DD2427 Image Based Recognition and Classification 6.0 credits
Bildbaserad igenkänning och klassificering
A course in computer science focusing on basic theory, models, and methods for classification of data with special emphasis on object recognition in digital images.
Educational level
Second cycleAcademic level (A-D)
DSubject area
Grade scale
A, B, C, D, E, FX, F
Course offerings
Spring 13 TMAIM for programme students
Periods
Spring 13 P4 (6.0 credits)
Application code
60243Start date
2013 week: 12End date
2013 week: 21Language of instruction
EnglishCampus
KTH CampusNumber of lectures
26 (preliminary)Number of exercises
Tutoring time
DaytimeForm of study
NormalNumber of places
No limitationSchedule
Schedule (new window)Teacher
Josephine Sullivan <sullivan@kth.se>
Target group
Compulsary: TMAIM Master Machine Learning, specialization Perception and Cognition but available for other programs.
Modular schedule in module D and I.
Part of programme
- Degree Progr. in Media Technology, year 4, Recommended
- Master (Two Years), Computer Science, year 1, CSCA, Conditionally Elective
- Master (Two Years), Human-Computer Interaction, year 1, HCIB, Optional
- Master (Two Years), Machine Learning, year 1, MAIA, Mandatory
- Master (Two Years), Machine Learning, year 1, MAIB, Conditionally Elective
- Master (Two Years), Machine Learning, year 1, MAIC, Conditionally Elective
- Master (Two Years), Systems, Control and Robotics, year 1, Recommended
- Master (Two Years), Systems, Control and Robotics, year 2, Recommended
- Master of Science in Engineering and of Education, year 4, MADA, Conditionally Elective
- Master's Program, Embedded Systems, year 1, Conditionally Elective
Spring 14 bik14 for programme students
Periods
Spring 14 P4 (6.0 credits)
Application code
60094Start date
2014 week: 13End date
2014 week: 23Language of instruction
EnglishCampus
KTH CampusNumber of lectures
26 (preliminary)Number of exercises
Tutoring time
DaytimeForm of study
NormalNumber of places
No limitationCourse responsible
Josephine Sullivan <sullivan@kth.se>
Target group
Compulsary: TMAIM Master Machine Learning, specialization Perception and Cognition.
Searchable for students at Master of Science in Engineering with at least 90 hp of which at least 50 hp from year 1 and for students at Master of Science in Engineering.
Part of programme
- Master (Two Years), Computer Science, year 1, CSCA, Conditionally Elective
- Master (Two Years), Computer Science, year 2, CSCA, Conditionally Elective
- Master (Two Years), Human-Computer Interaction, year 1, HCIB, Conditionally Elective
- Master (Two Years), Machine Learning, year 1, MAIA, Mandatory
- Master (Two Years), Machine Learning, year 1, MAIB, Conditionally Elective
- Master (Two Years), Machine Learning, year 1, MAIC, Conditionally Elective
- Master (Two Years), Media Technology, year 1, META, Conditionally Elective
- Master (Two Years), Media Technology, year 2, META, Conditionally Elective
- Master (Two Years), Systems, Control and Robotics, year 1, Recommended
- Master (Two Years), Systems, Control and Robotics, year 2, Recommended
- Master of Science in Engineering and of Education, year 4, MADA, Conditionally Elective
- Master's Program, Embedded Systems, year 1, Conditionally Elective
Learning outcomes
After the course you should be able to:
- identify basic notions, terminology, theories models and methods for classification of data,
- develop and systematically evaluate a number of basic methods for classification of data,
- experimentally evaluate algorithms for classification and recognition of objects in digital gray value images,
- choose appropriate method in order to automatically solve a given classification problem,
- know about theories of how the brain processes visual information for classification,
in order to
- be able to solve general problems of data representation and classification,
- be able to implement, analyze and evaluate simple systems for automatic classification of images,
- obtain a broad base of knowledge in order to be able to acquire information about and read literature in the field.
Course main content
- Representation and feature extraction in digital images
- principles of recognition and classification, Bayesian decisions
- discriminant functions, neural networks, support vector machines
- learning, optimization of classifiers
- overview of recognition in biological systems
- examples of recognition: handwritten text, faces, objects.
Eligibility
Single course students: 90 university credits including 45 university credits in Mathematics or Information Technology. English B, or equivalent.
Prerequisites
Knowledge corresponding to the compulsory courses on mathematics, computer science and numerical analysis on D-, E- or F-programme.
Literature
Material produced at the department.
Examination
- INL1 - Assignment, 1.5 credits, grade scale: P, F
- LAB1 - Laboratory Work, 1.5 credits, grade scale: P, F
- TEN1 - Examination, 3.0 credits, grade scale: A, B, C, D, E, FX, F
In this course all the regulations of the code of honor at the School of Computer science and Communication apply, see: http://www.kth.se/csc/student/hederskodex/1.17237?l=en_UK.
Requirements for final grade
Laboratory assignments (LAB1; 1,5 university credits)
Hand in exercise (INL1; 1,5 university credits)
Examination (TEN1; 3 university credits )
Offered by
CSC/Computer Science
Contact
Josephine Sullivan, tel: 790 6136, e-post: sullivan@kth.se
Examiner
Josephine Sullivan <sullivan@kth.se>
Add-on studies
Discuss with course leader.
Version
Course plan valid from:
Autumn 09.
Examination information valid from:
Autumn 07.
