1

Prof. Shih-Fu Chang

Digital Video and Multimedia LabColumbia University

Jan. 22 2004http://www.ee.columbia.edu/dvmm

Digital Image Processing

Lecture 1: Introduction

Most images are downloaded from the web site of the textbook

2

Digital Image ProcessingProcessing of digital images by computersDigital images

Digital photos, image sequences, multi-sensor data like satellite images, medical images etc.

3

Some images from Mars Rover “Spirit”

Image feature registration is used to align the landing trajectory of the rover.

taken by Mars Global Surveyor

taken by Rover’s descent imaging

motion estimation (DIME) system

Color mosaic imageOf Mars surface

Images downloaded from the NASA/JPL web site

4

Some images by visible lights

Images by infrared lights – Visualize electricity energy consumption

6

Scanning Electronic Microscope (SEM)

7

Components of DIPRepresentation (Chap 2 and 6)

Human perceptual modelsHow to represent halftone, grey-scale, color images on the computers?How to determine spatial-temporal resolutions?

Enhancement (Chap 3)Contrast, noise, smoothness, sharpness

(online demos)

8

Components of DIP (2)Image Transform (Chap 4 and 7)

Pixel domain vs. transform domainFourier, Discrete Cosine Transform, KLT, Wavelet

Restoration (Chap 5)Remove degradation/blurring due to atmospheric interference, motion, noise, etc.

9

DIP Components (3)Feature extraction & segmentation (Chap 10 and 11)

Edge detection, connectionRegion segmentation and representationMotion estimation

Morphological Image Processing (Chap 9)Image/Video compression (Chap 8)Image Reconstruction from Projections (Jain Chap 10)

X-ray CT scanning

10

LogisticsRequired background

Signals and SystemsProbability, Linear Algebra, and some Random Process

TextbookGonzalez and Woods, Digital Image Processing, 2nd edition, Prentice Hall, 2001. (Required)Anil K. Jain, Fundamentals of Digital Image Processing, Prentice Hall, 1989. (reference)

Office HoursMondays 2-3:30pm, CEPSR Rm 709

Bi-weekly assignments (40%) including both analytical and programming experimentsOne midterm (30%), one final (30%), open books

11

Logistics (2)Software and data

MATLAB suggested, or other languagesA MATLAB recitation (Wed. Jan. 28 2004), EE Conference Room 1312 Mudd, 4:30pm-6pm.Computers available in

Mudd Rm 251 (PCs, ACIS accounts needed)EE Teaching Labs: Rm 1218 (SUN), Rm 1235 (Linux), EE access and accounts needed

Web SitesResources and bulletin board online on course web page Companion web site of the textbook

Background reviewSuggested projectsImages used in the book

12

Image Acquisition Models

13

Visual Perception Models

14

Image Acquisition Systems (2)

CT, PE, MRI, etc.

15

Image Acquisition Systems (1)

16

Sampling and Quantization

17

Sampling and Quantization

18

Matrix Representation