metrology and sensing - iap.uni-jena.demetrology+and+sensing... · system properties bssic...

60
www.iap.uni-jena.de Metrology and Sensing Lecture 14: OCT 2018-02-01 Herbert Gross Winter term 2017

Upload: others

Post on 04-Sep-2019

32 views

Category:

Documents


2 download

TRANSCRIPT

www.iap.uni-jena.de

Metrology and Sensing

Lecture 14: OCT

2018-02-01

Herbert Gross

Winter term 2017

2

Preliminary Schedule

No Date Subject Detailed Content

1 19.10. Introduction Introduction, optical measurements, shape measurements, errors,

definition of the meter, sampling theorem

2 26.10. Wave optics Basics, polarization, wave aberrations, PSF, OTF

3 02.11. Sensors Introduction, basic properties, CCDs, filtering, noise

4 09.11. Fringe projection Moire principle, illumination coding, fringe projection, deflectometry

5 16.11. Interferometry I Introduction, interference, types of interferometers, miscellaneous

6 23.11. Interferometry II Examples, interferogram interpretation, fringe evaluation methods

7 30.11. Wavefront sensors Hartmann-Shack WFS, Hartmann method, miscellaneous methods

8 07.12. Geometrical methods Tactile measurement, photogrammetry, triangulation, time of flight,

Scheimpflug setup

9 14.12. Speckle methods Spatial and temporal coherence, speckle, properties, speckle metrology

10 21.12. Holography Introduction, holographic interferometry, applications, miscellaneous

11 11.01. Measurement of basic

system properties Bssic properties, knife edge, slit scan, MTF measurement

12 18.01. Phase retrieval Introduction, algorithms, practical aspects, accuracy

13 25.01. Metrology of aspheres

and freeforms Aspheres, null lens tests, CGH method, freeforms, metrology of freeforms

14 01.02. OCT Principle of OCT, tissue optics, Fourier domain OCT, miscellaneous

15 08.02. Confocal sensors Principle, resolution and PSF, microscopy, chromatical confocal method

Temporal coherence

Principle of optical coherence tomography

Light sources

Dispersion

Time domain OCT

Spectral domain OCT

Examples

OCT in biological tissue

3

Contents

Temporal Coherence

t

U(t)

c

duration of a

single train

Damping of light emission:

wave train of finite length

Starting times of wave trains: statistical

| ( ) |

c

Time-Related Coherence Function

( ) lim ( ) ( ) ( ) ( )* *

TT

T

TTE t E t dt E t E t

1

2

( ) ( ) ( )*0 E t E t IT

( )( )

( )

( ) ( )

( )

*

02

E t E t

E t

Time-related coherence function:

Auto correlation of the complex field E

at a fixed spatial coordinate

For purely statistical phase behaviour: = 0

Vanishing time interval: intensity

Normalized expression

Usually:

decreases with growing symmetrically

Width of the distribution: coherence time c

t /1

t

tA

sin)(

deAtE ti2)()(

Temporal Coherence

I()

Radiation of a single atom:

Finite time t, wave train of finite length,

No periodic function, representation as Fourier integral

with spectral amplitude A()

Example rectangular spectral distribution

Finite time of duration: spectral broadening ,

schematic drawing of spectral width

6

0

)( dSI

deS i2)()(

1c

dc

2)(

cc cl

Time-Related Coherence Function

Intensity of a multispectral field

Integration of the power spectral density S()

The temporal coherence function and the power

spectral density are Fourier-inverse:

Theorem of Wiener-Chintchin

The corresponding widths in time and spectrum are

related by an uncertainty relation

The Parceval theorem defines the coherence time

as average of the normalized coherence function

The axial coherence length is the space equaivalent of

the coherence time

7

8

Lateral and Axial Resolution

Intensity distributions

Aberration-free Airy pattern:

lateral resolution

axial resolution

lateral axial

Ref: U. Kubitschek

NADAiry

22.1

2NA

nRE

9

Lateral Resolution vs Depth of Focus

Gaussian beam as example:

Lateral resolution wo coupled to depth of focus zo

Increase of depth resolution : tight focussing

Measurement dilemma:

measurement of deep bore

holes with large divergence

impossible,

large depth of focus only for

bad lateral resolution

Imaging dilemma:

large spreaded light cone

gathers light from different

depth in volume imaging,

bad contrast conditions

2

0

0

0

0

2

wwz o

o

r

zwo

depth of focus

lateral resolution

weak

focussing

tight

focussing

zo=wo2/

Resolution in OCT

1. Axial resolution limited by spectral bandwidth

2. Lateral resolution: diffraction limited, improvement by confocal setup

3. Usually low NA

22

4413.02ln2

cohzLow NA High NA

2zdiff

2x 2x

2zcoh

2zdiff

10

Lateral resolution:

Airy profile

Penetration depth:

axial resolution

u

fx

sin

24

22ln2resz

Resolution of OCT

Log x

100 m

10 m

1 m

100 m 1 mm 1 cm 10 cm depth

lateral

resolution

confocal

microscopy OCT

ultra

sound

Log z

11

OCT Setup

Basic principle of OCT

Michelson interferometer

Time domain signal

receiverfirst mirrorfrom

source

signal

beam

reference

beam

beam

splitter

second

mirror

moving

overlap

lc

z z

relative

moving

I(z)

wave trains

with finite

length

-4 -2 0 2 4

0

0,2

0,4

0,6

0,8

1,0

-4 -2 0 2 4

0

0,2

0,4

0,6

0,8

1,0

primary

signal

filtered

signal

t

12

z

I

measured

signal

filtered

signal

measured

position

axial length of coherence

m mn

mnmnm IIII cos2

2

42

zzk

0,,2)()()( 2121 rrrIrIrI

)()(

),,(

21

21

minmax

minmax

rIrI

rr

II

IIK

Interference Contrast

Superposition of plane wave with initial phase

Intensity:

Radiation field with coherence function :

Reduced contrast for partial coherence

Measurement of coherence in Michelson

interferometer:

phase difference due to path length

difference in the two arms

(Fourier spectroscopy)

13

Basic method of optical coherence tomography:

- short pulse light source creates a coherent broadband wave

- white light interferometry allows for interference inside the axial coherence length

Measured signal:

- low pass filtering

- maximum of envelope gives

the relative length difference

between test and reference arm

For Gaussian beam:

axial coherence length

High frequency oscillation

depends on z

Principle of OCT

z

I(z)

signal

measured

filtered

signal

measured position

coherence length

14

24ln 2 o

cl

2

ol

2

42

zzk

Example:

sample with two reflecting surfaces

1. Spatial domain

2. Complete signal

3. Filtered signal

high-frequency content removed

15

Optical Coherence Tomography

Ref: M. Kaschke

Signature of OCT signale for thin layer measurements at the resolution limit

16

Optical Coherence Tomography

Ref: R. Leach

1. Decreas of contrast as a result of the coherence gating

17

Optical Coherence Microscopoy

Ref: R. Leach

Signal

spectral profile

Basic setup:

Michelson interferometer

OCT - White Light Interference

z-scan by moving

mirror

white light

source

spectrumI

signal

surface

under

test

z

I

18

Fiber Based OCT Interferometer

Basic setup

LED

source

source

spectrum

I

fiber coupler

reference arm

measuring

arm

surface

under test

z-scan

detector

fiber fiber

fiberfiberI signal

z

19

Achronyms in literature

20

Optical Coherence Tomography

Ref: R. Leach

Examples:

SLD = super luminescent diodes

21

OCT Light Sources

Ref: M. Kaschke

Typical light sources used for OCT

Light Sources of OCT

No Type of source wavelength

[nm]

axial resolution

remark

1 Superluminescent diode 800-830 10 m

2 Swept laser source 1050-1070 2.8 kHz swept rate

3 Supercontinuum fiber laser 450-1700

4 Photonics crystal fiber (PCF) illuminated by a fs-Ti:Sapphire laser

550-950 < 1 m [3]

6 PCF source 1300 2 m

5 Ti-Sapphire laser 675-975 1 m [4]

22

Left column:

optical spectrum

Right column:

signal in the spatial

domain

23

OCT Sources and PCI Signal

Multimode

laser diode

Super

Luminescent

Diode

Ti-Saphir-

Laser

Ref: M. Kaschke

Pulse transmission through dispersive medium

1. input pulse

2. after propagation with dispersion

24

Dispersion

Ref: M. Brezinsky

Dispersive material in OCT:

- wavelength-dependent phase delay

- group velocity dispersion

- degradation of the axial resolution

- the dispersion causes a distortion of the pulse

shape during propagation

Dispersion:

k not linear changing with /

Group velocity

1st derivative

Group velocity dispersion

2nd derivative

GVD Dispersion

0( , )

i t kzE z t E e

t k z

3 2(2)

2 2''

2

z d nk z

c d

3 2

2 22j

j

d nt D z z

c d

2 2

2 2 2

1 2

gr

d c d k d nD

d v d c d

0

2 ( ) ( )( )

n nk

c

25

oc

nk

)(2)(

d

dk

vgr

2

11

2 3 2

2 2 2

1 1

2 gr

d k d d nD

d dv v c d

Dispersion relation

Expansion

Rearrangement with as variable

introduction of group velocity and GVD

GVD Dispersion

2

00000

2

02

2

00

)(''2

1)(')(

)(

2

1)()()(

00

kkk

kkkkS

00 0

0

2 ( )( )

nk k

26

gvd

dnn

c

n

d

d

cd

dk

d

d

d

dk 11)(2

vDd

nd

cd

dk

d

d

d

kd

2

1)(

2

'2

2

2

3

2

2

Numerical stable calculation with Sellmeier formula

Refractive index

Derivatives

GVD Dispersion

3

12

2

1)(j j

j

L

Kn

2

2

1 j j

jj

K Ldn

d n L

222

2 32 3 2 2

31 1 j j jj j

j jj j

K L LK Ld n

d n nL L

27

Example:

n, n' and n'' for three types of glasses

GVD - Group Velocity Dispersion

BK7 SF6 FK54

0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2

n( )

1.51

1.52

1.53

dn/d

-150

-100

-50

0

0

2

4

6

8

1.75

1.8

1.85

1.9

-600

-400

-200

01.430

1.435

1.440

1.445

-80

-60

-40

-20

0

-2

0

2

4

6

0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2

x 105

d2n/d 2

0

20

40

60

0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2

28

Reference arm:

- allows for z-scan in the depth

- z-discrimination by axial length of coherence, defined by the bandwidth

(coherence gating)

- Large spectral width of illumination source:

good time/spatial z-resolution

Measured signals:

- reflected light and scattered light

- SNR above 10-10 can be resolved

Problems:

- refractive interfaces are dispersive

- group refractive index is important

Typical:

- fast axial scan by moving mirror or rotating cube (A-scan)

- slow lateral x-y-scans (B-scans)

Properties of OCT

29

30

OCT Measurement of Distances

Technical application of white light interferometry

OCT Enables 3D Imaging in Ophthalmology

OCT combines unique

resolution with high

sensitivity

• axial resolution is independent

of the numerical aperture (pupil)

• by heterodyne detection a

photon-noise limited sensitivity is

achieved (R 10-10)

optical resolution

z = lc 0.44 /

5 µm

z 2/NA²

1.7 µm (1.33 1/24)-2

555 µm

within Rayleigh range of imaging optics

Ref: M. Kempe

Dimensions of OCT imaging:

a) only depth (A-scan), one-dimensional

b) depth and one lateral coordinate (B-scan), two-dimensional

c) all three coordinates, volume imaging

32

Optical Coherence Tomography

Ref: M. Kaschke

Example of OCT Imaging

Example: Fundus of the human eye

33

OCT for Fundus Imaging

Ref: M. Kempe

OCT imaging in Ophthalmology

- between 800 nm and 1100 nm reasonable contrast

- here no dazzling of the eye

- lateral and depth resolution appropriate for single cells

35

White Light Interferometry

Examples

Ref: R. Kowarschik

Basic assumption: Michelson interferometer

Coherent field of the reference arm

factor 2: double pass in the arm

Coherent field of the sample/test arm

The spectral distributions E() correspond to a low coherence light source

(ultra short pulses)

Simplification: only a single mirror is assumed as sample

Interference signal

integrating over all spectral components

of the modulated signal part

Phase difference determines the fast

oscillating signal

depends on z-difference and on dispersion

36

Theory of Time Domain OCT

tizik

rorrreEE

)(2

tizik

sossseEE

)(2

*22Re2)( srsr EEEEI

deEE

dEEI

i

soro

srTD

)(*

*

Re2

Re2

rrss zkzk )()(2)(

With expansion of the dispersion functions

Definition of the spectral cross correlation

S() between refernce and sample arm

assumed to be symmetrical

Rs: reflectivity in sample arm

Phase difference with phase and group

velocity

Final result of TD-OCT signal

Interpretation:

- cos-prefactor: high oscillating term, proportional to time difference

- integral term: envelope of signal, depends on dispersion

is determined by the light source spectrum bandwidth

37

Theory of Time Domain OCT

gopo

g

o

p

o ttcc

z

)(2

od

dkkk os

)()()( 0

*)( soros EESR

deStRzI go ti

posTD )(cos)(

Spectral Domain OCT

Spectral Domain-OCT:

- broad band source

- reference mirror fixed in position, no A-scan necessary

- signal splitted by spectrometer

The high-frequency content of the signal is analyzed

The frequency is proportional to the depth z,

measured is the overlay beat-signal of all scatterers

Broadband

source

spectrometer

reference mirror

Sample

in z fixed

interferogramm

frequency domainFourier transform

spatial domain

data

processingz

z

38

Fourier Domain OCT

Fourier Domain-OCT:

setup

Signals:

a) intensity spectrum

b) spatial intensity distribution

39

Ref: M. Kaschke

Principle of OCT

Comparison of

1. time domain

2. spectral domain

source

k

z

z, t

z beam

splitter

intensity

Probe

reference mirror

Interference signal

TD:

sequentiell detection

with photodetector

SD:

parallel detection with

spectrometer

)cos(2~ zkIII rsIF

Intensity spectrum

Fourier-transform

and

Envelope

detection

envelope

detection

A-scan

A-scan

z

z k

Ref: M. Totzeck

40

Frequency Domain OCT has dramatic signal-to-

noise advantage

Time Domain OCT Frequency Domain OCT

Shot-

noise-

limited

detection

Spectral Domain OCT Swept Source OCT

Detection in M wavelength channels

TD-OCT

FD-OCT

R. Leitgeb et al., Opt.Expr. 11(2003), 889

𝑆𝑁𝑅𝑇𝐷 =𝜂𝑅𝑠𝑃0𝐸𝑝𝐵

𝑆𝑁𝑅𝑆𝐷 =𝜂𝑅𝑠𝑃0𝑀𝜏𝑖𝑛𝑡

2𝐸𝑝 𝑆𝑁𝑅𝑆𝑆 =

𝜂𝑅𝑠𝑃0𝑀𝜏𝑠𝑤𝑒𝑒𝑝2𝐸𝑝

• coherent addition of M

waves

• signals from all depths

contribute

1

𝐵= 𝜏𝑖𝑛𝑡 = 𝜏𝑠𝑤𝑒𝑒𝑝 for

𝑆𝑁𝑅𝑆𝐷 = 𝑆𝑁𝑅𝑆𝑆 =𝑀

2𝑆𝑁𝑅𝑇𝐷

Ref: M. Kempe

Properties of Fourier Domain OCT

the modulation frequency depends on the path length difference

simultaneous measurement of all backscattering contributions: larger sensitivity

- Fourier transform adds signals coherent

- noise is added incoherent

faster image processing due to missing A scan

signal drop-off with increasing depth

spectrometer resolution changes over depth

positive and negative z cannot be distinguished

42

Field in the reference arm

Field in the sample arm

Interference field

Discrete scattering model:

OCT signal

with reflectivities rj

Final signal evaluation

1st term: DC signal (underground)

2nd term cross correlation with

reference, interesting term

3rd term autocorrelation between

scatterers (small)

43

Fourier Domain OCT Signal

Rkzi

RR erkE

E20

2

),(

Skzi

SS ezrkE

E20 )(

2

),(

2

SRFD EEI

2

22

2

0

2

),(

2),(

j

tkzi

Sj

tkzi

RFDSjR erer

kEkI

j

SjSSjSS zzrzr )(

22( , ) ( )

4

( ) cos 2 ( )2

( ) cos 2 ( )4

FD R Sj

j

R Sj R Sj

j

Sj Sm Sj Sm

j m

I k S k r r

S k r r k z z

S k r r k z z

Fourier Domain OCT Signal

Typical Fourier Domain

OCT signal

only z-part

First term: DC

Second term: interference, cross-correlation, contains information

Third term: autocorrelation between scatterers

0.8 0.85 0.9 0.95 1 1.05 1.1 1.15 1.20

0.2

0.4

0.6

0.8

1

1.2

1.4

interference

signal

DC signal

contrast

mj

SmSjSmSj

j

SjRSjR

j

SjRFD

zzkrrkS

zzkrrkSrrkSkI

)(2cos)(4

)(2cos)(2

)(4

),(22

44

Fourier Domain OCT Signal

Signal complexity depends on

scatter-distribution

0.8 0.85 0.9 0.95 1 1.05 1.1 1.15 1.20

0.2

0.4

0.6

0.8

1

1.2

1.4

0.8 0.85 0.9 0.95 1 1.05 1.1 1.15 1.20

0.2

0.4

0.6

0.8

1

1.2

1.4

0.8 0.85 0.9 0.95 1 1.05 1.1 1.15 1.20

0.2

0.4

0.6

0.8

1

1.2

1.4

1 scatterer

2 scatterer

3 scatterer

rS(z

S)

zSz

S2z

S1z

R

reflectivity

samplereference

ID(z)

z-2(z

R-z

S1)0

A-scan

DC term

auto

correlation

cross correlation mirror image artifacts

-2(zR-z

S2)2(z

R-z

S2) 2(z

R-z

S1)

45

Fourier Domain OCT Signal

)(16

1

2)(

2)(ˆ)(ˆ)(

2

211 zrA

nn

rrz

n

rkSFkIFzI SC

SS

SR

S

S

)(

)()( 1

kS

kIF

r

nzr

R

SS

)(

114

)(

0

2

24

2

22

22

22

2

22

zI

eew

weee

ww

we

w

PPzI

zzp

S

nonzzzp

Snon

nonz

non

bSR

cohgate

ssbsssbs

Signal evaluation: inverse Fourier transform

DC-term subtracted by difference measurement

Auto-correlation: mostly negligible

Heterodyne efficiency:

Decrease in signal strength due to scattering underground for gaussian beams

46

Fourier Domain OCT Heterodyne Efficiency

Typical z-dependence

Dependence on scattering coefficient

for different focus locations

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 210

-7

10-6

10-5

10-4

10-3

10-2

10-1

100

z

(z)

single

scattering

multiple

scattering

0 2 4 6 8 10 12 14 16 18 2010

-7

10-6

10-5

10-4

10-3

10-2

10-1

100

(s)

s

z = 0.5

z = 1.0

z = 1.5

z = 2.0

47

Fourier Domain OCT Heterodyne Efficiency

Dependence on numerical aperture

for different scattering coefficients

Dependence of the mean signal

strength on z

(NA)

s = 2

0 0.05 0.1 0.15 0.2 0.2510

-7

10-6

10-5

10-4

10-3

10-2

10-1

NA

s = 10

s = 10

s = 40

<P>

NA0 0.05 0.1 0.15 0.2 0.25

102

103

104

105

106

48

OCT Heterodyne Signal Strength

Decrease with growing z

Peak in focus

z0 0.5 1 1.5 2 2.5

10-9

10-8

10-7

10-6

10-5

10-4

10-3

I(z)

focus

49

Fourier Domain OCT Example Calculation

k /ko

inte

nsit

y

0 50 100 150 200 250 3000

0.2

0.4

0.6

0.8

1FD Signal / IFT of deconvolved dI(k)

Depth z in my

refl

ecti

vit

y

0.9 0.95 1 1.05 1.10

0.2

0.4

0.6

0.8

1Interferogram

Only z-dependence

2 discrete scatterers

50

Fourier Domain OCT with Swept Source

Special setup for Fourier domain OCT:

- SS-OCT: swept source instead of broad spectral source

- source has tunable wavelength

- advantage: no spectrometer necessary

- usually faster than classical Fourier domain OCT

- tunable laser expensive

Narrow source

frequency sweep

detector

reference mirror

Sample

in z fixed

primary signal

frequency domain spatial domain

data

processingz

z

frequency

time

samplereference

51

Overview on OCT Setups

52

Different realization options for OCT:

Properties Time domain Spectral domain /

Fourier Spectral swept source

Source broad band broad band tunable laser

z-Scan, moving reference mirror yes no no

speed slow fast very fast

data gathering sequential parallel, spectrometer sequential

detector photodetector line camera photodetector

signal processing filtering, envelope Fourier transform Fourier transform

data encoding time frequency frequency

signal to noise behavior bad good good

Overview on OCT Setups

53

Ref: M. Kaschke

54

Scattering in Turbid Media

Different strengths of interaction

Ref: M. Gu

a) ballistic photon

b) snake photons

c) multiple

scattered photons

55

Scattering in Turbid Media

Change of light properties

Ref: M. Gu

a) spectral shift

frequency

frequency

b) spatial broadening

x

w

x

c) temporal broadening

time t t time t

snake

d) polarization

snake

56

Scattering in Turbid Media

Imaging of a circular disc through

a turbid medium with

growing scattering strength

a)...d)

Ref: M. Gu

Scattering in Tissue

Description of the light propagation in tissue:

1. Coefficient of absorption a

Loss of energy on the path.

2. Coefficient of scattering s

Probability of directional change per unit length of the path

3. Phase function p(q)

Mean angle distribution of the scattering process.

Frequently used model: Henyey-Greenstein

The sum of both coefficients is called the total extinction coefficient

sat

57

Henyey-Greenstein Scattering Model

Henyey-Greenstein model for human tissue

Phase function

Asymmetry parameter g:

Relates forward / backward scattering

g = 0 : isotropic

g = 1 : only forward

g = -1: only backward

Rms value of angle spreading

Typical for human tissue:

g = 0.7 ... 0.9

)1(2 grms

2/32

2

cos21

1

4

1),(

gg

ggpHG

forward

30

210

60

240

90

270

120

300

150

330

180 0

g = 0.5

g = 0.3

g = 0.7

g = 0.95

g = -0.5

g = -0.8

g = 0 , isotrop

z

58

Light Propagation in Tissue

Extended Huygens-Fresnel principle

Greens function with statistical

phase term

Coherent primary beam without scattering

Incoherent scattered intensity contributions

tissue

surface

n n1

signal

layerlens f

illuminating

beam

back

scattering

d

z

a

)',(

''22

22

2)',( rri

DrrrArB

ik

eeB

ikrrG

Pill

(z)

z

coherent

incoherent

straylight

')',()'()( 0 rdrrGrErE

59

Propagation in Tissue with Gaussian Beams

log Iill

(r,z)z

r

g = 0 g = 0.9 g = 0.99

0 1 2 3 410

-4

10-3

10-2

10-1

100

log Iill

(z)

z

g = 0

g = 0.5

g = 0.9

g = 0.99

log Iill

(r,zF)

r-0.4 -0.2 0 0.2 0.4

10-4

10-3

10-2

10-1

100

g = 0

g = 0.5

g = 0.9

g = 0.99

60