| Responsible Department | Department of Basic Science and Environment | ||||||||||||||
| Course Dates | Block 4 (late April through June), even years (2012, 2014, ...). | ||||||||||||||
| Course Abstract | The couse is an introduction to automatic image analysis, with examples and applications mainly from LIFE research groups. Participants will learn principles and techniques for extracting qualitative and quantitative information from digital images. | ||||||||||||||
| Course Home Page | http://matdat.life.ku.dk/ia | ||||||||||||||
| Course Registration | To sign up for the course, please send an e-mail to Morten Larsen at ml@life.ku.dk. Please also remember to add the course to your PhD plan. | ||||||||||||||
| Deadline for Registration | April 1st 2012, 2014, ... | ||||||||||||||
| Credits | 7.5 (ECTS) | ||||||||||||||
| Level of Course | PhD course | ||||||||||||||
| Organisation of Teaching | The course will take place on the Frederiksberg campus. | ||||||||||||||
| Language of Instruction | English | ||||||||||||||
| Course Content | |||||||||||||||
| The following subjects will be covered: Digital image representation and -compression, filters, image improvement (low level image processing), simple characterisation of images (e.g. histograms), edge detection, texture, segmentation and counting of objects, morphological transformations, characterisation of objects in images, transformation to the frequency domain (Fourier transformation), analysis in the frequency domain, colour spaces and multispectral imagery, template matching, 2D image registration. Examples will be partly from research groups at KU/LIFE using image analysis; these groups are consulted during the development of the course. An open source / free standard tool (ImageJ) will be used in the course, and apart from acquiring basic skills using the tool the participants will also have to learn to make simple extensions and adaptations (through simple programming). In the last part of the course the participants will be working on projects, preferably with image data from their own PhD work. A project could for example focus on making an application which in a user friendly fashion solves a specific image analysis problem with a "known" algorithm (e.g. counting cells in a microscopy image) or a project could be more "experimental" in nature and seek algorithms to extract information from a certain type of images (e.g. segment different kinds of tissue in ultrasound images). At the end an oral presentation of each project must be made for the other participants. | |||||||||||||||
| Teaching and learning Methods | |||||||||||||||
| In the first part of the course, days will have a mixture of lectures and hands-on computer exercises. In the second part of the course, during which the participants are working on projects, there will be scheduled time for consultation on the projects. | |||||||||||||||
| Learning Outcome | |||||||||||||||
| The objective of the course is to introduce the participants to automatic image analysis and digital images, mainly with biological applications. The participants shall learn about fundamental problems in image analysis and methods to solve them. They shall furthermore learn to use and through simple programming extend and adapt a standard program for image analysis. After completing the course the student should be able to: Knowledge: -Describe the effect/behaviour of key image analysis methods, including the meaning of their parameters. -Explain on a general level the theory behind key image analysis methods. -Reflect about the appropriateness of an algorithm proposed for a given image analysis problem. -Display knowledge of basic principles of programming on the script level. Skills: -Select and apply low level image processing methods to achieve a desired effect. -Implement a simple image analysis algorithm through scripting of a standard software tool. Competences: -Design and develop an image analysis algorithm to solve an image analysis problem through the combination of standard image analysis tools. -Communicate to peers the principles of an image analysis algorithm developed. -Communicate to end users the scope, use and limitations of an image analysis solution developed. | |||||||||||||||
| Course Coordinator | |||||||||||||||
| Morten Larsen, ml@life.ku.dk, Department of Basic Sciences and Environment/Mathematics & Computer Science, Phone: 353-32390 | |||||||||||||||
| Type of Evaluation | |||||||||||||||
| A written project report and oral presentation (for all course participants) of the project. | |||||||||||||||
| Work Load | |||||||||||||||
| |||||||||||||||
| Other Remarks | |||||||||||||||
| The course also serves as a master course (280004). | |||||||||||||||