closed-book, closed-notes) during the term. Use of simple, non-programmable calculators with no text storage facilities will be permitted.

 

You are required to complete up to ten laboratory exercises to be assigned during the course. A report, including illustration (one page each) and discussion of the results obtained in each experiment (one page each), must be submitted at the end of the course.

 

In lieu of a final exam, you are required to work on a Digital Image Processing Project of your choice. Projects must involve the development of algorithms for digital image processing, computer programming for implementation of the algorithm, and testing of the methods with real images from any application area of your choice (such as medical imaging, remote sensing, robotics, or geophysical exploration). The algorithms need not be original, but must be technically advanced and sophisticated. If the project is a continuation or extension of previous work, you should state clearly your additional work and findings in the course project. The project must be completed before the end of the course.

 

A full-fledged written project report and a seminar must be presented at the end of the course. The project report must include a brief introductory review of the subject area and problem, complete technical details of the methods (equations, procedures, and algorithms) developed, critical analysis and discussion of the results obtained, and references. More attention should be paid to the image processing techniques developed than to the specific application of interest in the project. The recommended length of the report is 15 double-space-printed pages, excluding illustrations and references.

 

 

Grading details and deadlines:

 

One-page project proposal: Monday, 25 February, 2008.

 

Two tests (1.5 hours and 20 marks each): TBA.

 

Lab report (10 marks): Monday, 14 April, 2008.

 

Project report (50 marks): Wednesday, 16 April, 2008.

 

Seminars: 17 – 18 April, 2008. Exact time to be fixed later.

 

All of the items listed above must be completed satisfactorily in order to obtain a passing grade in the course.

 

Pre-requisites: ENEL 327 Signals and Transforms OR: A working knowledge of linear algebra (vectors and matrices), advanced calculus (complex variables, the Fourier transform), probability and statistics, computer programming, linear system theory, and digital signal processing.