441-1 | Pattern Recognition | Computer Science | S9 | ||||||
---|---|---|---|---|---|---|---|---|---|
Lessons : 18 h | TD : 4 h | TP : 8 h | Project : 0 h | Total : 30 h | |||||
Co-ordinator : Luc Brun |
Prerequisite | |
---|---|
Bachelor in mathematics of computer science | |
Course Objectives | |
The objective of this course is to provide students with the basics of shape recognition: characterize objects (shapes, images, molecules,...), define measures of similarities between these objects and finally perform regression, indexing or classification tasks between these objects. |
|
Syllabus | |
The course is logically divided into two parts. The first part is dedicated to structural shape analysis and concerns the characterization and measurement of similarity of objects defined either by chains (music, DNA, movement,...) or by graphs (shapes, molecules, graphs of adjacency of an image,...). The course will study purely algorithmic methods as well as methods to create bridges between structural and numerical vision. The second part of the course is dedicated to statistical pattern recognition and will study all the statistical measures (such as moments) usually used in this field. |
|
Practical work (TD or TP) | |
1) Hungarian method 2) String and Image matching/embedding through the calculation string/graph edit distance and multidimensional embedding. |
|
Acquired skills | |
Be able to analyse a media (image/song, video) in order to take a decision. | |
Bibliography | |
Dupe, Francois-Xavier & Brun, Luc. Classification de formes avec un noyau sur graphes flexible et robuste au bruit. In Proceedings of RFIA'2010 , Caen , January 2010 .* https://en.wikipedia.org/wiki/Syntactic_pattern_recognition Richard O. Duda, Peter E. Hart, David G. Stork, Pattern classification, Wiley-interscience, 2001 |
© 2024 - ENSICAEN ( Legal Notices - Credits )