Početna stranicaVisoka učilištaKorisničke stranice
Vision Systems
Abbreviation: VISU PILoad: 30(L) + 0(E) + 15(LE) + 0(CE) + 0(PEE) + 0(FE) + 0(S) + 0(DE) + 0(P) + 0(FLE) + 0()
Lecturers in charge: prof. dr. sc. Bojan Jerbić
doc. dr. sc. Tomislav Stipančić
doc. dr. sc. Petar Ćurković
Lecturers: dr. sc. Bojan Šekoranja ( Laboratory exercises, Lectures )
dr. sc. Marko Švaco ( Laboratory exercises, Lectures )
dr. sc. Filip Šuligoj ( Laboratory exercises )
Josip Vidaković mag. ing. mech. ( Laboratory exercises )
Course description: Course objectives:
Gaining knowledge necessary for application of vision systems in automatic manufacturing, including digital image obtaining techniques, picture editing methods, edge and contour detection and object recognition.

Enrolment requirements and required entry competences for the course:


Student responsibilities:
Regulary attendance of exercises and lectures. Written and oral exam.

Grading and evaluation of student work over the course of instruction and at a final exam:
Attendance on lectures and exercises, final exam success rate.

Methods of monitoring quality that ensure acquisition of exit competences:
Coverage of quality:
immediately in the classroom, in direct communication with the students (student questions and discussion);
periodically, according to the processed education content, by grading of colloquiums and project;
exposition/discussing of results (scores) instructing students how to achieve better results.

Upon successful completion of the course, students will be able to (learning outcomes):
analyse the image, measure and recognise patterns by using the vision algorithms and techniques;
independently program and utilise the vision tools that are available at the market within various applications.

Lectures
1. Introduction: Artificial vision, artificial intelligence and robots, perception, vision, the role of the vision.
2. Pictures: Obtaining picture, optics, ccd. Digital image representation. Distorsions, filtering.
3. Taxonomy and paradigms of vision systems: Pixels, lines, borders, regions, objects, bottomup, topdown, neural networks, "low level", "intermidiate level" and "high level" vision.
4. Taxonomy and paradigms of vision systems: Pixels, lines, borders, regions, objects, bottomup, topdown, neural networks, "low level", "intermidiate level" and "high level" vision.
5. Detection and processing of regions: "Elementary" regions detection, partition, joining, quadtree structures morphologic region operations, features.
6. Edges: Convolution, gradients, laplace, filters, Roberts, Cany, examples.
7. Contours: Hough transform, boundary tracing, line fitting.
8. Recognition: Paradigms and strategies.
9. Recognition using neural networks: Neural networks, networks for shape recognition, (bpn... ).
10. Surrounding representation and matching (recognition by comparing): Clouds, generalized cylinders, semantic nets, matching group of lines and objects, object representations.
11. Surrounding representation and matching (recognition by comparing): Clouds, generalized cylinders, semantic nets, matching group of lines and objects, object representations.
12. Depth estimation: Triangulation, raster projection, shadow shape.
13. Stereovision: Systems with two and more cameras, depth from comparison of pictures from several cameras.
14. Active vision systems: Active coagency, stereo simulation, shape from movement dynamic vision.
15. Application of vision systems.

Exercises
1. Digital image representation.
2. Picture editing, morphologyc operations.
3. Picture editing, morphologyc operations.
4. Filter application.
5. Noise elimination.
6. Application of convolution masks for edge detection.
7. Application of convolution masks for edge detection.
8. Middle level algorythms detection of object contours.
9. Middle level algorythms detection of object contours.
10. Robot and vision system presentation of application.
11. Robot and vision system presentation of application.
12. Student worksolving of given problem using vision systems.
13. Student worksolving of given problem using vision systems.
14. Student worksolving of given problem using vision systems.
15. Student worksolving of given problem using vision systems.
Lecture languages: hr
Compulsory literature:
1. Machine vision; Ramesh Jain, Rangachar Kasturi, Brian G. Schunck
2. Computer Vision and Fuzzy Neural Systems; Arun D. Kulkarni
3. Uvod u raspoznavanje uzoraka, Ludvik Gyergyek, Nikola Pavešić, Slobodan Ribarić
Recommended literature: - - -
Legend
L - Lectures
FLE - Practical foreign language exercises
-
E - Exercises
LE - Laboratory exercises
CE - Project laboratory
PEE - Physical education excercises
FE - Field exercises
S - Seminar
DE - Design exercises
P - Practicum
* - Not graded
Copyright (c) 2006. Ministarstva znanosti, obrazovanja i športa. Sva prava zadržana.
Programska podrška (c) 2006. Fakultet elektrotehnike i računarstva.
Oblikovanje(c) 2006. Listopad Web Studio.
Posljednja izmjena 2019-06-07