Machine Vision Methods and Applications (4cr)

Evaluation scale

1-5

Objective

After completing the course the student will know the essential machine vision methods and understand their common applications. Application examples are shown in the fields of technology, medicine and biology, to name a few. The student is able to implement image processing and machine vision applications by using OpenCV library and Python programming language.

Content

- Image formation and the structure of the digital image
- Preprocessing algorithms
- Segmentation algorithms
- Morphology algorithms
- Shape and feature detection and identification
- Image transformations

Location and time

[The schedules can be found in the timetable at https://lukkarikone.seamk.fi/. Timetables are published for the next six weeks. The first six weeks of autumn are published by Midsummer and the first six weeks of spring by Christmas. Timetables may be subject to changes.

Materials

Will be given during the course.

Teaching methods

The study involves contact teaching and requires attendance at the SEAMK campus.Hybrid studying is also possible in some extent.

The course is conducted in the Moodle learning environment. The course requires independent work and scheduling.

- The course includes laboratories at the SEAMK campus.
- The recordings of the teaching sessions are available for later viewing in Moodle.
- The course also includes independent work online.

Completion alternatives

This implementation can be completed in the following alternative ways. The completion methods are described in more detail in the Moodle learning environment. 

- Demonstration

Student workload

The workload of the study is designed so that one credit corresponds to an average of 27 hours of student work to achieve the learning objectives. The actual time required varies individually, e.g., due to prior knowledge.

Rough estimate of the workload:

Contact lessons: 35 t
Laboratory works: 30 h
Independent studying: 43 h

Assessment criteria, approved/failed

- At least 30 % of weekly exercises done
- Passed laboratory works
- Passed exam

The weekly exercises earn you extra points to a passed exam.

Assessment criteria, satisfactory (1)

The student can programmatically read a digital image and perform simple pre-processing and post-processing operations on it, as well as perform simple segmentation. The student knows the basics of the structure of a digital image.

Assessment criteria, good (3)

The student can also analyze a segmented image using different methods and perform measurements from it. The student knows several different pre-processing, post-processing and analysis methods and knows how to apply them. The student has extensive knowledge of machine vision applications and can describe them. The student's programming style is clear.

Assessment criteria, excellent (5)

The student has a strong command of the methods studied in the course and knows how to apply them widely for their own purposes and beyond the applications taught in the course. In addition, the student implements their method very clearly, following a good programming style.

Qualifications

Basics of Programming 1