Title: ISA: Image Synthesis and Analysis
Short Code: MTE7901
ECTS Credits: 2
UAS: BFH
Organizer Details: BFH HuCE
Evaluation:
Final exam 2 weeks after the last course session. The lecture will provide additional information at the beginning of the course.
Decision Date: 1 September 2023 
Start Date: 22 September 2023 
End Date: 3 November 2023 
Date Details:

Date KW SW HS23 morning (4L) HS23 afternoon (4L)
22.09.23 38 1 ISA ISA
29.09.23 39 2 ISA ISA
06.10.23 40 3 ISA ISA
13.10.23 41 4 ISA ISA
20.10.23 42 5 Self-study
27.10.23 43 6 Self-study
03.11.23 44 7 Self-study Exam ISA
Type:

Full day course at 6 Fridays per semester + 1.5 days individual preparation for the exam.

Language(s):

English by default, but deviations according to the wishes of the students.

Description (max. 300 characters):
This course is part of the HuCE EVA course series. The course topics are Image Synthesis and Image Analysis.
Contents and Learning Objectives:

Image Synthesis (Prof. Marcus Hudritsch)

This course introduces modern GPU programming using the OpenGL GLSL Shading Language. We thereby get to know the basic principles of computer graphics. In particular, we look at the working of the graphics rendering pipeline and its processing steps like vertex and geometry shader, hidden surface removal, rasterization, fragment processing, and depth buffering. Several demo applications and small programming exercises will give the participants hands-on knowledge of tools and the GLSL language. The course evaluation will be based on a small, student-defined defined, and implemented project and its short presentation.

Image Analysis (Prof. Marcus Hudritsch)

This course block gives a broader insight into digital image analysis. After a short wrap-up over image processing, we start with the classic feature engineering approach where the software engineer first segments an image into desired regions and represents them in processable data structures. After that, we learn different methods to extract meaningful and discriminative features that we can use in the final step to classify the image. The course block closes with machine learning for image analysis where we learn how to use unsupervised and supervised learning with large labeled datasets to classify images.

All material (Scripts, slides, exercises) is distributed with a GitLab repository.

Admission:
Literature:
Conditions:

50% theory and 50% labs

Contact:
  • For administrative questions and module enrolment, please contact the BFH MSE office directly. mse@bfh.ch
  • Students who are not enrolled at BFH must first create a guest account and then send an email to the BFH MSE office. Link to registration form!
  • Lecturer: Prof. Marcus Hudritsch
 
Contact Person E-Mail: marcus.hudritsch@bfh.ch
Status: finished
 
Specialization: Industrial Technologies (InT)

Computer Science (CS)

Data Science (DS)

Electrical Engineering (ElE)

Mechatronics & Automation (MA)

Medical Engineering (Med)

 

[Responsible for this text: Hudritsch Marcus]