Deblurring of Motion-Blurred Photographs
Using
Extended-Range Holographic Fourier-TransformDivision,
G.
W.
Stroke, F. Furrer and
D.
Larnberty, Optics Corrununications, Vol.
A,
No.
3,
July/August
1969.
Technical Report No. 188
Volume 1, number
3
OPTICS COMMUNICATIONS July/August 1969
DEBLURRING OF MOTION-BLURRED PHOTOGRAPHSUSING EXTENDED-RANGE HOLOGRAPHIC FOURIER-TRANSFORM DIVISIONGeorge
W.
STROKE, Franz
FURRER
*
and Donald R. LAMBERTY
**
Electro-Optical Sciences Center, State University
of
New York,Stony Brook, New York
11790,
USA
Received
2
June 1969The method of a posteriori image-correcting deconvolution by holographic Fourier-transform division(Stroke and Zech, Physics Letters 25A (1967) 89) has been successfully extended to the deblurring ofphotographs blurred by motion with the aid of
a
photographic (positive
+
negative) nmaskingn process,used to obtain the required great linear dynamic range (especially in the Fourier-transform domain), inanticipation of extended-range high-resolution films now under development.
There has arisen an increasing interest inmethods which permit one to extract
a
posterioria sharpened image from
a
photograph blurred
as
a result of various instrumental and recordingimperfections, including blurring by atmosphericturbulence and indeed by motion. In a generalway optical image deblurring methods are basedon the principles of 'spatial filtering' describedby Mar6chal and Croce
[I].
Considerable successin implementing these principles were demon-strated already
a
decade ago by Tsujiuchi [2]who showed that the required complex (amplitudeand phase) filters [3] could be realized by
a
com-bination of two filters: the amplitude filter,realized by photography, and the phase filter,realized by vacuum evaporation. The Tsujiuchimethod
is
particularly suitable for the prepara-tion of filters which may be described in analyti-cal form. It has been used, for instance, to de-blurr images blurred by artificial (laboratory)turbulence and indeed for certain types of motion[4]. Several other image-deblurring methodshave also been demonstrated, including digital-computer implementation of the required Fourier-transform division [5] as well
as
purely elec-tronic methods, particularly suitable in conjunc-tion
with
television devices
[6,7].
The relativemerits of these and other methods are formingthe subject of current investigations
by
a
numberof authors. Somewhat in analogy with the resultsof the investigations of Jacquinot and Chabbal,with regard to relative merits of various formsof spectroscopy, it
is
probable that each of theimage-deblurring methods may be particularlysuitable for some types of applications, while
not
necessarily being the most powerful generalmethod. However,
in
keeping with the suggestionby Stroke and Zech
[8]
it
has become increasing-ly apparent
that
the most general method for op-tical image deblurring will no doubt be mostreadily implemented by holographic means. Thegeneral background for the holographic imagedeblurring methods may be found
in
refs. [8-111,in which we also discuss the difference betweenthe holographic image restoration methods (solu-tion of an integral equation) on the one hand, and,on the other, the correlative character recogni-tion methods [12] Vormation of an integral equa-tion).
In
addition to the methods of refs. [8-111,which use a holographically synthesized filter forthe purpose of solving the integral equation (con-volution integral) by division in
the
spatial Four-ier-transform domain, Stroke [13,14] has alsoproposed
a
different
type
of holographic methodwhich permits one to achieve 'image-deblurring'as well as 'aperture synthesis' simply by takinga hologram of the blurred photograph and by
il-
luminating
it
with the light from the spread func-tion
h(x,
y),
in one of the arrangements. Thesemethods [13,14] are particularly suitable for thecase when the auto-correlation function of thespread function
is
sharply peaked. Detailedcomparison of the various holographic methodsare given in ref. [15]. It may be of interest tofurther note that spatial Fourier-transform
divi-
*
Visiting staff, from the Photographisches Institut(Director, Professor Dr.
W.
F.
Berg), Eidgen6s-sische Technische Hochschule, Zurich.
**
Visiting staff, from the CBS Laboratories, Stam-ford, Connecticut.
Volume 1, number
3
OPTICS COMMUNICATIONS July/August
1969
Fig.
1.
a) Photograph of 4
x
4 mm test chart, deliberately blurred by motion of the 240 mm Schneider Symmar lens(f/5.6) used
in
the recording of the photograph in white light (zirconium arc). The dimensions of the blurred photo-graph are 10
%
10 mm. The lens was deliberately moved by hand with the aid of a micrometer and ball-slide arrange-ment. The type of motion can be evaluated by looking at the "spread function" h(x,
y)
indicated by the small verticalarrow. It can also be seen by the "doubling" of the numerals (for instance that indicated by the slanted arrow), show-ing that the motion was in fact not linear with time.b) Holographically deblurred photograph, using method of Stroke and Zech
[8]
with "extended-range" filter as de-scribed
in
text. The clearly apparent considerable degree of 'deblurring' may be evaluated by noting that the spreadfunction of fig.
la
has been reduced almost to
a
point (indicated by vertical arrow),
and
also that the "doubled" nume-rals, which had almost equal intensity in fig. la, have been restored to a single numeral. It
is
important to notethat restoration of a
single
image from such ndoubledn motion-blurred photographs, as achieved holographicallyhere, could not possibly be achieved by the well-known purely photographic 'masking' and printing methods, as usedin reproduction, among other applications. It
is
also important to note that nothing more than the spread functionh(x,y) (fig. la) was required to synthesize the image-deblurring deconvolution filter (see fig. 2).
sion filters may in some instances be generated
in
binary form by computer: as shown by Loh-
mann
and Paris [16],
it is
necessary for
this
purpose to have some way of first obtaining ananalytical expression for the required filter.
In
the methods which we discuss here, the requiredFourier-transform division filter is directlygenerated by an analogue optical method [8]starting from the spread function
h(x,
y) (seefigs.
1
and
2).
Before giving the experimental details for theresults shown in figs.
1
and 2, we give the theor-etical principles of our method [8] as it applieswhere
h(x,
y), the spread function,
is
the blurredimage of a single point
in
the object domain.Within the band limits, discussed in ref.
[15],
eq.
(1)
may be written
in
the spatial Fouriertransform domain
[3]
in the form
G=FH.
(2)Within the same limitations, we may extract thefunction
FBL
rom
G
by Fourier-transform divi-sion, in the formto
this
work. Let
g(x,
y) describe -the exposure
E
FBL
=
G/H
G
-1
.
[proportional to the intensity distribution
I(x,
y)]
(3)
in the original blurred photograph. Let f(x, y) beThe symbol
FBL
s
used to indicate the well-the image which would be formed by the givenknown fact that no direct image-restorationoptical instrument in the absence of image-de- method
is
capable of restoring those spatial fre-grading blurring. We recall that
f
(x, y) may be quencies in
F
which have been degraded to zeroconsidered as he "diffraction limitedn image in the original image-recording (i.e. photography)which would be formed by the given instrumentprocess. Nevertheless, extensive experience
if
it
had no aberrations or other blurring imper-has now shown without any doubt that image-fections, notably motion, which we consider restoration methods, such as that which we de-here. The blurred photograph may therefore bescribe, result
in
a considerable visual improve-described by the equation ment of a blurred image (see e.g. fig.
I),
even
Search History:
Result 00 of 00
00 results for result for
p.
More From This User
Notes
Load more
