U.S. patent application number 09/264006 was filed with the patent office on 2002-02-28 for method for aligning an object with an image capture apparatus and a coupling eyepiece.
Invention is credited to GUILLEMAUD, REGIS.
Application Number | 20020024657 09/264006 |
Document ID | / |
Family ID | 9524079 |
Filed Date | 2002-02-28 |
United States Patent
Application |
20020024657 |
Kind Code |
A1 |
GUILLEMAUD, REGIS |
February 28, 2002 |
METHOD FOR ALIGNING AN OBJECT WITH AN IMAGE CAPTURE APPARATUS AND A
COUPLING EYEPIECE
Abstract
The present invention relates to a method for aligning an
object, whose image is to be recorded, with an image capture
apparatus and a coupling eyepiece. Said method consists in using a
light source, said light source being positioned in the place of
the object and emitting luminous dots through a mask and through
the coupling eyepiece towards the capture apparatus. The invention
also consists in verifying whether the images of said luminous dots
are split into two on the image plane, then, using all the
development functions determined for the various alignment
settings, in determining a particular setting in which the
distances between the images of the luminous dots are minimal.
Inventors: |
GUILLEMAUD, REGIS;
(GREMOBLE, FR) |
Correspondence
Address: |
OBLON SPIVAK MCCLELLAND MAIER & NEUSTADT PC
FOURTH FLOOR
1755 JEFFERSON DAVIS HIGHWAY
ARLINGTON
VA
22202
US
|
Family ID: |
9524079 |
Appl. No.: |
09/264006 |
Filed: |
March 8, 1999 |
Current U.S.
Class: |
356/138 |
Current CPC
Class: |
G02B 7/28 20130101 |
Class at
Publication: |
356/138 |
International
Class: |
G06K 009/36; G01B
011/26; G01C 001/00 |
Foreign Application Data
Date |
Code |
Application Number |
Mar 16, 1998 |
FR |
98 03185 |
Claims
1. Method for aligning an object whose image is to be recorded with
an image capture apparatus and a coupling eyepiece, characterized
in that it consists in: defining an object plane (6); emitting at
least three luminous dots (1a, 1b) in the object plane (Po);
positioning an opaque mask perforated with at least two holes in
the optical path; effecting (10-12) a plurality of image captures
of the luminous dots, each capture being made for a different
alignment setting whose parameters are known; establishing for each
luminous dot and each alignment setting a list of distances between
the images of the luminous dots on the image plane defined by the
capture apparatus; determining a development function of the
distance between the luminous dots for each luminous dot and each
alignment setting; and determining a particular setting for which
the distances between the images of the luminous dots are minimal
using all the development functions for the various alignment
settings.
2. Method of claim 1 characterized in that the setting parameters
comprise a translation movement parameter and two rotation
parameters that move in a direction parallel to the optical axis of
the coupling eyepiece.
3. Method of claim 2 characterized in that the setting parameters
comprise a fourth parameter relative to the overall translation
movement of the image plane along the optical axis.
4. Method of claim 1 characterized in that it consists in emitting
three luminous dots and effecting at least four image captures of
said luminous dots for different alignment settings.
5. Method of claim 1 characterized in that, where a large number of
luminous dots are used, pre-selecting the luminous dots to be used
in the subsequent stages.
Description
DESCRIPTION
Field of the invention
[0001] The present invention relates to a method that uses a
coupling eyepiece and an image capture apparatus to align an object
whose image is to be recorded in order to produce an optimal
quality shot.
[0002] The invention may have applications in any sectors concerned
with the creation of optimal quality images, particularly in the
field of astronomy and in the medical sector and more particularly
for X-ray shots.
BACKGROUND ART
[0003] In the field of image capture, specialists in the field
attempt to achieve images of the highest quality possible, in other
words with satisfactory focussing of the capture system and the
best possible image definition.
[0004] In order to obtain an image of excellent quality, the image
plane, i.e. the plane in which the image capture apparatus lies,
must be perfectly aligned with the coupling eyepiece on the one
hand and with the object on the other. However, the alignment is
difficult to effect as the orthogonality of the image plane must be
adjusted according to the optical axis and to the focal
distance.
[0005] Also, in order to create images of an object on an image
plane through a welding eyepiece it is necessary to obtain optimal
positioning between the object, the eyepiece and the image plane.
The optimal positioning can be achieved by means of settings that
are usually made between the positions of the eyepiece and the
capture system (i.e. the image plane). The object is generally
positioned so that it is fixed. There are two kinds of
settings:
[0006] a setting, by means of rotation, of the perpendicularity of
the image plane in relation to the optical axis of the eyepiece;
and
[0007] a setting of the image plane position i.e. the capture
apparatus, in relation to the eyepiece; this is the standard type
of focus setting used in an ordinary camera. The setting is made by
means of a translation movement along the optical axis, and enables
the plane of the capture apparatus or the image plane to be focused
on the object plane.
[0008] Generally, the object is at a variable distance from the
eyepiece. The eyepiece/image plane setting can therefore be made
either approximately or visually by the operator using range
finding or reflex means.
[0009] Some image recording or capture apparatuses are equipped
with a system for measuring the object/eyepiece distance that
measures the distance between the object and the eyepiece using an
ultrasound or infrared method. In this technique the eyepiece/image
plane setting is made automatically.
[0010] However, in these standard apparatuses, the depth of field
at the eyepiece is significant. The eyepiece/image plane setting is
therefore not always accurate.
[0011] Moreover, applications exist in which the eyepiece/image
plane setting is finer, for example in electron microscopy. In
electron microscopy the operator can adjust the eyepiece/image
plane setting visually. The operator can also make the setting
using a technique known as the "wobber focusing aid". The technique
is described on pages 29 to 31 of "The Principles and Practice of
Electron Microscopy" by Ian M. Watt (Cambridge University Press).
This technique consists in oscillating a luminous beam between two
positions in relation to the lens of the image recording apparatus.
This produces a double response of the object observed on the image
plane until the system reaches optimal focusing.
[0012] However, these "wobbling" methods are difficult to implement
as it is not easy to deflect the light source.
[0013] In astronomy, it is also useful to have fine setting,
particularly when a CCD camera is used in combination with an
astronomic eyepiece as described in the article "Une Mission Haute
Rsolution au T60" (A High Definition Assignment at T 60) by J.
Dijon et al., published in Pulsar magazine No. 707, March-April
1995. The article describes a method that enables a relatively
sensitive eyepiece/image plane setting to be achieved. The method
consists in selecting a single star, positioning a two-holed mask
at the entry of the astronomic eyepiece and verifying the number of
patches of light that appear on the CCD image recording apparatus.
If the CCD camera is not correctly focused in relation to the
eyepiece, two patches appear on the camera. If, however, the
focusing is correct only a single patch appears on the CCD image
recording apparatus. By displacing the projection of the star over
different areas of the CCD camera, the alignment of the
eyepiece/CCD camera can be adjusted, in other words it is possible
to adjust the alignment between the eyepiece and the image
plane.
[0014] However, this method can only be used for an object located
at infinity, as is the case in astronomy.
[0015] Other optical methods enable the orthogonality of the image
plane with the optical axis to be adjusted. One method consists in
using an autocollimation eyepiece that is placed on the entry of
the eyepiece. In this example, the image plane must provide a
reflection for this kind of setting. Moreover, difficulties may
arise when other reflective planes are present on the trajectory
between the eyepiece and the image plane, as is the case in a CCD
camera being used through a glass window.
[0016] A method of this kind therefore proves difficult to put into
operation. Also, this method only provides the setting required for
surface evenness, it does not provide the setting for focusing
required to align the system.
DISCLOSURE OF THE INVENTION
[0017] The aim of the invention is to overcome the drawbacks of the
techniques described above. In order to do this, the invention
provides a method for aligning an object, of which an optimal
quality shot is to be taken, with an image capture apparatus and a
coupling eyepiece. This method consists in using a light source
that is positioned in the place of the object and that emits
luminous dots through an opaque mask, perforated with at least two
holes, and through the coupling eyepiece in the direction of the
capture apparatus. The method also consists in verifying whether
the images of the luminous dots are split into two on the image
plane of the capture apparatus.
[0018] More precisely, the invention relates to a method for
aligning an object whose image is to be recorded with an image
capture apparatus and a coupling eyepiece. This method is
characterized in that it consists in:
[0019] defining an object plane
[0020] emitting at least three luminous dots in the object
plane;
[0021] positioning an opaque mask perforated with at least two
holes in the optical path;
[0022] effecting a plurality of image captures of the luminous
dots, each image capture being made for a different alignment
setting, using known parameter settings;
[0023] producing a list of distances between the images of the
luminous dots on the image plane that is defined by the capture
apparatus, said list being produced for each luminous dot and each
alignment setting;
[0024] determining a development function of the distance between
luminous dots for each luminous dot and each alignment setting;
and
[0025] determining a particular setting for which the distances
between the images of the luminous dots are minimal that is based
on all the development functions for the various alignment
settings.
[0026] Advantageously, the parameter settings comprise a
translation movement parameter and two rotation parameters that
move in a direction parallel to the optical axis and the coupling
eyepiece.
[0027] The parameter settings can also include a fourth parameter
that is relative to the overall translation movement of the image
plane along the optical axis.
[0028] In one embodiment of the invention, the method consists in
emitting three luminous dots and effecting at least four image
capture operations of the luminous dots fore four different
alignment settings.
[0029] In another embodiment of the invention, when the number of
luminous dots is greater than 3, the method consists in
pre-selecting those luminous dots that are to be included in the
subsequent stages of the method of the invention.
BRIEF DESCRIPTION OF THE FIGURES
[0030] FIG. 1 is a diagram of the optical trajectory of the
luminous dots from the object plane to the image plane;
[0031] FIG. 2 is a functional diagram of the method of the
invention in a first embodiment; and
[0032] FIG. 3 is a functional diagram of the method of the
invention in a second embodiment.
DETAILED DESCRIPTION OF EMBODIMENTS OF THE INVENTION
[0033] The invention relates to a method for aligning an object
whose image is to be recorded with an image capture apparatus and a
coupling eyepiece.
[0034] Throughout the description that follows, reference will be
made to the "image plane" and the "object plane":
[0035] the image plane, referred to as Pi in the figures, is the
plane in which the image of the object is formed in the image
recording apparatus (or the image capture apparatus);
[0036] the object plane, referred to as Po in the figures, is the
plane in which the object lies at the time the shot is taken. It
should be noted, however, that the method of aligning these planes
is implemented without the object being present; no object is
required to perform the alignment, only the object plane i.e. the
plane in which the object will lie when the shot is taken, is used
to implement the method. During alignment the object plane is
defined either virtually or materially by the support on which the
object will later be placed when the image is recorded.
[0037] It should be noted, moreover, that throughout the setting up
of the alignment method, and during the shot itself, the coupling
eyepiece (more simply called "the eyepiece")focuses on the object
plane.
[0038] According to the method of the invention, the alignment of
the image capture apparatus (in other words the object, the
coupling eyepiece and the capture apparatus) is successively
achieved using several luminous dot images emitted by a light
source through an opaque mask perforated with at least two holes,
such as that described in the patent application filed today by the
present applicant and entitled "System for Determining and
Quantifying the Alignment of an Object with a Coupling Eyepiece and
an Image Recording Apparatus". The image of each of these luminous
dots is either a single dot, in which case the alignment of the
system is considered to be achieved, or a dot split into two, in
which case there is considered to be no alignment of the system.
Where a dot is split into two, the method of the invention consists
in determining the optimal setting of the elements of the system in
relation to each other (i.e. the coupling eyepiece, the object
plane and the image capture apparatus), using the distance that
separates the two dots (or patches) of a dot split into two.
[0039] To enable the method of the invention to be fully
understood, FIG. 1 shows the optical trajectory of two luminous
dots 1a and 1b that are emitted by the light source. The two
luminous dots 1a and 1b are projected as 2a and 2b onto the image
plane Pi, in other words on the image capture apparatus, along an
optical path through coupling eyepiece 3.
[0040] The example in FIG. 1 shows an aligned image capture
apparatus i.e. a system in which the image capture apparatus is
aligned with eyepiece 3 and with the object plane Po shown during
the method to align it with the light source. The alignment is
correct because the beams that are emitted from points 1a and 1b of
the object plane converge towards a single point on image plane 2a
and 2b. In FIG. 1 it may be noted that the incident beams of light
that are parallel to the optical axis of the coupling eyepiece 3,
shown by unbroken lines, all pass through the focal point Fa. It
may also be noted that the incident beams that are oblique and
parallel to each other and shown by dotted lines in FIG. 1, pass
through a different single focal point Fb. Therefore, both focal
points Fa and Fb of focal plane Pf correspond to a direction of
incident beams.
[0041] Luminous dot 1a that passes through focal points Fa and Fb
will now be considered. Providing that the capture apparatus 2 is
correctly positioned in relation to coupling eyepiece 3 and to
object plane Po, in other words providing these three elements are
aligned, the image of luminous dot 1a is then point 2a. The image
of the luminous dot is a single point.
[0042] However, if the capture apparatus is badly focused i.e. if
these three elements are not aligned, the image of luminous dot 1a
then comprises two dots (or patches) on the capture apparatus, in
other words a dot that is split into two. In this situation, the
distance between the two patches of the split dot is in proportion
to the focusing distance of the capture apparatus. This distance,
and at least two other distances that are determined identically
for other settings of the system, are used to determine the optimal
setting of the system and particularly the optimal focusing
position of the capture apparatus.
[0043] In other words, the aim of the method of the invention is to
characterize and calibrate the distance between two patches or
images of the same luminous dot relative to the focusing of the
capture apparatus, said method being used when the capture
apparatus, the coupling eyepiece and the object plane have been
correctly defined.
[0044] If then, the distance between two patches of a double dot is
referred to as Dpoint and the distance between the eyepiece and the
capture apparatus is referred to as Deyepiece, the following
equation can be written:
Dpoint=a*Deyepiece+b,
[0045] this function being cancelled when the optimal focus of the
capture apparatus is determined.
[0046] The alignment method of the invention consists in an initial
stage 6 of defining an object plane by choosing a position where
the object whose image will be recorded is later placed, said
position also being occupied by the source of luminous dots when
the method is implemented.
[0047] The method consists in then emitting luminous dots in the
object plane in the direction of the image plane (stage 8). There
are a minimum of three luminous dots. Advantageously, more than
three luminous dots are used to ensure a high degree of accuracy in
the method and a relatively short processing time.
[0048] The subsequent stages of the method of the invention can be
carried out according to two different embodiments.
[0049] The first embodiment is schematically shown in FIG. 2.
[0050] According to this first embodiment the subsequent stages of
the method are carried out iteratively. In this embodiment the
method initially comprises a stage 10 that provides a choice of
alignment settings. An "alignment setting" comprises focusing of
the capture apparatus on the one hand, and positioning of the
coupling eyepiece and positioning of the capture apparatus on the
other. Positioning of the object plane is defined at the beginning
of the method and remains fixed for the rest of the method and for
the shots of the object.
[0051] Therefore setting of the system is achieved using:
[0052] rotation to adjust the evenness of the capture apparatus in
relation to the coupling eyepiece; and
[0053] a translation movement along the optical axis in order to
adjust the focus.
[0054] The setting is chosen randomly for the first operation.
Subsequent settings are chosen according to information collected
at the end of previous operations (in other words at stage 16)
depending on the quality of alignment obtained.
[0055] The method continues with step 12 that consists in capturing
the image of the luminous dots on the image plane for the alignment
setting chosen at stage 10.
[0056] A stage 14 then consists in analyzing the image of the dots.
During the analysis it is determined whether the image of a single
dot is single or double; if the image is double, the distance
between the two images of the luminous dot i.e. between the two
patches that constitute the double dot, is calculated in order to
establish what setting of the elements of the system must be
applied to improve the alignment of said elements.
[0057] A stage 16 then consists in a quality test used to verify
the quality of the alignment thus obtained during analysis stage
14. If the quality is satisfactory, optimal setting is considered
to be achieved i.e. the setting required to obtain minimal
distances between the images of luminous dots.
[0058] If this is not the case, the method is repeated from stage
10 during which a new alignment setting is chosen depending on the
result of quality test 16.
[0059] FIG. 3 schematically shows the method of the invention
according to a second embodiment. In addition to stage 6 that
defines the object plane and stage 8 that emits luminous dots in
the object plane, the method of the invention comprises a stage 11
for capturing the images of the luminous dots (or luminous patches)
using the capture apparatus. According to this embodiment of the
invention an image of each of these luminous dots is made for each
setting of the system. These image captures are made simultaneously
or successively for various alignment settings.
[0060] The method of the invention then comprises a stage 13 that
consists in calculating the distances between the images of the
luminous dots i.e. between the impact patches of each luminous dot,
when the dot is split into two. The distances may be calculated in
a number of ways; for example, for every luminous dot on the object
plane a sub-image can be defined in the capture apparatus in which
the image of the luminous dot is to be projected. The luminous
patches can then be positioned, for example, by calculating the
center of gravity of each patch.
[0061] Another method of calculation consists in performing a
gaussian estimation on each of the patches that thus enables the
distribution of the levels of gray in the sub-image containing the
images to be modeled satisfactorily. This estimation can be made
using an Expectation Maximisation-type algorithm such as that given
in the article "Parameter Estimation and Tissue Segmentation from
Multispectral MR Images" by Z. Liang et al., IEEE Trans. Med. Im,
vol. 13, No. 3, September 1994.
[0062] The method of the invention may then comprise a stage 15 for
verifying the quality of the dots measured. This stage is not
necessary for the method to be satisfactorily achieved but it does,
however, improve the quality of the result. This stage consists in
verifying that the behavior of the various luminous dots is
coherent. For example, if two or three translation movement
settings have been made in stage 11, stage 15 consists in ensuring
that the behavior is the same for all the luminous dots; therefore
luminous dots that may appear aberrant during this stage are
eliminated and are not used in the subsequent stages.
[0063] The method then consists in a calibration stage 17 for each
of the luminous dots. This stage 17 provides the information needed
to determine the optimal alignment for the system in stage 19.
Calculation of the optimal setting consists in minimizing all the
distances between the images of the luminous dots in relation to
alignment setting parameter of the image plane.
[0064] Stages 17 and 19 can be implemented using mathematical
methods that will be described further on. In particular, stages 17
and 19 each consist in solving opposite systems of equations.
[0065] It should be noted that the image plane of the capture
apparatus is constituted by a plane that is defined by three
setting points. Optimal focusing of the capture apparatus is
achieved when the plane is perpendicular to the to the optical axis
and at the correct focusing distance. Setting of the position of
the plane is therefore made by positioning the three setting points
that define it.
[0066] Therefore, the position of these three setting points
defines how the system must be adjusted. In other words, the three
points set the parameters of the translation movement and the
rotation that are needed to align the system. Preferably, these
three points move in a direction parallel to the optical axis and
the parameters of the points are set by defining their position on
the line along which the system moves. This line is advantageously
parallel to the optical axis of the lens. A point of origin is
randomly defined on each line and remains immobile throughout
alignment.
[0067] For each luminous dot of the light source a distance between
the dots (i.e. the distance between the two patches that are images
of the luminous dots on the capture apparatus) is defined. The
positions of the setting points are referred to as D1, D2 and D3
along the axis of displacement of the system; there exists
therefore a linear relation:
Dpoint=a1*D1+a2*D2+a3*D3+b.
[0068] For each luminous dot a calibration can therefore be
calculated as a function of the distance Dpoint as a function of
the values of D1, D2 and D3. In order for this to be achieved
different image capture apparatuses are used that are intended for
different settings, a distance Dpoint.sub.i being associated with
each setting. The following equation system can therefore be
written:
Dpoint.sub.1=a1*D1.sub.1+a2*D2.sub.1+a3*D3.sub.1+b
Dpoint.sub.2=a1*D1.sub.2+a2*D2.sub.2+a3*D3.sub.2+b
[0069]
Dpoint.sub.n=a1*D1.sub.n+a2*D2.sub.n+a3*D3.sub.n+b,
[0070] in which the unknowns are a1, a2, a3 and b.
[0071] This system can be inverted using known mathematical methods
such as the pseudo-inversion function of the MATHEMATICA.RTM.
software, or using the methods described on pages 52-60 of
"Numerical Recipes, the Art of Scientific Computing", by W. H.
Press et al. (Cambridge University Press). It should, however, be
noted that in order to find a solution to this system it is
necessary for the series of settings to include at least three
independent linear settings.
[0072] After all these points have been calibrated, the following
equations are therefore obtained:
Dpoint.sub.pt1=a1.sub.pt1*D1+a2.sub.pt1*D2+a3.sub.pt1*D3+b.sub.pt1
Dpoint.sub.pt2=a1.sub.pt2*D1+a2.sub.pt2*D2+a3.sub.pt2*D31+b.sub.pt1
[0073]
Dpoint.sub.ptn=a1.sub.ptn*D1+a2.sub.ptn*21+a3.sub.ptn*D31+b.sub.pt1
[0074] In this system the variables are Dpoint, D1, D2 and D3.
[0075] The optimal solution of this system corresponds to a
Distance Dpoint that is equal to 0 for all the luminous dots used
to put implement the method (Dpoint=0: no image splitting of the
luminous dots).
[0076] As seen above, setting of the system is achieved using
rotation in order to adjust the evenness of the capture apparatus
in relation to the coupling eyepiece and using a translation
movement along the coupling eyepiece axis to adjust the focus of
the capture apparatus. However, in this second embodiment of the
invention the parameters for all the rotation and translation
movements are set by the position of the three setting points that
define the image plane. Therefore, an identical movement of the
three setting points enables a translation movement and different
settings to be made according to the three setting points used for
rotation.
[0077] According to another version of the invention, positioning
of the capture apparatus can be defined using a fourth parameter;
this parameter is an overall translation movement of the image
plane along the optical axis that is equivalent to a direct focus
setting of the eyepiece. In this version the fourth parameter is
referred to as D4 and the linear relation is as follows:
Dpoint=a1*D1+a2*D2+a3*D3+a4*D4+b
[0078] The method used to find the optimal setting is equivalent to
that in the embodiment described above that comprises, however,
five shots used for five different alignment settings.
[0079] When a great number of luminous dots are emitted by the
light source the method of the invention can comprise an additional
stage that consists in choosing the luminous dots that are to be
used to implement the method. This stage is implemented immediately
after the luminous dots are emitted during stage 8.
[0080] In order to ensure a higher definition quality of the
system, additional processing may be added during stage 19 used to
calculate the optimal setting:
[0081] in the image captures effected during stage 11 a series of
image captures called "SP" is made for identical setting variations
called "VD" on all the D1, D2 and D3 setting coefficients; this
therefore corresponds to translation movements along the optical
axis;
[0082] in a sub-stage that exists between stage 11 and stage 13,
verification is made to ensure that all the differences in spacing
between the patches are small. For luminous dots with small spacing
differences a mean spacing difference .DELTA.D is calculated for
each setting variation VD i.e. const=mean (.DELTA.D/VD);
[0083] the following requirement can therefore be added to the
definition of the system:
a1+a2+a3=const.
* * * * *