ELE489: DIGITAL IMAGE PROCESSING

Department of Electrical Engineering
University of Southern Maine

Instructor: Professor Mariusz Jankowski
Office: 127 John Mitchell Center
Telephone: (207) 780-5580
E-mail: mjankowski@usm.maine.edu


Course Description: The theory and practice of digital processing of images by computer. Introduction to two-dimensional signal processing theory. Image acquisition and representation, enhancement methods, image coding, image analysis, and image processing hardware.

Course Objectives:
Introduce fundamental principles and techniques for digital image processing and to provide hands-on experience in using software for processing digital images.

Course Goals:
1. Understand the effect of sampling and quantization on image quality.
2. Obtain a variety of statistical measures of an image.
3. Convert between standard color spaces.
4. Perform halftoning and error diffusion operations.
5. Implement 2-D finite impulse response (FIR) image filtering operations.
6. Implement 2-D binary morphological filtering operations.
7. Implement 2-D non-linear noise reduction filters such as the median filter.
8. Design point operations for image enhancement, including histogram equalization.
9. Compute 2-D image transforms.
10. Understand and implement block transform coding techniques for image compression.
11. Derive and implement the linear Wiener restoration filter, as well as its adaptive extension.

Laboratories: This course is part of the "Integrating Mathematica into the electrical engineering curriculum" project. Included are a series of Mathematica notebooks. Students will be expected to use Mathematica to solve problems in class, as well as in take-home assignments and projects.

Prerequisites: ELE314, COS160 or equivalent. Lecture 3 hrs. 3 Cr.

Textbook:  none

Suggested reading:

"Digital Image Processing," by R. Gonzalez and R. Woods, Prentice Hall, 2nd Edition, 2001.
"Digital Image Processing Algorithms and Applications," I. Pitas, Wiley, 2000.
"Fundamentals of Electronic Image Processing," A. Weeks, SPIE Press/IEEE Press, 1996.
 

Lecture Topic Reading
1-2 Digital image fundamentals
- digital representation of images (monochrome and color)
- elements of matrix theory
- digital imaging hardware and software
- sampling, quantization		 1, 2.1-4
3-6 Image enhancement in the spatial domain
- point processing
- image histogram
- spatial filtering
 		3.1-8
7-8 Color image processing
- color models
- color transformations 		6.1-5
9-11 Image geometry and spatial transformations
- interpolation and decimation
- spatial transformations (rotation, warping) 		5.11
12-15 Image transforms
- frequency domain representation of LSI systems
- 2-D digital filters
- discrete Fourier transform
- other unitary transforms (cosine, wavelet) 		4.1-4, 4.6
16-18 Image restoration
- noise removal
- inverse filtering		 5.1-10
19-21 Image compression and coding
- predictive coding
- transform coding 		8.1-6
22-24 Image morphology
- binary morphology
- grayscale morphology		 9.1-6
25-28 Image segmentation
- thresholding
- edge based
- region based 		10.1-6


Grading policy: The final grade will be based on the results of two mid-semester exams, a final exam, and homework problems/computer projects. These will be weighed equally, each part contributing approximately a third of your final grade.

 

If you need course adaptations or accommodations because of a disability please contact the Office of Academic Support for Students with Disabilities, 237 Luther Bonney Hall, 780-4706.