Saltar al contingut Saltar a navegacio
  • Enlarging letters
  • Home
  • Information
  • Contacting
  • Map

Computer Vision (VC)

Credits Dept.
7.5 (6.0 ECTS) ESAII

Instructors

Person in charge:  Alicia Casals Gelpi (alicia.casals@upc.edu)
Others:Antoni Grau Saldes (antoni.grau@upc.edu)
Jose Fernàndez Ruzafa (josep.fernandez@upc.edu)
Josep Amat Girbau (josep.amat@upc.edu)
Juan Aranda López (joan.aranda@upc.edu)
Juan Climent Vilaró (juan.climent@upc.edu)
Manel Frigola Bourlon (manel.frigola@upc.edu)

General goals

The aim of this subject is for students to acquire an understanding of the techniques used in processing and analysing images, which are basically oriented towards controlling them and enabling interaction with the environment. They should also understand the features and limitations of the algorithms and strategies used in order to be able to design vision and image processing systems. Furthermore, in the classes they define the specifications and select the components for a system, in addition to determining its configuration. Finally, they design and develop the process algorithms needed to satisfy the requirements of each application.

Specific goals

Knowledges

  1. The structure of a vision system. Stages and components
  2. Basic physical principles in forming images.



    Image acquisition and calibration.
  3. Digital processing of images. Local operators, linear transformation of images, filtering.
  4. 2D image processing techniques for interpreting or extracting information of interest from a scene.
  5. Techniques for extracting 3D/movement information.
  6. Vision techniques for guiding and controlling robots and robotic systems.

Abilities

  1. Drawing up technical specifications based on objective needs.



    Selection of components, designing the configuration of an application-oriented vision system.
  2. Ability to design and implement image processing algorithms.
  3. Ability to create procedures based on known generic algorithms and aimed at satisfying the application specifications.
  4. Characterising applications. Scenes - configuring image acquisition devices.
  5. Use of development environments and the use of image treatment APIs.

Competences

  1. Ability to express design solutions for systems through schemes, diagrams, graphics, and calculations.

Contents

Estimated time (hours):

T P L Alt Ext. L Stu A. time
Theory Problems Laboratory Other activities External Laboratory Study Additional time

1. Introduction to computer visualisation and image treatment
T      P      L      Alt    Ext. L Stu    A. time Total 
2,0 1,0 0 0 0 2,0 0 5,0
Introduction of concepts, definitions, and nomenclature related to computer-aided visualisation and image treatment. Understanding the structure, stages, and components of a visualisationsystem. Review of the physical and geometric concepts involved in the process of forming an image.

2. Image acquisition. The need for calibration and calibration techniques
T      P      L      Alt    Ext. L Stu    A. time Total 
4,0 1,0 2,0 0 2,0 4,0 0 13,0
introduction to various image acquisition equipment, with special reference to video cameras (which will be used in the practical work and in most of the exercises and problems). Study of image characteristics. Lighting systems. Introduction to the programme environment and the skeleton application to be developed.







  • Laboratory
    Development of basic functions at the pixel level in grey and colour images.
  • Additional laboratory activities:
    Preparation of the functions for implementation. Study of the programming environment.

3. Signal treatment at the pixel level
T      P      L      Alt    Ext. L Stu    A. time Total 
2,0 1,0 2,0 0 1,0 3,0 0 9,0
Understanding image format and content and study of image transformation techniques at the pixel level.







  • Laboratory
    Creating a histogram of an image and transforming the image. Binarisation.
  • Additional laboratory activities:
    Design of algorithms producing image histograms and image transformations

4. Digital processing of images. Local operators, linear transformations of images, filtering.
T      P      L      Alt    Ext. L Stu    A. time Total 
5,0 1,0 4,0 0 4,0 8,0 0 22,0
Study and development of basic functions in binary and multi-level images at the pixel-sized level. Implementation of filters and extraction of outlines.







  • Laboratory
    Introduction to the use of a given 2D or 3D image treatment or visualisation system.

5. Image segmentation techniques
T      P      L      Alt    Ext. L Stu    A. time Total 
2,0 0 2,0 0 1,0 4,0 0 9,0
Classification of segmentation techniques and study of the processing techniques based on parametres, statistical calculations, structural analysis of images, etc. Algorithms for labelling the various regions obtained.











  • Additional laboratory activities:
    Design of labelling algorithms.

6. Extracting characteristics
T      P      L      Alt    Ext. L Stu    A. time Total 
2,0 2,0 2,0 0 1,0 3,0 0 10,0
Study and design of algorithms for parameterizing certain characteristics of objects or regions contained in the image.











  • Additional laboratory activities:
    Design of algorithms for extracting results and presenting the results.

7. Pattern recognition. Classification of objects
T      P      L      Alt    Ext. L Stu    A. time Total 
2,0 0 0 0 0 4,0 0 6,0
Study of object recognition techniques based on the characteristics of an object or image region.

8. Detection and localisation
T      P      L      Alt    Ext. L Stu    A. time Total 
2,0 0 0 0 0 2,0 0 4,0
Study of detection and localisation of scene elements from extracted characteristics or from the segmented image.

9. Extracting 3D information
T      P      L      Alt    Ext. L Stu    A. time Total 
3,0 2,0 0 0 0 4,0 0 9,0
Study of the extraction of 3D data through stereo vision, laser triangulation, light structure or movement.

10. Analysis of movement
T      P      L      Alt    Ext. L Stu    A. time Total 
1,0 1,0 2,0 0 3,0 4,0 0 11,0
Study of movement detection and measurement based on image differences or optical flux in a sequence of images. Acquisition of moving images and monitoring of an object in a sequence of images.











  • Additional laboratory activities:
    Design of an object-monitoring algorithm and the need to optimise the algorithm to work in real time.

11. Computer vision applications. Methodology for handling computer vision projects.
T      P      L      Alt    Ext. L Stu    A. time Total 
5,0 3,0 6,0 0 4,0 10,0 0 28,0
Study of the working conditions, requirements and techniques found in the various fields of computer vision (with particular reference to robot guidance, inspection, detection, and dimensional measurement).







  • Laboratory
    Producing documentation for the application: specifications and stages.
  • Additional laboratory activities:
    Study of the problem requiring solution and the design of the algorithm to be implemented.

12. Enhanced reality
T      P      L      Alt    Ext. L Stu    A. time Total 
1,0 0 2,0 0 0 2,0 0 5,0
Identifying the need to provide an image that stresses certain characteristics in order to make it easier for someone to interpret. This is achieved by superimposing graphic elements on the sparsest areas of the image, or on areas of special interest after localising the regions to which information needs to be added.

13. Work environments, processing libraries and commercial systems.
T      P      L      Alt    Ext. L Stu    A. time Total 
2,0 0 4,0 4,0 2,0 2,0 0 14,0
  • Laboratory
    Information search for and study of commercial vision systems.
  • Additional laboratory activities:
    Reading and comprehension of technical documentation/manual.


Total per kind T      P      L      Alt    Ext. L Stu    A. time Total 
33,0 12,0 26,0 4,0 18,0 52,0 0 145,0
Avaluation additional hours 5,0
Total work hours for student 150,0

Docent Methodolgy

Given that the course is design-oriented, no distinction will be made between theory classes and problems. Theory classes are supplemented with examples of real applications, illustrating design alternatives. This methodology develop students" critical faculties and learn how to make the right trade-offs in order to optimise design.

Evaluation Methodgy

Final grade = 0.7*Theory grade + 0.3*Lab grade
Theory grade = 0.4*P + 0.6*F o F (if F > Theory grade).
P : Partial Exam
F : Final exam.
Lab grade:= The weighted average of the scores of each practice. The weight depends on the practice complexity and will be indicated at the beginning of each course.

(All grades out of 10)

Assessment of practical work

Assessment of the practical work will be based on the following:
1.Presentations in each of the above work sessions 0.3
2. Final presentation of the software produced and the resulting application. 0.5
3. Reports and the developed software 0.2

Basic Bibliography

  • Ramesh Jain, Rangachar Kasturi, Brian G. Schunck Machine vision, McGraw-Hill, 1995.
  • Gonzalo Pajares Martinsanz, Jesús M. de la Cruz García Visión por computador : imágenes digitales y aplicaciones, Ra-Ma, 2001.
  • Arturo de la Escalera Hueso Visión por computador : fundamentos y métodos, Prentice Hall, 2001.

Complementary Bibliography

  • John C. Russ The Image processing handbook, CRC Press, 2002.
  • Javier González Jiménez Visión por computador, Paraninfo, 2000.
  • Rafael C. González, Paul Wintz Digital image processing, Addison-Wesley, 1987.
  • Carme Torras [ed.] Computer vision : theory and industrial applications, Springer-Verlag, 1992.
  • Milan Sonka, Vaclav Hlavac and Roger Boyle Image processing, analysis and machine vision, PWS, 1999.

Web links

  1. Obrir nova finestra http://homepages.inf.ed.ac.uk/rbf/BOOKS/BANDB/bandb.htm


  2. Obrir nova finestra http://www.eeng.dcu.ie/~whelanp/ivsi/IVSI.pdf


  3. Obrir nova finestra http://homepages.inf.ed.ac.uk/rbf/CVonline/CVentry.htm


Previous capacities

The knowledge acquired on previous courses. No additional preparation is required.


Share this page
  • Forum
  • Twitter
  • Menéame
  • Delicious
  • Yahoo! bookmarks
  • Google bookmarks
  • LinkdedIn
  • Facebook

 
logo FIB © Barcelona school of informatics - webmaster@fib.upc.edu - RSS