EENG445a/BENG445a/ENAS912a

Digital Image Processing

Home Page

Fall 2002

For information from your TA, see: the TA page.

For information on course resources for computer assignments, including matlab and account setup, see:
the Resources page.

Course Number	EENG/BENG 445, ENAS 912
Course Title	Digital Image Processing
Instructors	James Duncan (james.duncan at yale.edu) Lawrence Staib (lawrence.staib at yale.edu)
Teaching Assistant	Jing Yang (j.yang at yale.edu)
Schedule	TTh 9.00-10.15 at 107 Mason

Overview

This course is an introduction to Digital Image Processing covering digital techniques for image representation, enhancement, compression and restoration. Students will learn the fundamentals behind image processing methods and algorithms. This course is open to undergraduate and graduate students. We assume students have an understanding of linear systems and calculus. In addition, it is also helpful to have a familiarity with elementary probability theory and linear algebra. There will be about ten homeworks and both a midterm and a final exam (during exam period). Grading will be based approximately 1/3 on the homeworks, 1/3 on the midterm and 1/3 on the final. Undergraduates and graduates are graded separately; in addition, assignments may differ.

Text:

A. Rosenfeld and A. Kak, Digital Image Processing, Volume 1, Academic Press, 1982.
R. Gonzalez and R. Woods, Digital Image Processing, Addison and Wesley, 1993

Both on reserve in the Engineering Library.

Additional readings to be distributed during class.

Course Objectives:

Having successfully taken this course, you will be able to

understand and apply stochastic image representations and linear systems theory for images
apply image sampling theory
understand image quantization
implement and analyze image enhancement techniques
analyze techniques in image compression and coding, including current standards (e.g. JPEG)
implement, evaluate and analyze image restoration methods
understand and mathematically analyze algorithmic methods of image reconstruction from projections
understand other current representations and methods in image processing such as wavelets, markov random fields, and warping

Course Outline (dates/topics approximate, Fall 2002)


   Sep  5    Intro/Organization

             Mathematical Background (Read R & K Ch. 1, 2 and 3) 
   Sep 10    Linear Systems and Convolution 
   Sep 12    Linear Systems; 2D Transforms 
   Sep 17    2D Continuous Transforms 
   Sep ??    Rescheduled lecture: 2D Discrete Transforms 
   Sep 19    No class
   Sep 24    No class
   Sep 26    Stochastic Representations 
   Oct  1    2D Stochastic Representations   

             Representation and Enhancement (Read R & K Ch. 4 and 6) 
   Oct  3    Sampling 
   Oct  8    Sampling 
   Oct 10    Quantization; Image Enhancement 
   Oct 15    Image Enhancement 
   Oct 17    Image Enhancement   

             Image Compression and Coding (Read R & K Ch. 5) 
   Oct 22    Transform Compression                                  
   Oct 24    No class
   Oct 29    Transform Compression                                 
   Oct 31    Predictive Compression
   Oct ??    Rescheduled lecture: Error-free Compression

   Nov  5    Midterm Exam, 8:45am    

             Image Restoration (Read R & K Ch. 7) 
   Nov  7    Image Restoration: Inverse and Wiener Filtering        
   Nov 12    A priori methods: Gerchberg; CIR                       
   Nov 14    Discrete Formulation   
   Nov 19    Discrete Formulation; Geometric Distortion 
   Nov 21    Advanced techniques 
   Nov 26    Thanksgiving Break                           
   Nov 28    Thanksgiving Break                                      

             Reconstruction from Projections (Read R & K Ch. 8) 
   Dec  3    Projection; Fourier slice theorem     
   Dec  5    Reconstruction            
             Reading Period                                         

   Dec 18    Final Exam, 9:00 am BCT C031

26 August 2002