CIS 350 – Fall 2003 (Undergrad. Seminar)

Principles and Applications of Computer Vision

Instructor: Dr. Rolf Lakaemper, 534 Wachman Hall, lakamper@temple.edu

Objective:

This seminar will give a comprehensive overview of the principles and applications of computer vision. Although there is often a deep mathematical background behind these techniques, the principles are mostly understandable by looking at the algorithmic solutions for special applications.

This seminar will focus on algorithms and applications (not on mathematics), forcing the students to become acquainted with many current techniques in computer vision by implementation.

The course will introduce basic techniques but will also present high level algorithms behind modern industrial applications. An introduction to MATLAB will be given.

Outline:

• Introduction to basic concepts in Computer Vision and Digital Image Processing
• Low Level Techniques
• linear filters: edge detection and noise reduction, application of math. morphology, detecting binary objects, feature detection in different spaces, histograms
• Mid Level Techniques:
• image and video compression: jpeg, gif, mpeg and other formats, image segmentation,basic movement detection
• High Level Algorithms:
• examples of industrial robots using computer vision, object tracking in video data, face detection and recognition, object recognition using global features, recognition of 2D shapes in images, building an autonome robot recognizing traffic signs

(not all topics might be listed here, some topics from the list might be changed)

PREREQUISITES:

by permission of instructor OR:

CIS166 and CIS223 and Math 76 or 86

Good programming skills in JAVA or C++ are mandatory, since the seminar will focus on the implementation of non-trivial algorithms !

Textbook:

(to be determined)

Grading:

Grades will be based on seminar presentations, projects and an exam.