Joseph Fourier (1768-1830) - PowerPoint PPT Presentation

1 / 26
About This Presentation
Title:

Joseph Fourier (1768-1830)

Description:

Joseph Fourier (1768-1830) Mathematical physicist in Napoleon s army Studied mathematical functions Any function (curve) can be decomposed into sine waves. – PowerPoint PPT presentation

Number of Views:74
Avg rating:3.0/5.0
Slides: 27
Provided by: arapahoN3
Category:
Tags: fourier | joseph

less

Transcript and Presenter's Notes

Title: Joseph Fourier (1768-1830)


1
Joseph Fourier (1768-1830)
  • Mathematical physicist in Napoleons army
  • Studied mathematical functions
  • Any function (curve) can be decomposed into sine
    waves.
  • Any curve can be created by summing sine waves.
  • Images are 2-D functions

2
Fundamental building blocks
  • Any image can be made from fundamental basic
    patterns
  • The fundamental patternsthe building blocks of
    any image are
  • sine wave gratings.

3
Sine wave grating parameters
  • Stripe size
  • Orientation (horizontal, vertical, etc.)
  • Contrast
  • Maximum (black on white) 1.0
  • Minimum (gray on gray) 0
  • Phase shift

4
Spatial frequency
  • Stripe size
  • Repeating pattern of black/white stripes
  • One black/white pair one cycle
  • Specify the number of cycles (black/white pairs)
    per degree of visual angle.
  • Spatial frequency
  • Low spatial frequency few big stripes
  • High spatial frequency many small stripes

5
Orientation
6
Contrast
1.00
0.40
0.20
0.00
Michelson formula
7
Build an image from 4 gratings




8
Fourier analysis of images
  • 50,000 sine-wave gratings
  • Each with appropriate
  • Spatial frequency
  • Contrast
  • Orientation
  • Phase shift

9
Test optical performance
  • All images are made up of sine wave gratings.
  • Optical engineers test how well an optical system
    works by
  • testing how it images sine-wave gratings.
  • Concept of contrast transfer

10
Contrast transfer
  • Optical systems transfer light from the object to
    the image.
  • Quality is always lost in the imaging process.
  • If the object is a sine-wave grating, the
    degraded image will be
  • a sine-wave grating with the same spatial
    frequency and orientation
  • But, some contrast will be lost.
  • Better optical systems transfer contrast better
    (lose less contrast)

11
How well do the optics transfer contrast?
1.0
1.0
1.0
1.0
Test pattern
Image
1.0
0.8
0.6
0.2
12
Contrast transfer function of the eye
13
Spatial frequencies in a image
Low spatial frequencies
High spatial frequencies
14
Requirements for good vision
  • Good optics good neural processing
  • MTF characterizes optics only
  • When we test vision we are testing the whole
    visual system (optics neural processing).
  • Contrast sensitivity testing
  • Like the MTF, uses sine-wave gratings
  • Differs in that vision (not only optics) is
    tested.

15
Contrast sensitivity
sensitivity
threshold
Spatial frequency
0.01
0.38
0.97
1
Sensitivity
threshold
16
Characteristics of normal CS
  • Peak at 4 c/d
  • Low frequency drop-off
  • Drops to zero at about 40-60 c/d
  • Cut-off frequency

Non-seeing
CS
seeing
Spatial frequency
17
Snellen size equivalent
E
E
CS
E
E
E
E
E
Spatial frequency
VA
1 cycle 2 arc minutes
18
Clinical contrast sensitivity tests
19
Improving visibility
  • Expand the curve
  • Correct refractive errors
  • Change non-visible objects to bring them inside
    the curve
  • Magnify
  • Increase contrast

Non-seeing
CS
seeing
Spatial frequency
20
Improving visibility by increasing contrast
Photo courtesy of Dr. Ralph Latimer
21
Visual system Fourier analyses images
Starry Night by van Gogh
Image without low and high spatial frequencies
22
Is VA obsolete? It depends.
  • It tests only one point on the contrast
    sensitivity functionthe cut-off frequency.
  • Uncorrected refractive errors primarily affect
    high spatial frequencies
  • So VA is good for clinical refraction.
  • Some diseases may affect low spatial frequencies
    more than high.
  • Contrast sensitivity better than VA for this

23
A better VA chartETDRS chart
  • Early Treatment of Diabetic Retinopathy Study
  • logMAR size progressionequal difficulty steps
    between all lines
  • Sloan letter setall letters equally difficult to
    read
  • Same number of letters (5) in every line
  • Same relative spacing between letters
  • Every letter counts as 1/5 of a line
  • Score VA to the letter (not just the line)

24
Scoring logMAR VA
  • 20/20 logMAR 0
  • Lower score better
  • Higher score worse
  • Each line 0.1 change
  • Each letter 0.02 change
  • Score VA to the letter

Negative values
25
Examples of logMAR scoring
  • Convert 20/402
  • 20/40 logMAR 0.3. Each additional letter
    subtracts 0.02 from score.
  • Therefore, 20/402 logMAR 0.26
  • Convert 20/40-2
  • 20/40 logMAR 0.3. Each letter missed adds 0.02
    to score.
  • Therefore, 20/40-2 logMAR 0.34

26
Snellen lt-gt logMAR conversion
X
x Snellen denominator
Write a Comment
User Comments (0)
About PowerShow.com