Convolution. Convolution Properties Commutative: f*g = g*f Associative: (f*g)*h = f*(g*h) Homogeneous : f*( g)= f*g Additive (Distributive): f*(g+h)=

Slides:



Advertisements
Similar presentations
CSCE 643 Computer Vision: Template Matching, Image Pyramids and Denoising Jinxiang Chai.
Advertisements

© by Yu Hen Hu 1 ECE533 Digital Image Processing Image Enhancement in Frequency Domain.
Slides from Alexei Efros
CS 691 Computational Photography
Linear Filtering – Part II Selim Aksoy Department of Computer Engineering Bilkent University
HCI/ComS 575X: Computational Perception Instructor: Alexander Stoytchev
Convolution. Spatial Filtering Operations g(x,y) = 1/M  f(n,m) (n,m) in  S Example 3 x 3 5 x 5.
Chapter 4 Image Enhancement in the Frequency Domain.
Psy 8960, Fall ‘06 Introduction to MRI1 Fourier transforms 1D: square wave 2D: k x and k y 2D: FOV and resolution 2D: spike artifacts 3D.
Image Filtering, Part 2: Resampling Today’s readings Forsyth & Ponce, chapters Forsyth & Ponce –
Sampling and Pyramids : Rendering and Image Processing Alexei Efros …with lots of slides from Steve Seitz.
Edges and Scale Today’s reading Cipolla & Gee on edge detection (available online)Cipolla & Gee on edge detection Szeliski – From Sandlot ScienceSandlot.
Fourier Transform Analytic geometry gives a coordinate system for describing geometric objects. Fourier transform gives a coordinate system for functions.
General Functions A non-periodic function can be represented as a sum of sin’s and cos’s of (possibly) all frequencies: F(  ) is the spectrum of the function.
Image transformations, Part 2 Prof. Noah Snavely CS1114
CSCE 641 Computer Graphics: Image Sampling and Reconstruction Jinxiang Chai.
Image Sampling Moire patterns
Lecture 4: Image Resampling CS4670: Computer Vision Noah Snavely.
Image Pyramids Pre-filter images to collect information at different scalesPre-filter images to collect information at different scales More efficient.
Lecture 3: Image Resampling CS4670: Computer Vision Noah Snavely Nearest-neighbor interpolation Input image 3x upsample hq3x interpolation (ZSNES)
2D Fourier Theory for Image Analysis Mani Thomas CISC 489/689.
Chapter 4 Image Enhancement in the Frequency Domain.
Image Sampling Moire patterns -
1.  Introduction  Gaussian and Laplacian pyramid  Application Salient region detection Edge-aware image processing  Conclusion  Reference 2.
ENG4BF3 Medical Image Processing
Image Representation Gaussian pyramids Laplacian Pyramids
Computer Vision Spring ,-685 Instructor: S. Narasimhan Wean 5409 T-R 10:30am – 11:50am.
2D Image Fourier Spectrum.
Signals and Systems Jamshid Shanbehzadeh.
Applications of Image Filters Computer Vision CS 543 / ECE 549 University of Illinois Derek Hoiem 02/04/10.
Image Processing in Freq. Domain Restoration / Enhancement Inverse Filtering Match Filtering / Pattern Detection Tomography.
Image Resampling CS4670: Computer Vision Noah Snavely.
Image Resampling CS4670: Computer Vision Noah Snavely.
Image Processing Xuejin Chen Ref:
Lecture 4: Image Resampling and Reconstruction CS4670: Computer Vision Kavita Bala.
Lecture 3: Edge detection CS4670/5670: Computer Vision Kavita Bala From Sandlot ScienceSandlot Science.
Lecture 7: Sampling Review of 2D Fourier Theory We view f(x,y) as a linear combination of complex exponentials that represent plane waves. F(u,v) describes.
Image Processing Basics. What are images? An image is a 2-d rectilinear array of pixels.
Computer Vision Spring ,-685 Instructor: S. Narasimhan Wean 5403 T-R 3:00pm – 4:20pm.
Gholamreza Anbarjafari, PhD Video Lecturers on Digital Image Processing Digital Image Processing Spatial Aliasing and Image Enhancement.
2D Image Fourier Spectrum.
Brent M. Dingle, Ph.D Game Design and Development Program Mathematics, Statistics and Computer Science University of Wisconsin - Stout Image Processing.
Lecture 5: Fourier and Pyramids
Computer vision. Applications and Algorithms in CV Tutorial 3: Multi scale signal representation Pyramids DFT - Discrete Fourier transform.
Projects Project 1a due this Friday Project 1b will go out on Friday to be done in pairs start looking for a partner now.
Digital Image Processing Lecture 8: Image Enhancement in Frequency Domain II Naveed Ejaz.
Convolution.
Jean Baptiste Joseph Fourier
Image Resampling & Interpolation
- photometric aspects of image formation gray level images
Degradation/Restoration Model
Image Sampling Moire patterns
(C) 2002 University of Wisconsin, CS 559
Convolution.
General Functions A non-periodic function can be represented as a sum of sin’s and cos’s of (possibly) all frequencies: F() is the spectrum of the function.
Image Sampling Moire patterns
Pyramid Coder with Nonlinear Prediction
Samuel Cheng Slide credits: Noah Snavely
Convolution.
Filtering Part 2: Image Sampling
Image Sampling Moire patterns
Convolution.
Nov. 25 – Israeli Computer Vision Day
Image Resampling & Interpolation
Resampling.
Chapter 3 Sampling.
Samuel Cheng Slide credits: Noah Snavely
Convolution.
Review and Importance CS 111.
Convolution.
Presentation transcript:

Convolution

Convolution Properties Commutative: f*g = g*f Associative: (f*g)*h = f*(g*h) Homogeneous : f*( g)= f*g Additive (Distributive): f*(g+h)= f*g+f*h Shift-Invariant f*g(x-x0,y-yo)= (f*g) (x-x0,y-yo)

The Convolution Theorem and similarly:

What is the Fourier Transform of ? Examples *

Image DomainFrequency Domain

The Sampling Theorem Nyquist frequency, Aliasing, etc… (on the board)

Gaussian pyramids Laplacian Pyramids Wavelet Pyramids Multi-Resolution Image Representation

Image Pyramid High resolution Low resolution

search Fast Pattern Matching Also good for: - motion analysis - image compression - other applications

The Gaussian Pyramid High resolution Low resolution blur down-sample blur down-sample

expand Gaussian Pyramid Laplacian Pyramid The Laplacian Pyramid - = - = - =

- = Laplacian ~ Difference of Gaussians DOG = Difference of Gaussians More details on Gaussian and Laplacian pyramids can be found in the paper by Burt and Adelson (link will appear on the website).

Computerized Tomography (CT) f(x,y) u v F(u,v)

Computerized Tomography Original (simulated) 2D image 8 projections- Frequency Domain 120 projections- Frequency Domain Reconstruction from 8 projections Reconstruction from 120 projections

End of Lesson... Exercise#1 -- will be posted on the website. (Theoretical exercise: To be done and submitted individually)