A potato farmer who wants to detect a disease in his crop as early as possible. The surgeon who wants more visibility into his operation. The processing company that wants to separate different plastics. These are all issues where AI can provide a solution.

• Do you want to get started with Artificial Intelligence?
• Do Big and Small Image Data have your interest?
• Would you like to develop technical solutions to real-world problems under the guidance of experts in the field?

Then the minor I'M A.I. - Artificial Intelligence for Image Data is of your interest! Specialize yourself in image acquisition, image processing and pattern recognizing using machine learning.

Minor content

In this minor you will learn the basics of image acquisition, image processing and pattern recognition using deep learning, among other tools. You learn by working on a real-world problem of one of our clients. This in a lab with state-of-the-art equipment such as a mini-supercomputer, 3D cameras, hyper- and multispectral cameras and JetBots.

You will get familiar with the following aspects: 

• Image acquisition and configuration of a vision system (camera, lens and illumination) 
• Image segmentation, image classification, object detection or object counting using deep learning and computer vision. 
• Validation and optimization of algorithmic performance on datasets. 
• Programming and using libraries and software tools. 
• Applied research methodology and writing a technical paper. 
• Project management using SCRUM. 
• Design Based Education in short prototyping cycles.

Structure of the minor

You study with us full-time for the length of a semester, you can start each semester (Sept or Feb). The semester starts with a kick-off course on Computer Vision & Data Science. During this course you will build on your basic knowledge and skills in the field. Topics that are covered include for example machine learning, neural networks and object detection. At the end of the second week projects are awarded, based on personal interests and technical background. For the remainder of the semester you will, in a group or individually, work on an real-life applied research projects from one of our partner companies. 

By working on the projects you deepen your knowledge and skills in the field, but also collect evidence for proving the learning outcomes. In monthly sprints you work towards an end product, consisting of a technical paper, poster, presentation and a demonstrator/proof of concept. During weekly SCRUM meetings the project progress is discussed and at the end of each sprint your educational progress. Your supervisor is one of our researchers. We are a small and flexible team and communication lines are short by working in two adjacent rooms. There is ample opportunity to ask questions.

This minor is part of the Mathematical Engineering curriculum and is offered by the Computer Vision & Data Science professorship.

Admission requirements

This minor is the place for students who want to learn more about image data analysis. This minor is therefore extremely suitable for students who have an affinity with programming, mathematics, applied research, computer vision and data science. This makes the minor a good match with various studies from the technical bachelor's education, such as (technical) computer science and mathematical engineering, but also chemical technology and electrical engineering

For admission we ask a motivation letter (in English), including CV. There will be an intake before admission to the minor.  Note, there is a maximum number of places available per semester. Admission is based on suitability and orderliness. It is therefore important to express your interest in time.  

Requirements are technical propaedeutic diploma and some engineering and programming experience. The student needs to have basic knowledge and experience with Python; Additional Computer Vision and  Data Science training is provided at the start of the minor. In this document, we sum up all the details of admission for you. 

This minor is offered by the Applied Mathematics degree program.

The recommended knowledge is:

• Demonstratable programming experience in Python;
• English language proficiency to communicate technical work adequately;
• Willing to dive into image data analysis and applied research.

To ensure the quality of the programme we also apply a maximum enrolment of students per semester. The first come - first serve principle is applied.

Indicators
The minor programme will look at these indicators in determining whether the recommended knowledge has been met.
Programming – The prospective student is able to demonstrate and explain code and is able to use:

• Debugging tools for Python.
• Libraries like NumPy.
• Package managers like Anaconda and pip.
• Operating systems Linux and Windows.

English – The prospective student is able to communicate technical work adequately in writing and verbally.

Motivation – The prospective student is motivated to:

• Learning more about image data analysis.
• Solving real-life research questions by programming.
• Doing a full-time minor programme of a semester.

Impression
The prospective student should be comfortable with Python programming of medium-sized Python scripts and especially has experience using PyCharm + Debugging. To give an impression of the required level of scientific programming skills we provide the following exercises.

For Python we advise to do this challenge https://github.com/Asabeneh/30-Days-Of-Python (at least the first 13 days, but also have a look at day 14) and for NumPy these exercises https://www.machinelearningplus.com/python/101-numpy-exercises-python/.

Within the minor we use PyCharm, so when practicing with Python we advise to use that IDE. Installing it from https://www.jetbrains.com/pycharm with an NHL Stenden University account gives a free download link to the professional edition.

Admission Procedure
During an intake interview, the minor programme determines whether the prospective minor student meets the required starting level expertise. The minor organizes a number of intake days per year, for which prospective students are invited in order of registration. All communication and documentation mentioned within this procedure is in English.
The basis for the intake is a motivation letter and CV that the prospective student submits in preparation for the intake interview. Within the motivation letter and the CV the prospective student provides evidence for meeting the admission requirements. In preparation for writing the motivation letter and the intake interview, prospective students are asked to study the exercises mentioned above.

First it is determined whether the student is admissible according to the quantitative requirements (maximum number of students per semester). If so, subsequently, the student is invited for an intake. During the intake interview the prospective student is asked to explain some of their own code to provide the admission board further insight on the level of programming experience. At the end of the intake, it is determined whether the student is admissible according to the qualitative requirements. The student will be either admitted or rejected.
The result of the intake will be communicated with the student within five working days following the intake interview.

Applying and contact information
To register for this minor, you can send your motivation letter and CV via the red 'online form' button below. You can also use this button for other questions.

 

The Learning Outcomes for the minor are:

1. The student selects, applies and tests, within a team and methodologically correct, machine-learning algorithms that automate visual inspections that meet the project’s specifications.
2. The student creates and manages, under supervision, a representative annotated and balanced dataset with the required quality to test machine-learning algorithms.
3. The student develops themself as a professional.

Within the minor programme knowledge and skills are integrated with applied research. Students explicitly work on the learning outcomes with the projects they complete in that semester. The portfolio created throughout the semester is used as evidence by the students to show they have achieved the learning outcomes. The semester is devided in five sprints of a month, in a way that the student iteratively works towards the end product and final portfolio.