Get documents about "Lecture 2 - Image acquisition, Digitization
Document Sample
Lecture 2 –
Image acquisition,
Digitization
Lennart Svensson
Centre for Image analysis
Swedish University of Agricultural Sciences
Uppsala University
2
Overview
the world
imaging
visualization image processing
image analysis
data image
computer graphics
lennart@cb.uu.se
3
Reading
● 2.2-2.6
lennart@cb.uu.se
4
Outline
● Acquisition process:
Detected signal
Sensors
Digitization, digital images
● Pixel relationships
● 3D imaging, tomography
lennart@cb.uu.se
5
Questions
● What is a digital image?
● How are they acquired?
● How are resolution, sampling and quantization
related?
● How can we measure distances in a digital
image?
lennart@cb.uu.se
6
Human vision
lennart@cb.uu.se
7
Electronic vision
● f (x,y)
lennart@cb.uu.se
8
What is detected?
● Electromagnetic waves (most technologies)
● Sound (ultrasound)
● Particles (electron microscopy)
● Mechanical contact forces
(scanning probe microscopy)
lennart@cb.uu.se
9
Electromagnetic waves
Image from Wikimedia Commons
lennart@cb.uu.se
10
Electromagnetic imaging
● Radio range: radio astronomy, MRI
● Microwave range: RADAR
● Visible range: Standard camera, light
microscopy
● X-ray range: CT, micro-CT
● Gamma range: Gamma camera
lennart@cb.uu.se
11
Imaging modalities, examples
MRI
Images from Wikimedia Commons
lennart@cb.uu.se
12
Imaging modalities, examples
● Infrared ● Confocal
microscopy
Image from Wikimedia Commons
lennart@cb.uu.se
13
Imaging modalities, examples
gold markers
antibody
antigen
complex
● Electron Tomography
Image from Wikimedia Commons
lennart@cb.uu.se
14
Sensors
● To image something we need sensors that can
capture the energy of the signal.
● The output from each sensor element is a
positive finite value proportional to the received
energy
Single sensor Sensor strips Sensor array
lennart@cb.uu.se
15
CCD and CMOS
● Photon sensitive elements that accumulate an
electrical charge
● Detecting visible light, IR, UV
● CCD – Nobel Prize in Physics 2009
lennart@cb.uu.se
16
From press release last week
● ”The CCD technology makes use of the
photoelectric effect, as theorized by Albert
Einstein and for which he was awarded the
1921 year's Nobel Prize. By this effect, light is
transformed into electric signals. The challenge
when designing an image sensor was to gather
and read out the signals in a large number of
image points, pixels, in a short time.”
lennart@cb.uu.se
17
CCD
lennart@cb.uu.se
18
CCD vs CMOS
● Dynamic range – ratio between the largest and
smallest measurable intensity value, favor CCD
● Shuttering, favor CCD
● Speed, favor CMOS
● Windowing, favor CMOS
lennart@cb.uu.se
19
Exposure
● Rolling shutter – exposure line-by-line, common
in ordinary video cameras
● Global shutter – all sensor elements are
exposed the same time period
lennart@cb.uu.se
20
Color sensors
● Filters (Bayer filter most common)
● Dichroic prism
● Different sensing elements (Foveon X3 sensor)
Images from Wikimedia Commons
lennart@cb.uu.se
21
Sensor data
● The continuous value needs to be possible to
represent in a computer, it needs to be digitized
● Discretization of amplitude values is called
quantization
● Sampling is the process of discretizing a
continuous function in terms of coordinate
values
lennart@cb.uu.se
22
Digitization
Uniform sampling Uniform quantization
Digitization
● Sampling rate – spatial resolution
● Quantization - grey level resolution
Images from Wikimedia Commons
lennart@cb.uu.se
23
Interpolation
The process of using known data to estimate
values at unknown locations
Images from Wikimedia Commons
lennart@cb.uu.se
24
Aliasing
● Aliasing refers to an effect that causes different
signals to become indistinguishable when
sampled
Images from Wikimedia Commons
lennart@cb.uu.se
25
Different sampling
lennart@cb.uu.se
26
Different quantization
lennart@cb.uu.se
27
Sampling arrangements
● Uniform sampling
Square grid, Rectangular grid, Hexagonal grid
● Non-uniform sampling
Closer where it is necessary, eye.
● Image size
128*128, 256*256, 512*512, 1024*1024
● The sampling is normally determined by the
sensor arrangement
lennart@cb.uu.se
28
Digital image
● f (x,y) – a (x,y) coordinate yields one or more
outputs
lennart@cb.uu.se
29
General digital image
● f (x,y,z,t,b)
x,y,z – spatial parameters
t – time
b – spectral
● Each dimension and the function value must be
quantized into a limited range of discrete values
lennart@cb.uu.se
30
Resolution
● Spatial resolution
▬
how many picture elements?
● Temporal resolution
▬
time of exposure
▬
number of images per second
● Spectral resolution
▬
range of wave-lengths
▬
number of colors
● Gray scale resolution (quantization)
▬
radiometric - measurement of electromagnetic radiation
▬
how many gray levels?
lennart@cb.uu.se
31
Volume images
Volume: f (x,y,z)
Image from Wikimedia Commons
lennart@cb.uu.se
32
Color image
f (x,y,b)
x,y – spatial parameters
b – spectral
lennart@cb.uu.se
33
Representation
lennart@cb.uu.se
34
lennart@cb.uu.se
35
Distances
lennart@cb.uu.se
36
Connectivity
lennart@cb.uu.se
37
3D imaging
● Distance measuring (laser, SEM)
● Multi-camera
● Tomography
lennart@cb.uu.se
38
Tomography
● Definition according to Merriam-Webster:
A method of producing a three-dimensional
image of the internal structures of a solid object
(as the human body or the earth) by the
observation and recording of the differences in
the effects on the passage of waves of energy
impinging on those structure
lennart@cb.uu.se
39
Tomography
Physical phenomenon Type of tomography
X-rays CT
Gamma rays SPECT
Electron-positron PET
annihilation
Electrons Electron tomography
lennart@cb.uu.se
Related docs
