DE19953966A1 - Method and equipment for measuring thickness of transparent layers of blood based on camera images recorded as specimen moves through focussing planes of objective lens - Google Patents

Abstract

A light beam (4) is deflected by a beam splitter (7) through an objective lens (5) on to a vessel containing the specimen (1). A focussing aid (9) focusses an image of itself on a boundary surface of the specimen which is recorded by the camera. The drive unit (14) moves the specimen in the z-direction so that successive boundary layers are traversed. Only at a boundary layer is the camera image sharp so that the distance between two sharp images gives the thickness of the intervening layer

Description

The invention relates to a method and a device for thickness measurement on transparent layers with the features of the independent claims 1 and 5.

The invention is used to measure the thickness of transparent, partially reflecting Layers. In particular, it is said to be used in blood analysis under another on the quantitative evaluation of fluorescence intensities one Blood layer based. The essential prerequisite for the evaluation is the precise knowledge of the thickness of the blood layer.

The blood is in a special disposable Microcuvette made of glass or plastic or a glass-plastic Combination. Manufacturing tolerances of the dimensions of the disposable Microcuvettes make it necessary to adjust the thickness of the blood layer to measure different places of the already filled disposable micro cuvette.

It would be conceivable to control the thickness tolerances of the disposable Micro cuvette, for example, with a mechanical button. The thickness of the blood layer is determined by the internal dimensions of the disposable  Micro cuvette determined, which may have other tolerance deviations than the outside dimensions. However, since the button only the outer dimensions probing solutions, it is for determining the layer thickness of the blood layer not suitable.

Another method to control the thickness tolerances of the disposable Microcuvette consists of an interferometric measurement of the blood layer. For this purpose, for example, an interferometer arrangement in an incident light Microscope used, its illuminating light from a spectrometer is brought out. The filled single-use micro cuvette is placed in the Interferometer beam path placed on the microscope stage. The microscope's illuminating light is switched on at all interfaces of the Microcuvette reflects and interferes. The measurement is for different Wavelengths of illuminating light that the spectrometer delivers carried out. From the wavelength-dependent interferences known refractive index determine the thickness of the individual layers. It however, proves to be very difficult when evaluating the measurement portions of light reflected from the various interfaces separate. Another disadvantage is that a spectrometer is required which is expensive and takes up a lot of space.

It is therefore an object of the invention to provide a method and an apparatus for Specify thickness measurement on transparent layers. You should in particular for transparent, liquid layers on a support or in a closed vessel, such as. B. a micro cuvette, are suitable.

This object is achieved by the features of the independent claims. Advantageous further developments are described by the subclaims.

According to the invention, in an illumination beam path of a lens a focusing aid arranged in a position conjugate to the focus plane. The Focus plane is imaged on a camera. The focusing aid will only then sharply imaged on the camera if a partially reflective Is the interface of an object in the focal plane. The one with the camera  controlled sharpness of the image of the focusing aid thus serves as "Focus indicator". This makes the invention for determining the Use layer thicknesses.

The illumination beam path is directed through the lens onto an object directed, which has a transparent layer. As an object For example, a single-use micro cuvette used for blood analysis with a transparent layer of blood is filled. Alternatively, one could Slides can be used that carry a transparent layer of blood.

By moving the object in the z direction relative to the lens, the focal plane of the lens is moved step by step through the object. The values of the positions z _i are measured. For each position approached, a camera image is recorded and its focus sharpness value is determined, the image of the structure of the focusing aid being used as a sharpness indicator.

For optimal image evaluation, the structure on the is advantageous Focusing aid at least one bar, the bar length of which is a multiple of Bar width is. The bar width must be a multiple of that Resolution of the lens and the camera. It can multiple bars can also be used. The use of a is advantageous Cross with the dimensions listed for the bar for each Ends of the cross. The Keuz can have different pixel widths and heights must be adjusted.

A small focus sharpness value arises from a blurring of the image large focus sharpness value due to a large sharpness of the image. This means: If the image of the structure of the focusing aid appears sharp in the camera image, is an interface of the single-use micro cuvette in the focal plane. Doing the focus value is a relative maximum.

The maxima of the focus values are therefore determined. The positions z _i with maximum focus sharpness values are assigned to the position of the various interfaces of the object. Interfaces are, for example, the air / glass transitions, the glass / liquid transitions, the liquid / glass transitions and the glass / air transitions in a liquid-filled microcuvette.

The different interfaces differ significantly with regard to the Reflectivity. Glass / air interfaces reflect e.g. B. by about one Factor 10 stronger than glass / liquid interfaces. By determining the focus sharpness values the intensities of the reflexes in the calculation from the size of the focus sharpness values determined on the Close the type of interface and make an assignment.

The thickness of the transparent layer, e.g. B. the liquid in the microcuvette, is then calculated from the difference of the positions z _i , which are assigned to their interfaces z ₁ , z ₂ . The thickness d is given by
d = (z ₁ - z ₂ ) .n _layer ,
where n _{layer is} the refractive index of the layer. The optical path length changed by the refractive index of the layer is taken into account by multiplying by n _layer .

When assigning the maxima of the focus sharpness values to the interfaces of the object with the layer to be measured, certain preliminary information about the type and / or the approximate position of the interfaces or the layer is received. This results in further options for the assignment. The following variants are possible, for example:

• - The travel path for the object is chosen to be sufficiently large so that the first interface is detected in any case and thus the first maximum the focus sharpness values can be assigned to the first interface. This is particularly advantageous for an object in a sandwich structure (i.e. with several layers of material) such as a microcuvette.
• - The interfaces of particular interest, for example the inner ones The walls of a microcuvette are in terms of their expected position predefined sufficiently precisely. The tolerance information of the  Manufacturer about the dimensions of the micro cuvette. If due this information, the outer interfaces are sufficiently large The distance to the inner interfaces of the microcuvette is one direct assignment of the maxima of the focus sharpness values to the inner ones Interfaces possible. Then the travel path of the object and the number of camera images can be reduced. This has the advantage that the measuring times are considerably reduced. The method is therefore suitable especially for routine laboratory examinations.

The method according to the invention can be used reliably for measuring the thickness of transparent layers of all kinds. The transparent layers can be:

• - Self-supporting solid layers in air or liquid - or around solid or liquid layers on a substrate (e.g. around medical smears on a glass slide)
• - or around solid or liquid layers on a substrate with a additional transparent layer covering (sandwich structure).

The layer, the substrate and any layer covers must be smooth Have surfaces. The interfaces of the layers only have to be partially reflective, d. H. they can have very small refractive index differences exhibit. The low reflections obtained from these interfaces are for an automatic image evaluation by means of image analysis, which the The method according to the invention already describes sufficient.

The method according to the invention and the device according to the invention are described below with the help of exemplary embodiments schematic drawings explained.

Show it:

. Figure 1 shows a device for layer thickness measurement;

Fig. 2 is detailed view of a microcuvette with schematic light intensities at the interfaces;

Figure 3 is a series of measurements of camera images with different image sharpness as a function of the z position.

Fig. 4 is diagram of a measurement series of focus sharpness values as a function of z-position;

Fig. 5 diagram of an analytic function to a measuring range of focus-focus values as a function of the z position.

Fig. 1 shows an apparatus for performing the method according to the invention. An object 1 is examined which has a transparent layer 2 , the thickness of which is to be determined. An illumination beam path 4 emanates from a light source 3 and is directed onto the object 1 through an objective 5 . The optical axis 6 of the illumination beam path 4 is oriented perpendicular to the layer 2 . The deflection of the illumination beam path 4 in the direction of the object 1 takes place on a beam splitter 7 . A focusing aid 9 provided with structures is arranged in the illumination beam path 4 in a position conjugate to the focus plane 8 of the objective 5 .

In the illustration, the focal plane 8 lies at the upper interface of a transparent layer 2 . This transparent layer 2 is enclosed in the object 1 . For this purpose, the object 1 consists of a base plate 10 which has a cavity in which the transparent layer 2 is located. On the base plate 10 there is a transparent cover 11 , which seals the cavity with the transparent layer 2 .

An imaging beam path 12 extends from the upper boundary surface of the transparent layer 2 , that is to say from the focal plane 8 . It passes through the beam splitter 7 and is aimed at a camera 13 which is in a position conjugate to the focal plane 8 .

Only when the focus plane 8 , as shown in the illustration, lies in an interface of the object 1 , is the structure of the focusing aid 9 with the imaging beam path 12 sharply imaged on the camera 13 . If the focus plane 8 lies outside an interface of the object 1 , the structure of the focusing aid 9 appears blurred in the camera image. The invention makes use of this.

The object 1 is connected to a drive unit 14 with which the object 1 can be moved in the z direction relative to the objective 5 . When moving the object 1 , the focal plane 8 can be passed through all the interfaces of the object 1 one after the other. A sharp image is generated on camera 13 only when passing through an interface. A desired layer thickness can then be determined from the difference between the z positions assigned to certain interfaces. For this purpose, the drive unit 14 is connected to a z-position measuring unit 15 . This measures the current position z of object 1 .

The camera 13 is connected to an image recording unit 16 , which in turn is connected to an image processing unit 17 and a computing unit 18 . The drive unit 14 , the z-position measuring unit 15 , the image recording unit 16 , the image processing unit 17 and the computing unit 18 are connected to a control unit 19 , with which the sequence of the method steps is controlled.

The sequence of the method according to the invention is described below with reference to FIG. 1.

The object 1 is illuminated with the illumination beam path 4 guided through the objective 5 . The object 1 is in an initial position, in which in this example the focal plane 8 lies above the top boundary layer of the transparent cover 11 . The image of the camera 13 is recorded by the image recording unit 16 . Since the focus plane 8 is not in a partially reflecting interface of the object 1 , the focusing aid 9 is imaged out of focus on the camera 13 . The image processing unit 17 determines a focus sharpness value f (z _i ) for the recorded camera image. z _i indicates the current position of the object 1 , which is determined with the z position measuring unit 15 .

This recording of a camera image and determination of an assigned focus sharpness value f (z _i ) is now carried out for different positions z _{i of} the object 1 relative to the lens 5 .

For this purpose, the object 1 is moved step by step in the direction of the objective 5 , taking discrete positions z _i one after the other. In each position z _i approached, a camera image is recorded with the help of the camera 13 and the image recording unit 16 . At the same time, in each position z _i approached, the associated measured value of the position assumed by object 1 is determined. The object 1 is moved step by step relative to the objective 5 until the focal plane 8 has passed through at least the lower boundary surface of the transparent layer 2 .

For each camera image recorded in the individual approached positions z _i , a focus sharpness value f (z _i ) is determined with the image processing unit 17 . The focus sharpness values f (z _i ) thus obtained are evaluated with the computing unit 18 . For this purpose, the different maxima of the focus sharpness values f (z _i ) are determined. The determined maxima are assigned to the interfaces of object 1 . A value of the associated z _i position can now be assigned to each maximum.

The thickness of the transparent layer 2 is determined from the difference between the z _i positions of the maxima, which are assigned to the interfaces of the transparent layer 2 . The layer thickness d is given by
d = (z ₁ - z ₂ ) .n _layer ,
where n _{layer is} the refractive index of the layer.

Fig. 2 shows an enlarged detail of the object 1. For example, it is a section through a liquid-filled microcuvette. The base plate 10 is again shown with a cavity which contains a transparent layer 2 , in this example a liquid. The base plate 10 is covered with a transparent cover 11 so that its cavity is sealed.

The illustration contains arrows which, due to their size and line thickness, schematically reproduce the reflected light components with respect to their direction and intensity at the individual interfaces of the object 1 . The reflected light components are symbolized by the upward-pointing arrows and the illuminating light by the strongly drawn upward-pointing arrow. It is clearly shown that the strongest reflections take place at the glass / air interfaces, that is to say on the top and on the bottom of the object 1 . The reflected light components are significantly smaller at the upper and lower interfaces of the transparent liquid layer to be measured. It becomes clear that the different interfaces differ significantly in terms of their reflectivity. Glass / air transitions reflect about 10 times more strongly than glass / liquid transitions. By also evaluating the intensity of the camera image for each recorded camera image when determining the focus sharpness value, the position of the interfaces can be inferred from the size of the focus sharpness values.

For this purpose, the difference in intensity between two adjacent pixels is preferably considered when calculating the focus sharpness value. This difference must be formed for the pixels of the rows and columns of the camera picture. A suitable function for calculating the focus values is, for example:

N indicates the number of pixels per row and m the number of pixels per column. The intensities of two neighboring pixels n, m are given with I _n , I _m . With this function, the power x <1 for the intensity differences | I _n - I _m | causes steep gradients to be considered more than flat gradients. That is, the function is non-linear.

The image evaluation of the camera images is to be clarified using FIG. 3. Fig. 3 shows a series of measurements of camera images that have been recorded for different positions of the object 1 z _i. It can be seen that a cross was used as the structure of the focusing aid 9 .

The cross is shown with different sharpness and different brightness in the individual camera images, i.e. with different light intensity. Since the focusing aid 9 is only imaged sharply on the camera when the focal plane 8 is in a boundary layer of the object 1 , the intensity and the sharpness of the cross in the camera image are the focus criteria. The cross must therefore be significantly larger in relation to the pixel widths and heights of the camera.

In the first camera image (top left), the cross appears extremely out of focus, that is, the focal plane 8 is located far away from an interface of the object 1 . With the camera images 2 , 3 and 4 (following to the right) the image sharpness of the cross gradually improves. This means that with the positions z _i assumed in these images, the focal plane 8 was gradually brought up to an interface of the object 1 . The fifth camera image (second row on the left) shows the greatest image sharpness for the cross of the focusing aid 9 . The cross becomes blurred again for all other camera images. It follows from this that the camera image 5 was recorded in a z _i position in which a boundary layer of the object 1 was in the focal plane 8 .

In order to avoid this visual assessment of the image sharpness, a focus sharpness value f (z _i ) for each camera image is determined for each camera image using the specified function f (z _i ). The largest focus sharpness value f (z _i ) is calculated for the camera image 5 , for the other camera images correspondingly lower values for f (z _i ) according to their sharpness and light intensity.

FIG. 4 shows the diagram of a series of measurements of calculated focus values as a function of the positions z _{i of} the object 1 . The focus values shown assume maximum values for two areas of positions z _i . In the left maximum, a focus value is obviously the maximum. At the right maximum, the determined focus values move around the actual maximum. This means that the position of the maximum image sharpness was not approached. With a sufficiently small step size when moving the object 1 with respect to the objective 5 , the determination of the actual maxima of the focus sharpness values can therefore be significantly improved. The layer thickness of the layer to be examined is then calculated from the difference between the two maxima. The maxima shown in Fig. 4 are about 30 microns apart. The two maxima shown here were clearly assigned to the boundary layers of the transparent layer 2 to be measured in one measurement, so that the thickness of the transparent layer 2 is thus determined.

So far it has been described that in order to improve the resolution when determining the maxima of the focus sharpness values, the largest possible number of measurements should be carried out with small distances between the individual positions z _i . However, this has the disadvantage that on the one hand the measurement takes a very long time because of the large number of individual measurements, and on the other hand very large amounts of data have to be processed.

In an advantageous embodiment, the method is therefore carried out with only relatively few positions z _i . An analytical function is then created from the few focus sharpness values f (z _i ) obtained.

An example of such an analytical function is shown in FIG. 5. In the left part of the curve, a strong maximum of the focus-sharpness function can be seen, in the right part of the curve two small maxima at a short distance from one another. As has already been explained with reference to FIG. 2, the strongest reflections and thus also the largest focus sharpness values are generated by the air / glass transitions of the object 1 . The two small maxima of the focus-sharpness curve therefore correspond to the glass / liquid transitions of the upper and lower interfaces of the transparent liquid layer 2 in object 1 . From the analytical function shown here, the two smaller maxima can therefore be determined with high accuracy, their associated positions z _i calculated and the layer thickness of the transparent liquid layer 2 determined from the difference between the two assigned positions z _i with high accuracy. This method works much faster because of the smaller number of camera images and the shorter computing time.

An additional reduction in the computing effort is achieved by calculating an approximating function F (z) only in the vicinity of the maximum focus sharpness values F _n (z _i ). To calculate an approximating function F (z), only those focus sharpness values F _n (z _i ) are used whose neighboring focus sharpness values F _nm (z _i ) and F _{n + m} (z _i ) have smaller values, whereby with m ≧ 1 the size of the environment is defined. With this method, a different approximating function F (z) can be calculated for each local maximum of the focus sharpness values F (z _i ). Since only a few focus sharpness values F (z _i ) are required for approximation, the computational effort is significantly reduced.

The present invention is described with reference to exemplary embodiments been. However, it is for anyone skilled in the art obvious that changes and modifications are made can, without sacrificing the scope of the following claims leave.  

Reference list

1

object

2nd

transparent layer

3rd

Light source

4th

Illumination beam path

5

lens

6

optical axis

7

Beam splitter

8th

Focus plane

9

Focusing aid

10th

Base plate

11

transparent cover

12th

Imaging beam path

13

camera

14

Drive unit

15

z-position measuring unit

16

Image acquisition unit

17th

Image processing unit

18th

Arithmetic unit

19th

Control unit

Claims (9)

1. Method for thickness measurement on transparent layers, characterized by the steps:

• a) illuminating an object ( 1 ) having a transparent layer ( 2 ) with an illumination beam path ( 4 ) guided through an objective ( 5 ) with an optical axis ( 6 ) oriented perpendicularly to the transparent layer ( 2 ), a structured focusing aid ( 9 ) are arranged in the illumination beam path ( 4 ) and a camera ( 13 ) in an imaging beam path ( 12 ) in positions conjugated to the focal plane ( 8 ) of the objective ( 5 ),
• b) stepwise movement of the object ( 1 ) relative to the objective ( 5 ) in the z direction at discrete positions z _i ,
• c) taking a camera image in each approached position z _i and measuring the associated z _i value,
• d) determining a focus sharpness value F (z _i ) assigned to the respective position z _i from each recorded camera image,
• e) determining the maxima of the focus sharpness values F (z _i ),
• f) assigning the maxima to the interfaces of the object ( 1 ),
• g) and determining the thickness of the transparent layer ( 2 ) enclosed in the object ( 1 ) from the difference between the two z _i positions z ₁ , z _{2 of} the maxima assigned to its interfaces, the thickness d being given by the transparent layer ( 2 ) is by d = (z, - z ₂ ) .n _layer , n _layer = refractive index of the layer.

2. The method according to claim 1, characterized by the further steps:

• a) determining an approximating function F (z) from the discrete focus sharpness values F (z _i ),
• b) determining the analytical maxima of the function F (z),
• c) and determining the thickness of the transparent layer () from the analytical maxima of the function F (z).

3. The method according to claim 1, characterized by determining an approximating function F (z) in the vicinity of those discrete focus sharpness values F _n (z _i ) whose neighboring focus sharpness values F _nm (Z _i ) and F _{n + m} (Z _i ) have smaller values, whereby with
m ≧ 1 the environment is defined. 4. The method according to claim 2 or 3, characterized in that the focus sharpness value F (z _i ) from the difference between the intensities I _n and I _{m of} adjacent pixels n, m of the camera image recorded in the position z _{i is} calculated by
where g (| I _n - I _m |) is nonlinear. 5. Device for carrying out the method for thickness measurement on transparent layers according to claim 1, characterized by:

• a) an object ( 1 ) which has a transparent layer ( 2 ),
• b) a light source ( 3 ),
• c) a lens ( 5 ) through which an illumination beam path ( 4 ) emanating from the light source ( 3 ) is directed onto the object ( 1 ) with an optical axis ( 6 ) oriented perpendicular to the transparent layer ( 2 ),
• d) a structured focusing aid ( 9 ) arranged in the illumination beam path ( 4 ) in a position conjugate to the focal plane ( 8 ) of the objective ( 5 ),
• e) a camera ( 13 ) arranged in the imaging beam path ( 12 ) in a position conjugate to the focal plane ( 8 ) of the objective ( 5 ),
• f) a drive unit ( 14 ) for moving the object ( 1 ) relative to the objective ( 5 ) in the z direction at discrete positions z _i ,
• g) a z-position measuring unit ( 15 ) for measuring the z _i values of the approached z _i positions,
• h) an image recording unit ( 16 ) connected to the camera ( 13 ),
• i) an image processing unit ( 17 ) for determining a focus sharpness value (Focus-Score) F (z _i ) assigned to the respective position z _i from each recorded camera image,
• j) a computing unit ( 18 ) for evaluating the focus sharpness values F (z _i ) and determining the thickness of the transparent layer ( 2 ),
• k) and a control unit ( 19 ) for sequence control of the method, which is connected to the other units ( 14 , 15 , 16 , 17 , 18 ).

6. The device according to claim 5, characterized in that the object ( 1 ) is a slide which carries a transparent liquid layer. 7. The device according to claim 5, characterized in that the object ( 1 ) is a micro-cuvette which contains a transparent liquid. 8. The device according to claim 5, characterized in that the structure on the focusing aid ( 9 ) is at least one bar whose bar length is a multiple of the bar width, the bar width being a multiple of the resolution of the lens and the camera. 9. The device according to claim 8, characterized in that the structure on the focusing aid ( 9 ) is at least a cross of two bars.