Course Project Image Processing Pattern Recognition

---

Responsible: Prof. Dr. rer. nat. Dietmar Kunz

Course

Meets requirements of following modules(MID)

• in active programs
  • Ba MT2012 BV3

Course Organization

Version created 2011-11_09 VID 1 valid from WS 2012/13 valid to

Course identifiers Long name Project Image Processing Pattern Recognition CID F07_BV3 CEID (exam identifier)

Contact hours per week (SWS) Lecture Exercise (unsplit) Exercise (split) Lab Project 2 Seminar Tutorial(voluntary)

Total contact hours Lecture Exercise (unsplit) Exercise (split) Lab Project 30 Seminar Tutorial (voluntary)

Max. capacity Exercise (unsplit) Exercise (split) Lab Project 18 Seminar

Total effort (hours): 180

Instruction language

• German

Study Level

• Undergraduate

Prerequisites

• Module Image Processing
• Module Pattern Recognition

Textbooks, Recommended Reading

• Burger/Burge: Digitale Bildverarbeitung
• Gonzales/Woods: Digital Image Processing

Instructors

• Prof. Dr. rer. nat. Dietmar Kunz
• Prof. Dr. Lothar Thieling

Supporting Scientific Staff

• tba

Transcipt Entry

Project Image Processing

Assessment

Type
oR	oral presentation of project results

Total effort [hours]
oR	10

Frequency: 1/year

---

Course components

Project

Objectives

Contents

• problem specific methods resulting from system model and literature search

Acquired Skills

• skilled use of software development environment
• skilled use of tools for image processing and image analysis
• skilled use of tools for training neural networks
• understanding of scientific texts in English
• presentation of project results in English

Operational Competences

• accomplish complex tasks in teams
• present project results
• Derive complex problem solutions that can be implemented using image processing and image analysis
  • analyse and understand complex problems
    • derive system behaviour from specifying texts
    • analyse systems
  • model system from subsystems
  • model, implement, and test subsystems
    • map subsystems as far as possible on available components (image processing modules), i.e. selection of models and parameters
    • implement and test required but not available image processing modules in C or Java using software development environment
  • implement, test, and validate entire system (problem solution)
    • Derive problem solution as chain of algorithms using image processing modules
    • parametrize image processing modules
    • test and validate solution
    • iteratively improve algorithmic chain

Additional Component Assessment

Type
fTP	presentations in regular team meetings, presentation and review of milestone results
wR	written report, software or application

Contribution to course grade
fTP	milestone presentation: rated (0 .. 10% of final marks)
wR	written report, software or application, rated (>50 % of final marks)

Frequency: 1/year