Course Title: Computer Vision - Universiti Putra Malaysia

lynxherringAI and Robotics

Oct 18, 2013 (3 years and 9 months ago)

57 views


1

FACULTY OF COMPUTER SCIENCE AND INFORMATION TECHNOLOGY

UNIVERSITI PUTRA MALAYSIA


Course Title
:

Computer Vision


Course Code
:

SMM5303


Credit
:


3(3+
0
)


C
ontact Hour
:

1 x 3 hour lecture per week


Prerequisite
:

None


Lecturer Name
:

Dr. Fatimah Khalid


Lect
urer Room
:

C2.03


Phone Number
:

603
-
89471730


Email

:


fatimahk@fsktm.upm.edu.my


Learning Outcome
:


At the end of the course
,

the student will be able to:

1.

compare recognition techniques from the perspective of its strength
and suitability in the computer
vision application. (C6) (A4)

2.

develop computer vision system. (P7)

3.

acquire idea and find out alternative solution for problems in computer
vision system. (CT3)


Synopsis:


This course
covers the discussion about components of digital image processing.
Adva
nced topics that are discussed include pattern recognition concept, colour
and shading, texture, motion from 2D image sequence, image segmentation,
matching in 2D, and 3D model and matching.


Course Contents:


Week


Topic

1

Vision System



Computer Vision
Application
s



Operations on images


presentation

on problems in Computer Vision Applications


2

2

Image Analysis



Introduction

o

System Model



Preprocessing

o

ROI Image Geometry

o

Image Algebra

o

Spatial Filters

o

Image Quantization


presentation/exercise

3

Image Analy
sis (cont’d)



Edge/Line Detection

o

Roberts Operator

o

Sobel Operator

o

Prewitt Operator

o

Kirsh Compass Masks

o

Robinson Compass Masks

o

Laplacian Operators

o

Frei
-
Chen Masks

o

Edge Operator Performance

o

Hough Transform


presentation/exercise

4

Image Analysis (cont’d)



Seg
mentation

o

Overview

o

Region Growing and Shrinking

o

Clustering Techniques

o

Boundary Detection

o

Combined Approaches

o

Morphological Filtering


presentation/exercise

5

Image Analysis (cont’d)



Discrete Transforms

o

Fourier Transform

o

Cosine Transform

o

Walsh
-
Hadamard Tra
nsform

o

Filtering

o

Wavelet Transform


presentation/exercise

6

Image Analysis (cont’d)



Feature Extraction and Analysis

o

Feature Vectors and Feature Spaces


3

o

Binary Object Features

o

Histogram Features

o

Color Features

o

Spectral Features

o

Pattern Classification


pres
entation/exercise

7

Image Restoration



Introduction



Noise



Noise Removal using Spatial Filters



Order Filters



Mean Filters



Adaptive Filters
-
Minimum Mean
-
Square Error
Filter



Frequency Domain Filters



Inverse Filter



Wiener Filter



Constrained Least
-
Squares Filte
r



Geometric Mean Filters



Notch Filter



Practical Considerations



Geometric Transforms



Spatial Transforms



Gray
-
Level Interpolation


presentation/exercise

8

Image Enhancement




Introduction



Gray
-
Scale Modification

o

Histogram Modification

o

Adaptive Contrast Enhan
cement

o

Color



Image Sharpening

o

Highpass Filtering

o

High
-
Frequency Emphasis

o

Homomorphic Filtering

o

Unsharp Masking



Image Smoothing

o

Mean and Median Filtering

o

Lowpass Filtering


presentation/exercise


4

9

Pattern Recognition Concept



Models for classification



Featu
re vector representation



Structural approach



Statistical approach



Syntactic approach



Artificial intelligence approach


presentation/exercise

10

Colour and Shading



Colour models



Colour Histogram



Color Segmentation



Shading


presentation/exercise

11

Texture



Texture and text cell



Texture description based on text cell



Texture measurement



Texture segmentation


presentation/exercise

12

Motion From 2D Image Sequences



Motion application



Image subtraction



Computing motion vectors


presentation/exercise

13

Matchi
ng in 2D



Representation of Points



Affine mapping



2D object recognition using Affine Transformation



2D Object Recognition using relational matching

Model and 3D Matching



Representation method



3D models



3D object recognition


presentation/exercise

14

Probab
ilistic and Inferential Methods



Finding Templates using Classifiers



Recognition by Relations between Templates

presentation/exercise


5

Evaluation:


1.

Course works




70%




Mid term




20%



Discussion in class




5% (individual)



Programming Assignments

25
%

(gro
up in 2)



Project




20%

(group in 4)


2.

Final examination




30%



References:


1.

Forsyth, D.A. and Ponce, J. (2008). Computer Vision: A Modern
Approach. New Jersey: Pearson Education International.


2.

Scott E Umbaugh. (2005).
Computer Imaging: Digital Image Analyis and
Processing

,

The CRC Press, Boca Raton, FL.


3.

Sonka, M., hlavac, V. and Boyle, R. (2007). Image Processing, Analysis
and Machine Vision. Cali
fornia: Brooks/Cole Publishing Company.