introduction to image administration processing - idc · introduction to image processing computer...
TRANSCRIPT
Introduction to Image Processing
Computer Science Semester A
Prof. Yacov [email protected]
Administration• Lectures / Tirgul
• Pre-requisites
• Course Home Page:– Syllabus, Reference List, “What’s new”
– Lecture slides and Handouts
– Homework, Grades
• Exercises: – Programming (Matlab), 5-6 Assignments
– Theoretical Assignments.
• Matlab software:– Available in PC labs
– Student version
– For next week: Run Matlab “demo” and read Matlab primer until section 13 .
Administration (Cont.)
• Grading policy: – Final Grade =
50% exam + 50% exercises– Exercises are compulsory– Exercises will be weighted – Final Exam with open material
• Office Hours: by appointment
Calendar
Oct 18Oct 25
Nov 01 Nov 08 Nov 15 Nov 22Nov 29
Dec 06Dec 13Dec 20Dec 27
Jan 03Jan 10
Digital Image ProcessingKenneth R. CastelmanPrentice Hall--------------------------------------Digital Image ProcessingRafael C. Gonzalez and Richards E. Woods, Addison Wesley --------------------------------------Digital Image ProcessingRafael Gonzalez and Paul Wintz
Addison Wesley --------------------------------------Fundamentals of Digital Image ProcessingAnil K. Jain
Prentice Hall, 1989. --------------------------------------
Course Books Image Processing - Lesson I
• Introduction to Image Processing
• Image Processing Applications
• Examples
• Course Plan
The Visual Sciences
Computer Vision
Rendering
ImageImage Processing
Model
3D Object
Geometric Modeling
Image Processing
Computer Vision
Low Level
High Level
Image Processing - Computer Vision
Acquisition, representation,compression,transmission
image enhancement
edge/feature extraction
Pattern matching
image "understanding“(Recognition, 3D)
Why Computer Vision is Hard?
• Inverse problems
• Apriori-knowledge is required
• Complexity extensive
– Top-Down v.s. Bottom-Up paradigm
– Parallelism
• Non-local operations
– Propagation of Information
Image Processing and Computer Vision are
Interdisciplinary Fields• Mathematical Models (CS, EE, Math)
• Eye Research (Biology)
• Brain Research:
– Psychophysics (Psychologists)
– Electro-physiology (Biologists)
– Functional MRI (Biologists)
ApplicationsRobotics
Object Recognition (assembly line)Autonomous VehiclesObstacle Avoidance
Arial photographyImage EnhancementMissile GuidanceGeological Mapping
AstronomyAstronomical Image EnhancementChemical/Spectral Analysis
MedicineComputerized Scanners (MRI,CT,etc)Radiological Organ Segmentation
MilitaryTrackingMappingDetection
GraphicsImage WarpingAnimationTexture/Image Mapping
Digital Camerasdemosaicingcolor manipulation
Robotics AVL
Astronomy: Meteorite detection
Military
Cruise Missiles
Military
Medical Imaging Medical Imaging
Compression Computer Vision + Computer Graphics
Image De-noising Image Enhancement
Image Enhancement
Original Noisy image Fourier Spectrum
Band Reject Filter
Image EnhancementFrequency Domain
Image Inpainting
Images of Venus taken by the Russian lander Ventra-10 in 1975
Image Inpainting
Video Processing
Y. Wexler, E. Shechtman and M. Irani 2004
Texture Synthesis
Example: 3D prior of 2x2 image neighborhoods
From Mumford & Huang, 2000
Mosaic Image from a digital camera
Pattern Matching Super Resolution
From: P. Milinfar MDSP software, UCSChttp://www.soe.ucsc.edu/~milanfar/SR-Software.htm
Video super-resolution
E. Shechtman, Y. Caspi, and M. Irani, 2002
Topics
• Image Acquisition• Image Operations
– Geometric Operations– Point Operation– Spatial Operation
• Frequency Domain and the FFT• Image Operations in Freq. Domain • Multi-Resolution • Feature Detection
Image Acquisition• Image Characteristics• Image Sampling (spatial)• Image quantization (gray level)
Using Different Number of Samples
N = 128
N = 64
N = 32
N = 16
N = 8
N = 4
bits=1 bits=2
bits=3 bits=4
bits=8
Using Different Number of Gray Levels
Image Operations
• Geometric Operations• Point Operations• Spatial Operations• Global Operations
Geometric Operations Point Operations
Geometric and Point OperationsSpatial Operations
Can we restore such an image?
Global Operations
• The Fourier Transform• Frequency Domain Operations
210 209 204 202 197 247 143 71 64 80 84 54 54 57 58 206 196 203 197 195 210 207 56 63 58 53 53 61 62 51 201 207 192 201 198 213 156 69 65 57 55 52 53 60 50 216 206 211 193 202 207 208 57 69 60 55 77 49 62 61 221 206 211 194 196 197 220 56 63 60 55 46 97 58 106 209 214 224 199 194 193 204 173 64 60 59 51 62 56 48 204 212 213 208 191 190 191 214 60 62 66 76 51 49 55 214 215 215 207 208 180 172 188 69 72 55 49 56 52 56 209 205 214 205 204 196 187 196 86 62 66 87 57 60 48 208 209 205 203 202 186 174 185 149 71 63 55 55 45 56 207 210 211 199 217 194 183 177 209 90 62 64 52 93 52 208 205 209 209 197 194 183 187 187 239 58 68 61 51 56 204 206 203 209 195 203 188 185 183 221 75 61 58 60 60 200 203 199 236 188 197 183 190 183 196 122 63 58 64 66 205 210 202 203 199 197 196 181 173 186 105 62 57 64 63
x = 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72y = 414243444546474849505152535455
Grayscale Image - Spatial Domain
= 3 + 5 +
+ 10 + 23 + ...
Grayscale Image - Frequency Domain
2 1 3
5 8 7
0 3 5
= 1 0 0
0 0 0
0 0 0
2 + 10 1 0
0 0 0
0 0 0
0 0 1
0 0 0
0 0 0
+ 3 + 50 0 0
1 0 0
0 0 0
+
+ ...
The Fourier Transform
Jean Baptiste Joseph Fourier 1768-1830
Freq. Domain Operations
High resolution
Low resolution
Multi-Resolution
Multiresolution Spline - Example Multiresolution Spline - Example
Original - Left Original - Right
Glued Splined
T h e E n d