Holographic recording method and filtering method

Details Holographic recording method and filtering method Holographic recording method and filtering method

2000-05-25

2001-09-04

Schuberg, Darren (Department: 2872)

Optical: systems and elements

Holographic system or element

Fourier transform holography

C359S010000, C359S011000, C359S561000, C356S495000

Reexamination Certificate

active

06285474

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a method and an apparatus for recording and reconstructing various images as holograms, and to a method and an apparatus for performing diverse kinds of spatial frequency filtering such as low-pass filtering, high-pass filtering, wavelet transform and matched filtering during the process of holographic recording and reconstruction.
2. Description of the Related Art
The so-called holographic memory is getting attention for its ability to rapidly record and reconstruct images in increments of pages, and to record plural pages of images in multiplexed fashion within the same volume of an optical storage medium thereby storing huge quantities of image data. As such, the holographic memory is sometimes referred to as a promising next-generation computer file memory.
The holographic memory is used not only to record and reconstruct images but also to execute spatial frequency filtering for such purposes as image analysis, image compression, image decomposition and image retrieval. Spatial frequency filtering is a technique that involves carrying out optical Fourier transform to acquire a two-dimensional spectrum of input images and getting the acquired spectrum changed by use of a spatial frequency filter. This is a typical parallel optical computing technique capable of performing diverse kinds of computation and thus considered an optical convolution.
Typical devices of spatial frequency filtering (abbreviated to “filtering” hereunder where appropriate) include low-pass filters, high-pass filters and band-pass filters.
Generally, low-frequency components in the spectrum of an image correspond to an approximate structure of the image while high-frequency components constitute edges and fine structures of the image. A low-pass filter allows only low-frequency components of the image spectrum to pass through while blocking high-frequency component noise. Conversely, a high-pass filter lets high-frequency components alone to pass in order to extract image edges and highlight fine structures of the image. A band-pass filter permits only a specific spatial frequency band to pass through in such applications as image compression and analysis.
Holographic technology affords spatial frequency filters complex amplitude characteristics that provide optical correlation. This makes it possible to perform pattern recognition and retrieval.
Usually, undulating complex amplitude distributions cannot be recorded directly. Only light intensity distributions can be directly measured, and photography merely records intensity distributions. In contrast, holography permits recording and reconstruction of complex amplitude distributions by introducing carrier components into the distributions.
A filter that takes advantage of the above characteristic of holography is what is known as the matched filter. In computing correlations between a reference image and input images, the matched filter checks to see whether a specific pattern exists in a two-dimensional image, and detects the location of such a pattern if it is found to exist.
More robust variations of the matched filter have been proposed to implement higher capabilities of recognition. These variations include a phase-only matched filter (J. L. Horner et al., APPLIED OPTICS Vol. 23, pp. 812-816 (1984)) and a wavelet matched filter (Y. Sheng et al., OPTICS LETTERS Vol. 18, pp. 299-301 (1993)).
The phase-only matched filter records only phase components of a Fourier transformed image of an input object and phase components of a Fourier transformed image of a reference object. In addition to its enhanced efficiency of light utilization, the phase-only matched filter has correlation values in a &dgr; function through autocorrelation. This affords the phase-only matched filter a higher recognition capability than that of conventional matched filters.
Features of an image tend to concentrate on its contour portions. Taking advantage of this fact, the waveletmatched filter removes low-frequency components of the reference and input images through wavelet transform and computes correlations between the high-frequency components involved, thereby attaining a high recognition capability.
Conventional methods for holographic recording and reconstruction as well as for filtering are described below with reference to
FIGS. 23 through 25
.
For holographic recording, parallel light
1
is applied to an image
91
to yield object light
2
as shown in FIG.
23
. The object light
2
is subjected to Fourier transform by a lens
92
. Transformed object light
3
is applied to an optical storage medium
93
while plane wave reference light
5
is emitted simultaneously to the storage medium
93
to record a hologram therein.
For holographic reconstruction, as depicted in
FIG. 24
, the same reference light
5
as that in recording is emitted to the optical storage medium
93
. In response, the stored hologram yields diffracted light
6
A onto an optical path of object light. The diffracted light
6
A is subjected to inverse Fourier transform by a lens
94
. Transformed diffracted light
7
A is sent to a photo detector
95
to form an image thereon.
The reconstructed image is filtered as follows: a filter
100
having a two-dimensional transmittance distribution is interposed between the optical storage medium
93
and the lens
94
. The filter thus positioned extracts only desired spatial frequency components from a Fourier spectrum of the diffracted light
6
A.
Illustratively, low-pass filtering is carried out by getting the filter
100
to let pass only the low-frequency spectrum components at a central position of the Fourier spectrum of the diffracted light
6
A while blocking peripherally located high-frequency spectrum components. High-pass filtering is performed in a reverse fashion. That is, the low-frequency spectrum components at the central position are blocked and the filter
100
allows only the high-frequency spectrum components on the periphery to pass through.
For computation of correlations between images, parallel light
1
is applied to an image
96
to obtain object light
2
a
as shown in FIG.
25
. The object light
2
a
is subjected to Fourier transform by a lens
97
. Transformed object light
3
a
is used as retrieval light that is emitted in the manner shown in
FIG. 23
to the optical storage medium
93
that contains the image
91
as a hologram. In turn, the stored hologram yields diffracted light
6
B onto an optical path of reference light. The diffracted light
6
B is subjected to inverse Fourier transform by a lens
98
. Transformed diffracted light
7
B is sent to a photo detector
99
to form an image thereon.
For the recording method of
FIG. 23
, it is assumed for the purpose of simplification that a wave number vector k of object light conforms to the wave number of reference light; and that the object light
2
before Fourier transform is expressed as Oexp(−ikr), the object light
3
after Fourier transform as oexp(−ik′r), and the reference light
5
as R(=R*). On that assumption, a hologram T held on the optical storage medium
93
is defined by the expression (1) below. In the expression (1) and subsequent expressions, &agr; is used to signify proportion.
T&agr;|R+oexp(−ik′r)|
2
=|R|
2
+|o|
2
+R*oexp(−ik′r)+R*oexp(−ik′r)  (1)
For the reconstructing method of
FIG. 24
, the hologram T may be subjected to the same reference light R(=R*) as the reference light
5
in effect for recording. In that case, diffracted light is defined by the following expression:
RT&agr;{R|R|
2
+R|o|
2
+RR*oexp(−ik′r) +RRo*exp(−ik′r)}  (2)
Because the reference light R(=R*) is plane wave light and because the third term in the expression (2) above is diffracted onto the optical path of object light, reconstructed diffracted l

Affiliated with

Also associated with

Rating

Holographic recording method and filtering method does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Holographic recording method and filtering method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Holographic recording method and filtering method will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2525933