Файл: Ersoy O.K. Diffraction, Fourier optics, and imaging (Wiley, 2006)(ISBN 0471238163)(427s) PEo .pdf

ВУЗ: Не указан

Категория: Не указан

Дисциплина: Не указана

Добавлен: 28.06.2024

Просмотров: 939

Скачиваний: 0

ВНИМАНИЕ! Если данный файл нарушает Ваши авторские права, то обязательно сообщите нам.

ONE-IMAGE-ONLY HOLOGRAPHY

265

Figure 15.22. A hologram generated with the FWZC method.

Figure 15.23. The reconstruction of concentric circles from the hologram of Figure 15.22.

15.10ONE-IMAGE-ONLY HOLOGRAPHY

For a plane perpendicular reference wave incident on the hologram, it is easy to observe that the existence of the twin images with various encoding techniques is due to the symmetry of the physical spaces on either side of the plane hologram. If we think of the object wave as a sum of spherical waves coming from individual object points, points that are mirror images of each other with respect to the hologram plane, correspond to the object waves on the hologram with the same

266 DIFFRACTIVE OPTICS I

amplitude and the opposite phase. Thus, when we choose a hologram aperture i that corresponds to the phase fi of the virtual image, the same aperture corresponds to the phase fi of the real image.

The conclusion is that the symmetry of the two physical spaces with respect to the hologram needs to be eliminated to possibly get rid of one of the images. This symmetry can be changed either by choosing a hologram surface that is not planar, or a reference wave that is not a plane-perpendicular wave. However, choosing another simple geometry such as an off-axis plane wave distorts only the symmetry and results in images that are at different positions than before.

An attractive choice is a spherical reference wave because it can easily be achieved with a lens [Ersoy, 1979]. If it becomes possible to reconstruct only the real image, the focal point of the lens can be chosen to be past the hologram so that the main beam and the zero-order wave can be filtered out by a stop placed at the focal point. In the following sections, we are going to evaluate this scheme, as shown in Figure 15.24.

In the encoding technique discussed in Section 15.7, the position of each

hologram aperture was chosen according to

 

jðxi; yiÞ þ kroi ¼ 2np þ f0

ð15:10-1Þ

In the present case, j(xi, yi) is the phase shift caused by the wave propagation from the origin of the reference wave front at (xc; yc; zc) to the hologram aperture at (xi, yi); kroi is the phase shift caused by the wave propagation from the aperture at (xi, yi) on hologram to an object point located at (xo, yo, zo). The radial distance roi is given by

 

¼ q

ð

 

Þ

roi

ðxo xiÞ2 þ ðyo yiÞ2 þ zo2

 

15:10-2

 

þ ( ) sign is to be used if the object is desired to be real (virtual).

For a spherical reference wave with its focal point at the position (xc, yc, zc), the phase of the reference wave jðxi; yiÞ can be written as

where

jðxi; yiÞ ¼ krci

ð15:10-3Þ

¼ q

 

 

 

 

ð

 

Þ

rci

ðxc xiÞ2 þ ðyc yiÞ2 þ zc2

 

15:10-4

 

Laser

Object

 

Hologram

Lens

Stop

Figure 15.24. The setup for one-image- only holography.


ONE-IMAGE-ONLY HOLOGRAPHY

267

where þ ( ) sign is to be used if the focal point of the lens is in the front (back) of the hologram.

Equation (15.10-1) can be written as

rci þ roi ¼ nl þ

f0l

ð15:10-5Þ

2p

Suppose that the position of the aperture is to be moved a distance in a defined direction, say the x-direction, such that Eq. (15.10-5) is satisfied. Then, the new lengths of the radius vectors roi0 and rci0 are given by

r0

¼ q

 

 

 

r2

 

 

2 x

0

x

i

Þ þ

2

oi

 

 

 

 

oi

ð

 

 

r0

¼ q

 

 

 

r2

 

 

2 x

 

x

iÞ þ

2

ci

 

 

 

 

ci

 

ð c

 

 

rci0 þ roi0 satisfy

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

rci0 þ roi0 ¼ rci þ roi B

 

Solving Eq. (15.10-8) for yields

 

 

p&

 

¼ C1

½

 

 

 

1

 

1 C2=C1

 

where

 

 

 

 

 

 

 

 

 

 

 

 

 

 

C1 ¼

F1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F2

 

 

 

 

 

 

 

 

 

 

 

C2 ¼

F3

 

 

 

 

 

 

 

 

 

 

 

F1

 

 

 

 

 

 

 

 

 

 

 

F1 ¼ F4ðX þ XcÞ þ Xrci2 þ Xcroi2

F2 ¼ ðrci þ roiÞ2 ðX XcÞ2 2F5

F3 ¼ 2F5roirci F52

 

 

 

 

 

F4 ¼ roirci F5

 

 

 

 

 

 

 

 

B2

 

 

 

 

 

 

 

 

 

F5 ¼

 

 

þ Bðrci þ roiÞ

 

 

 

 

 

2

 

 

 

 

ð15:10-6Þ

ð15:10-7Þ

ð15:10-8Þ

ð15:10-9Þ

ð15:10-10Þ

ð15:10-11Þ

ð15:10-12Þ

ð15:10-13Þ

ð15:10-14Þ

ð15:10-15Þ

ð15:10-16Þ

We note that amplitude modulation of the object points can still be achieved by varying the number of hologram apertures for each object point proportional to the desired amplitude at the object point. Even though the amplitude of the spherical wave varies slowly at the hologram, the average effect from randomly distributed apertures can be considered to be constant.



268

DIFFRACTIVE OPTICS I

15.10.1Analysis of Image Formation

In order to analyze the method, the lengths of the radii will be expanded in the paraxial approximation [Meier, 1966]. Instead of a spherical reference wave, suppose that a plane reference wave given by

Ur ¼ AejkðaxþbyþgzÞ

ð15:10-17Þ

is used. Then, Eq. (15.10-5) can be written in the paraxial approximation as

x

 

 

x0

 

 

y

 

y0

 

xi2 þ yi2

 

n

 

f0l

 

15:10-18

 

i

a z0

 

þ

i b z0

þ

2z0

¼

l þ

2p

ð

Þ

 

 

 

 

where the sign of z0 should be chosen the same as the sign used in Eq. (15.10-2). If a spherical reference wave as determined by Eq. (15.9-3) is used instead,

Eq. (15.10-5) can be written as

i

zc

þ z0

 

 

i

zc

þ z0

þ

 

2

z0 þ zc

¼

 

l þ

2p

 

0c

x

 

xc

 

x0

 

y

 

yc

y0

 

xi2

þ yi2

1 1

 

 

n

 

f0l

F

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ð15:10-19Þ

where

F

0c ¼

z

0

þ

x02 þ y02

þ

z

c þ

xc2 þ yc2

ð

15:10-20

Þ

2z0

2zc

 

 

 

 

Above the signs of z0 and zc should be chosen the same as the signs used in Eqs. (15.10-2) and (15.10-3).

An image forms whenever Eq. (15.10-18) or (15.10-19) is valid for arbitrary n, disregarding the constant terms f0l=2p and F0c. It is observed that if Eq. (15.10-18) is valid for the object point at ðx0; y0; z0Þ, the reference wave with direction cosines ða; b; gÞ, and wavelength l, it is also valid for an object point at ðx00; y00; z00Þ, the reference wave with direction cosines ða0; b0; g0Þ, and wavelength l0 such that

z00

¼

l z0

 

 

 

 

 

 

 

 

 

 

 

l0

 

m

 

 

 

 

 

 

 

 

 

 

 

 

x0

 

l

 

a0

ma

 

z

 

 

 

x

 

 

 

¼ l0

 

 

 

0

þ

0

0

b0

m

 

y0

 

l

 

mb

z

 

 

y

 

 

 

¼ l0

 

 

 

0

þ

0

0

 

 

m

 

 

 

ð15:10-21Þ

ð15:10-22Þ

ð15:10-23Þ

where m is an integer. m equal to 1 and –1 corresponds to real and virtual images while m equal to 0 corresponds to the zeroth order wave, a plane wave traveling in the same direction as the original reference wave. Other values of m correspond to higher order images.


ONE-IMAGE-ONLY HOLOGRAPHY

269

By the same token, if Eq. (15.10-19) is valid for the object point at ðx0; y0; z0Þ and the reference wave with focal point at ðxc; yc; zcÞ, whose wavelength is l, it is also

valid for the object point at ðx00 ; y00 ; z00 Þ and the reference wave with focal point at

ðxc0 ; yc0 ; zc0 Þ, whose wavelength is l0 such that

 

 

 

 

1

l0

 

 

1

 

 

 

1

 

 

 

 

 

 

 

 

 

¼

 

m

 

þ

 

 

 

 

 

 

ð15:10-24Þ

z00

l

z0

zc

 

 

 

 

 

 

 

 

l

 

 

xc

 

x0

x0

 

 

x00

¼ z00

 

m

 

 

þ

 

 

 

c

 

ð15:10-25Þ

l0

zc

z0

zc0

 

 

 

 

 

l

 

 

yc

 

y0

y0

 

 

y00

¼ z00

 

m

 

 

þ

 

 

 

c

 

ð15:10-26Þ

l0

zc

z0

zc0

where m is an integer. We note that m ¼ 0 corresponds to a wave that is the same as the reference wave and focuses at the focal point. If z0 zc, and assuming the initial and the final reference waves are the same, the image positions other than m ¼ 1 are

approximately given by

 

 

zc

 

 

 

 

 

 

 

 

z00

 

 

 

 

 

 

ð15:10-27Þ

 

 

 

 

 

m 1

z0

 

x0

x

c þ

m

0

x

0

ð

15:10-28

Þ

 

0

 

 

 

z0

 

y0

 

 

 

 

 

z0

 

 

 

 

y

c þ

m

0

y

0

ð

15:10-29

Þ

 

0

 

 

 

z0

 

In other words, if the focal point of the lens is sufficiently close to the hologram, all the images other than the one for m ¼ 1 have to be so close to the hologram that they become completely defocused in the far distance where the desired image is.

However, it is possible to focus one order at a time. If the mth order is desired to be at the position ðx00; y00; z00Þ, the corresponding values of ðx0c; y0c; z0cÞ are simply found such that Eqs. (15.10-24), (15.10-25), and (15.10-26) are satisfied. Then, all the other images become completely defocused and out of view in the far distance where the desired image is, because they are located very close to the hologram.

The above analysis is not always reasonable because the paraxial approximation becomes difficult to justify as the focal point of the spherical reference wave approaches the hologram. Thus, the images other than the one desired may not be well defined. It is also possible that the encoding technique used loses its validity for very close distances to the hologram, especially due to the registration errors of the hologram apertures.

In order to support the above arguments, consider what happens when z0c ! 1 in Eqs. (15.10-24)–(15.10-26); in other words, a plane wave is used for reconstruction.

Again assuming z0 zc, and l equal to l0, we get

 

z00

zc

ð15:10-30Þ

m

 

zc

x00

xc þ

ð15:10-31Þ

 

x0

z0

y00

yc þ

zc

ð15:10-32Þ

 

y0

z0