Announcements

  1. (Sep 17) Course page is online.
  2. (Sep 17) Syllabus is available.
  3. (Sep 18) Slides for introduction are available.
  4. (Sep 24) Slides for digital image fundamentals are available.
  5. (Sep 27) Slides for binary image analysis are available.
  6. (Oct 4) Slides for linear filtering are available.
  7. (Oct 11) Homework assignment 1 is available.
  8. (Oct 11) Slides for edge detection are available.
  9. (Oct 18) Slides for local feature detectors are available.
  10. (Oct 30) Slides for color image processing are available.
  11. (Oct 30) Slides for texture analysis are available.
  12. (Nov 8) Slides for image segmentation are available.
  13. (Nov 10) Homework assignment 2 is available.
  14. (Nov 12) Slides for representation and description are available.
  15. (Nov 22) Slides for pattern recognition are available.
  16. (Dec 10) Homework assignment 3 is available.
  17. (Dec 11) Slides for image classification and object recognition are available.
  18. (Dec 11) Slides for image retrieval are available.
  19. (Dec 28) Project description is available.

Personnel

Instructor: Selim Aksoy (Office: EA 422, Email: )
TA: Hüseyin Gökhan Akçay (Office: EA 425, Email: akcay[at]cs.bilkent.edu.tr)

Course Information

Schedule: Tue 13:40-15:30, Thu 15:40-17:30 (EB 102)
Office hours: Selim Aksoy (TBD)
Catalog description: Image acquisition, sampling and quantization. Spatial domain processing. Image enhancement. Texture analysis. Edge detection. Frequency domain processing. Color image processing. Mathematical morphology. Image segmentation and region representations. Statistical and structural scene descriptions. Applications.
Course emphasis and goals: This course provides an introduction to image analysis and computer vision for undergraduates. We will start with low-level vision (early processing) techniques such as binary image analysis, filtering, edge detection and texture analysis. Then, we will cover mid-level vision topics such as image segmentation and feature extraction in detail. Finally, we will do case studies on several applications such as image retrieval, image classification, and object recognition. The emphasis will be on feature extraction and image representations for recognition.
Prerequisites: Good programming background, data structures, linear algebra, vector calculus, basics of signal processing. No prior knowledge of image processing or computer vision is assumed.
Syllabus: Make sure you read the syllabus for course details.

Texts

Lecture Schedule

Chapters

Contents

Introduction

[ Slides: pps | pdf ]

(Sep 18, 20)

Topics:
  • Overview
  • Example applications

Digital Image Fundamentals

[ Slides: pps | pdf ]

(Sep 25, 27)

Topics:
  • Acquisition, sampling, quantization
  • Image enhancement
  • Image formats
  • Linear algebra and MATLAB review
Readings:
  • SS Ch 1, 2
  • GW Ch 1, 2, 3.1-3.4
References:
Software:

Binary Image Analysis

[ Slides: Part 1 (pps | pdf) | Part 2 ]

(Oct 2, 4)

Topics:
  • Pixels and neighborhoods
  • Mathematical morphology
  • Connected components analysis
  • Automatic thresholding
Readings:
  • SS Ch 3.1-3.5, 3.8
  • GW Ch 2.5, 9.1-9.5, 10.3
References:
Software:

Linear Filtering

[ Slides: Part 1 (pps | pdf) | Part 2 (pps | pdf) ]

(Oct 9, 11)

Topics:
  • Spatial domain filtering
  • Frequency domain filtering
  • Image enhancement
Readings:
  • SS Ch 5.1-5.5, 5.10-5.11
  • GW Ch 3.5-3.8, 4
Software:

Edge Detection

[ Slides: pps | pdf ]

(Oct 16, 18)

Topics:
  • Edges, lines, arcs
  • Hough transform
Readings:
  • SS Ch 5.6-5.8, 10.3-10.4
  • GW Ch 10.1-10.2
References:
Software:

Local Feature Detectors

[ Slides: pps | pdf ]

(Oct 30, Nov 1)

Topics:
  • Corners and other interest points
  • Invariants
References:
Software:

Color Image Processing

[ Slides: pps | pdf ]

(Nov 6)

Topics:
  • Color spaces and conversions
Readings:
  • SS Ch 6.1-6.5
  • GW Ch 6

Texture Analysis

[ Slides: pps | pdf ]

(Nov 8, 13)

Topics:
  • Statistical approaches
  • Structural approaches
Readings:
  • SS Ch 7
  • GW Sec 11.3.3

Image Segmentation

[ Slides: pps | pdf ]

(Nov 15, 20)

Topics:
  • Histogram-based approaches
  • Clustering-based approaches
  • Region growing
  • Split-and-merge
  • Morphological approaches
  • Graph-based approaches
Readings:
  • SS Ch 10.1
  • GW Ch 10.4-10.5
References:
Software:

Representation and Description

[ Slides: pps | pdf ]

(Nov 22, 27)

Topics:
  • Image representations and descriptors
  • Region representations and descriptors
Readings:
  • SS Ch 10.2, 3.7
  • GW Ch 11
References:

Pattern Recognition Overview

[ Slides: Part 1 | Part 2 ]

(Dec 4, 6)

Topics:
  • Brief introduction to pattern recognition
Readings:
  • SS Ch 4
  • GW Ch 12.1-12.2
References:
Software:

Case Studies

[ Slides: Part 1 (pps | pdf) ] | Part 2 (pps | pdf) ]

(Dec 11, 13, 18, 20, 27)

Topics:
  • Image retrieval
  • Image classification
  • Object recognition
References:

Exams

Homework

  1. Homework assignment 1: description | data (Due: October 31, 2012 as online submission)
  2. Homework assignment 2: description | data (Due: November 28, 2012 as online submission)
  3. Homework assignment 3: description | data (Due: December 28, 2012 as online submission)

Please make sure you fully understand the late submission policy and the honor code for assignments in the syllabus. Cheating and plagiarism on homework assignments will be punished according to the regulations of the University as described in the Bilkent University Policy on Academic Honesty / Öğrenci Disiplin İlke ve Kuralları.

Project

The goal of the project is to develop an object recognition system based on clustering of segments represented using the bag-of-words model.

Grading Policy

Homework:35%
Quiz:10%
Exam:25%
Project:25%
Class participation:5%

Related Links