In this course, you will be introduced to advanced image analysis methods enabling to further enrich your geoinformation problem solving abilities. Image processing and analysis methods treated in previous courses, such as conventional hard pixel based classification do not take into account spatial correlations in images and therefore do not exploit information contained in images to full extent. In addition, such methods cannot correctly treat mixed pixels, uncertain class definitions and data from various sources, making them insufficient for reliable geoinformation extraction. In this course we aim to treat more specialized image analysis methods. In particular, Support Vector Machines and Random Forest will be introduced for multisource classification at pixel level whereas Convolutional Neural Networks and Markov random fields will be introduced for contextual classification. These methods will be applied to analysis of satellite images at a pixel and some methods at a sub-pixel levels. The methods introduced in this course will be applied on real case studies.
- Support Vector Machines for classification
- Random Forest for classification
- Deep Learning with Convolutional Neural Networks applied for classification
- Markov Random Fields for imageclassification and filtering
Image analysis requires theoretical concepts and practical skills. Lectures will be used to briefly introduce the subjects, followed by reading textbook material and flipped classroom discussions. In addition, on some other subjects reading research articles will be recommended after the lecture to go deeper into the subject.
Practical classes will consist of a mixture of a demo by an instructor, individual work following written instructions and summarizing the outcome of the exercise in a class. In practical class students are supposed to work with existing program codes and modify these (to a limited degree). In this way the students can get insight in the intermediate stages of the image analysis algorithms and make decisions on the outcomes. In the summaries reflection on theoretical concepts will be made. In this way a solid integration of theory and practice will be achieved.
Upon successful completion of this course, you will receive a Certificate which will include the name of the course.
Along with your Certificate you will receive a Course Record providing the name, and if applicable, all the subjects studied as part of the course. It states: the course code,
subject, EC credits, exam date, location and the mark awarded.
This certificate course is part of the accredited Master’s programme Geo-information Science and earth Observation at ITC. If you decide to take the full Master’s programme Geo-information Science and Earth Observation programme at ITC, the Examination Board will give you in principle exemption from the course you followed successfully as a certificate course.
Academic level and background
Applicants for the Certificate programme should have a Bachelor degree or equivalent from a recognized university in a discipline related to the course, preferably combined with working experience in a relevant field.
Some courses in the Certificate programme or separate modules require knowledge of, and skills in, working with GIS and/or digital image processing of remotely sensed data.
Skills in taught or related subjects are a prerequisite for some courses in the Certificate programme or separate modules. Even if the applicant satisfies the overall admission requirements, acceptance is not automatic.
The faculty accepts transcripts, degrees and diplomas in the following languages: Dutch, English, German, French and Spanish. It is at the discretion of the faculty to require additional English translations of all documents in other languages as well.
As all courses are given in English, proficiency in the English language is a prerequisite.
If you are a national of one of the countries in this list (PDF), you are exempted from an English language test.
Please note: the requirements when applying for fellowships may vary according to the regulations of the fellowship provider.
English language tests: minimum requirements
Only internationally recognised test results are accepted.
TOEFL Paper-based Test (PBT)
TOEFL Internet-based Test
British Council / IELTS
C2 Proficiency / C1 Advanced
Applicants lacking computer experience are strongly advised to follow basic courses in their home countries.