EECS 332 - Digital Image Analysis
CATALOG DESCRIPTION: Introduction to computer and biological vision systems, image formation, edge detection, image segmentation, texture, representation and analysis of two-dimensional geometric structures, and representation and analysis of three-dimensional structures.
REQUIRED TEXT: None
REFERENCE TEXT: R. Jain, R. Kasturi, and B. G. Schunck, Machine Vision , McGraw-Hill, Inc. 1995.
READINGS : Papers from journals, conference proceedings, or book chapters will be assigned.
COURSE COORDINATOR: Prof. Ying Wu
COURSE GOALS: The goal of this course is to provide students with a basic understanding of the fundamentals and applications of digital image analysis (or computer vision) techniques including 2-D and 3-D paradigms to solve real world applications.
PREREQUISITES : None.
PREREQUISITES BY TOPIC :
- Linear algebra
- Computer programming in C
DETAILED COURSE TOPICS :
- Introduction to image formation (1 week)
- Binary image processing (2 weeks)
- Color and color segmentation (1 week)
- Region segmentation (1 week)
- Edge, contour, Hough transform and texture (2 weeks)
- Motion and tracking (1 week)
- 3D geometry, calibration, pose and stereo (1 week)
- Lighting and applications (1 week)
- Implementation of connect component analysis
- Implementation of morphological operators
- Implementation of histogram equalization and lighting compensation
- Implementation of color segmentation
- Implementation of canny edge detector
- Implementation of Hough transform.
- Implementation of camera calibration
- Implementation of 3D pose determination
Based on the machine problems, the course involves a final project to design a vision-based interface system, i.e., a “virtual gun,” where the cursor moves with your fingertips. The idea is to locate and track a fingertip through a video sequence accurately and robustly. The project consists of three parts: (1) a working demo, (2) a 15-minute presentation, and (3) a 15-page report.
- Machine problems – 50%
- Final project – 50%
COURSE OBJECTIVES: When a student completes this course, s/he should be able to:
- Understand the projection geometry in the image formation process.
- Design and implement computer programs to perform image feature extraction.
- Design and implement computer programs for image segmentation.
- Design and implement computer programs for motion analysis and tracking.
- Understand the basic techniques and issues in 3-D computer vision.
- Design and build a real vision-based interaction system.
ABET CONTENT CATEGORY: 100% Engineering (Design component).