EE 576 – Machine Vision Spring 2018 Course Information
The aim of this course is to provide an overview of machine vision. This 1-semester course covers the fundamentals of the following topics:
Images, extraction of low-level features, boundary and region based analysis, segmentation and grouping, lightness and color, shape from shading. photometric and binocular stereo, optical flow and motion estimation, tracking, texture analysis.
Lectures: Mondays 11-12 @ Fourier, Wednesdays 9-11 Fourier
Instructor: Prof. H. Isil Bozma
Teaching Assistant: Kadir Türksoy kadir.turksoy at yandex.com
Prerequisites : C or C++, Working knowledge of Matlab & Simulink,
Image Processing (Advised, but not required)
There is no specific text book specified for the course. The recommended reading for this course includes:
- Robot Vision by Berthold Klaus Paul Horn (MIT Press, 1986)
- Machine Vision: Theory, Algorithms, Practicalities by E. R. Davies (Morgan Kaufmann Publications 2005) ·
- Machine Vision by Ramesh Jain, Rangachar Kasturi and Brian Schunk (McGraw Hill 1995).
- Introductory Techniques for 3-D Computer Vision by Emanuele Trucco and Alessandro Verri (Prentice Hall 1998)
Some additional references:
- Image Processing, Analysis, and Machine Vision by Sonka, Hlavac, and Boyle (ITP, 1999) .
- Digital Image Processing by Rafael C Gonzalez and Richard E Woods (Addison Wesley, 1992) MPSL 621.367 1992 DIG.
Grading will be based on class participation and performance, projects , one midterm and a final. The weights of each will be roughly.
Class participation 10%
The following are required to be able to take the final exam:
Projects – All projects must be turned in.
Pls note that I will be updating the notes as we progress throughout the semester.
Pls make sure that you get the most updated notes.
Tentative Schedule (Updated: 28.03.2018)
- Week 1 (5-9 Feb)
-Lecture presentation: Intro
Camera models and calibration
Lecture presentation: Camera and Image Formation
- Week 2 ( 12-16 Feb )
hw1: Pls do the exercise on opencv calibration exercise
- Week 3 (19 – 23 Feb )
Lecture presentation: Light and Image Formation
- Week 4 ( 26 Feb -2 Mar )
Lecture presentation: :color
Enhancement, Morphology, Binary images
Lecture presentation: :Enhancement,
- Week 5 ( 5 – 9 Mar)
Feature presentation: Features
- Week 6 (12 – 16 Mar )
- Week 7 ( 19 – 23 Mar)
Lecture notes: Segmentation
- Week 8 (26– 30 Mar)
Lecture presentation: Shape descriptors
Reference material: Kuhl-Giardina
- Week 9 (2 – 6 April)
Lecture presentation: Object Recognition
Single Image Viewing
Lecture presentation: Single image viewing
Midterm 1 – 2 April 2018 17-19 Monday
- Week 10 ( 9 – 13 April)
Motion Field and Optical Flow
Lecture presentation: Optical Flow
- Week 11 ( 16 – 20 April)
- Week 12 (23- 28 April)
Lecture presentation: TrackingPres
- Week 13 (30 Apr – 4 May)
Lecture presentation: 3D Motion Estimation
Lecture presentation: Binocular Stereo
- Week 14 (7 – 11 May)
In your projects, BE SURE TO REFER TO ANY LITERATURE/EXTERNAL CODE you have examined or used.
A separate REFERENCE section must be added to the end of all your reports.
Again, I assume that the projects have not been or are currently being done for other courses, etc.· You must present simulation results done in a statistical manner·
Pls send your source code, ( if applicable) data and report in a rar file with a name containing your name and project number – ie. LastnameFirstnamePrX.rar both to me and the course TA Kadir Turksoy. The report must be handed in a printed form either to me or to the course TA.
You must present a conclusion regarding your work.
The grading of the projects will be done as follows:
|Checking and verification of execution of code||30 pts|
|Header of the code inc. author, course,project no, etc.||5 pts|
|Documentation –where every class and method must be commented including comments of statements themselves.||15 pts|
|Programming discipline (indentations and ease of following code)||5 pts|
|Code ( Content of code, creativity, etc.)||45 pts|
OpenCV Intel’s OpenCV library, a very comprehensive open source vision library of C functions. A free source code version is available from SourceForge which you are recommended to download and install.