|
Search our Site
ON THIS PAGE:
Course Objectives
Prerequisites
Approach
Typical Text
Course Outline
|
|
COMP 775 [254]: Image Processing
and Analysis (BMME 775)
(3 hours)
Course
Objectives
Introduces statistical and structural pattern recognition as applied to
the analysis of digital images. Prerequisites
Prerequisites: MATH 547, STOR 435 and ability to program in Matlab.
Approach
Lecture with mathematical homework and programming lab assignments.
Typical Text
Coggins, J. Image Pattern Recognition (notes)
Course Outline
Numbers in parentheses indicate approximate number of weeks
- Geometry (2)
Scale, derivatives. Filter implementations. Significant geometric
structures: elliptical and hyperbolic regions, parabolic lines,
gradient and isophote, watersheds, ridges.
- Statistical Pattern Recognition (5)
Bayes' Theorem, Maximum Likelihood (ML) and Maximum
A Posteriori (MAP) estimation; Classification; Density estimation;
Evaluating feature sets; eigenanalysis; Applications to images.
- Optimization (2)
Problem characterization, Energy function definitions, Gradient descent,
simulated annealing, genetic search, snakes, conformable surfaces.
- Advanced image operations (2)
Warping via thin plate spline, morphological operators, masking
operations
- Structural Pattern Recognition (3)
Identifying primitives; search; matching methods
|
