Pattern recognition

Course description

  • Introduction: definitions, pattern representations, pattern recognition by classification and analysis, applications of pattern recognition in economy, traffics, medicine, robotics, banking, forensics, man-machine communication, etc.
  • Pattern pre-processing: restoration, enhancement, normalization.
  • Pattern segmentation: basic idea,
    image segmentation, and
    auditory signals segmentation.
  • Features: generation of features by heuristic and mathematical methods.
  • Analysis of learning sets: pattern similarity measures, pattern clustering test, crisp and fuzzy clustering, clustering techniques, deep learning of generative models.
  • Pattern classification: classification of feature vectors by matching, decision, inference, and artificial neural networks; classification of sequences by dynamic programming and Hidden Markov Models; classification by graph matching; classification of statistically dependent samples.
  • Combining and fusing classifiers.

Course is carried out on study programme

Objectives and competences

To acquaint students with the advanced mathematical and computational approaches to pattern recognition by classification and analysis.

Learning and teaching methods

  • lectures,
  • individual consultations,
  • seminar projects.

Intended learning outcomes

After completion of the course the student will be able to demonstrate knowledge and understanding of:

  • developing systems based on recognition of external signals,
  • modelling rational capabilities of human beings (e.g. perception and cognition of the environment, learning),
  • state-of-the-art methods for pattern segmentation, feature extraction, clustering and classification.

During the course the student will gain and improve transferable skills such as:

  • use of information technology: the use of development tools (OpenCV, Weka Data Mining Toolkit), programming environments (Matlab, GNU Compiler Collection, Netbeans), programming languages (Matlab, C++, Java); and

problem solving: problem analysis, algorithm design, implementation and testing of a program.

Reference nosilca

  • Dobrišek S, Žibert J, Pavešić N, Mihelič F (2009) An edit-distance model for the approximate matching of timed strings. IEEE transactions on pattern analysis and machine intelligence, 31:736-741
  • Gajšek R, Mihelič F, Dobrišek S (2013) Speaker state recognition using an HMM-based feature extraction method. Computer speech & language, 27:135-150
  • Križaj J, Štruc V, Dobrišek S (2013) Towards robust 3D face verification using Gaussian mixture models. International journal of advanced robotic systems, 9:1-11
  • Dobrišek S, Gajšek R, Mihelič F, Pavešić N, Štruc V (2013) Towards efficient multi-modal emotion recognition. International journal of advanced robotic systems, 10:1-10
  • Justin T, Mihelič F, Dobrišek S (2014) Intelligibility assessment of the de-identified speech obtained using phoneme recognition and speech synthesis systems, Lecture Notes in Computer Science – Springer Verlag, 8655:529-536.

Study materials

  • Pavešić , N (2012) Razpoznavanje vzorcev : uvod v analizo in razumevanje vidnih in slušnih signalov –  3., popravljena in dopolnjena izdaja,  Založba FE in FRI, Slovenija
  • Murphy , KP (2012) Machine learning: a probabilistic perspective, MIT Press, Cambridge, MA
  • Theodoridis S, Koutroumbas K (2009) Pattern Recognition, Fourth Edition, Academic Press
  • Bishop, CM (2009) Pattern recognition and machine learning, Springer, New York

Bodi na tekočem

Univerza v Ljubljani, Fakulteta za elektrotehniko, Tržaška cesta 25, 1000 Ljubljana

E: T:  01 4768 411