DE102006051868B4 - Three-dimensional single-shot holography - Google Patents

Abstract

Holographic method for high-resolution three-dimensional imaging of an object (5), in which the object (5) is illuminated with at least one illumination light source (1) and the transmitted and / or reflected light (6) for storing holograms in a holographic storage medium (7) is superimposed on a reference beam (8) of a reference light source on the storage medium (7), wherein the reading of the stored holograms with a read-out light source, characterized in that the illumination, reference and read light source spectrally broadband light sources are that as a storage medium (7 ) an optically thick holographic storage medium is used and that all three-dimensional information of the object (5) by a single shot defined time duration of the illumination and reference light source (1) in the holographic storage medium (7) are stored, the stored information for reconstruction a s three-dimensional image of the object (5) are read out optically and the reading is carried out using at least one transformation and wherein the depth information on the internal structure of the object (5) from the spectrally resolved intensity distribution of the stored in the storage medium interference pattern is reconstructed.

Description

The invention relates to a holography method for high-resolution three-dimensional imaging of an object according to the preamble of claim 1. The field of application of the invention is the high-resolution three-dimensional imaging in or by non-scattering and in particular scattering materials.

The acquisition of three-dimensional images in a short time is becoming increasingly important in medical research and diagnostics as well as in metrology in general. Particularly in biomedical diagnostics, there are extremely high demands with regard to a high resolution in the micrometer range with preferably short image acquisition times, whereby these are as far as possible in the millisecond range.

Especially in the medical field, a fast data generation, resulting in a short recording time, is of great importance, because it is in vivo imaging, so living people. On the one hand, long recording times entail the risk of shaky pictures - the pulse rate of the patient is decisive for living objects - on the other hand, the patient is expected to be more uncomfortable, which reduces the acceptance of the method.

In addition to the classical methods such as X-ray technology, magnetic resonance imaging and ultrasound optical methods for high-resolution three-dimensional imaging of an object, with which good results, especially in terms of resolution can be achieved and without health endangering ionizing radiation, for example, the X-rays, get along. However, optical imaging in biological tissue is made extremely difficult by strong light scattering. For biomedical imaging, various methods are currently used that do not meet these high requirements and are briefly outlined below.

From the US Pat. No. 3,013,467 A is known as the so-called confocal microscopy. With this special microscopy method, high-resolution images of three-dimensional objects can be generated, whereby a raster method, ie a spatial scanning, is performed in all three dimensions.

Further, from the scientific paper, "Volume holographic hyperspectral imaging", W. Liu, G. Barbastathis, D. Psaltis :, Applied Optics 43, pp. 3581-3599, 2004, and US Patent Application No. US 2004/0 021 871 A1 discloses a method which uses volume holograms as special filters and thus provides one or more two-dimensional projections from a four-dimensional space consisting of the three spatial dimensions and the spectrum.

Confocal microscopy and volume hologram filtering are unsuitable for scattering media. In confocal microscopy, the image information is massively disturbed in strongly scattering media by the strong scattered light background, and it is also a time-consuming screening method. Although the method of filtering by means of volume holograms can rapidly generate images of three-dimensional objects without rastering, it is practically limited to the representation of a small number of pixels since, in principle, for each individual pixel a volume hologram, for example via a CCD chip or a camera, is provided separately must be read so that in practice the temporally parallel mapping of different pixels is geometrically limited to a number of pixels, which is substantially less than z. B. in the three-dimensional representation of a multi-cubic millimeter tissue sample with micrometer resolution would be required. On the other hand, the method of filtering by means of volume holograms in strongly scattering media does not work because due to the multiple scattering of the hologram diffracted photons can no longer be clearly assigned to a specific, acting as a point source point in space. The volume hologram serves as a filter by only diffracting light when the Bragg condition is met. However, this is true for any light having the correct wavelength and wavefront characteristic and is not dependent on the optical coherence property of the light. Therefore, it does not provide the necessary filtering function with respect to stray light.

In contrast, concepts are known in which it is attempted to suppress the strong light scattering and thus high-resolution three-dimensional imaging in scattering materials such. B. to improve biological tissue. The most successful concepts are interferometric or holographic methods with light sources of short coherence lengths, which will be briefly discussed below. However, as will be seen in the presentation below, all of these methods have in common that they do not operate without resorting to the temporal dimension when recording the three spatial object dimensions. By this is meant that the inclusion of the three local dimensions requires operations that must be sequential in time, or in other words temporally serial or sequential. It should already be noted at this point that hereinafter by the terms "simultaneously" or "simultaneously" or "temporally parallel" an operation is understood that does not consist of further successive operations. It is called as a simultaneous recording of all three object dimensions understood a shot in a single shot, ie, in the literal sense simultaneous recording of all three dimensions.

From Optical Coherence Tomography, Science 254, pp. 1178-1181, Huang et al., 1991, the so-called Optical Coherence Tomography (OCT) is known. It is a powerful process that provides three-dimensional imaging of highly diffuse-scattering objects using optical, non-ionizing radiation. The imaging in biological tissues is not limited by the light attenuation due to the strong scattering but by the fact that the image information is destroyed by a strong scattered light background. The central achievement of OCT, also called standard or time-domain OCT (TD-OCT), and the interferometric and holographic techniques based thereon, is to mask out the scattered light background and to detect only the light portions containing image information. These light components, which are not scattered many times, but are directly transmitted or reflected, are called ballistic photons. This suppression of the multiple scattered light components is achieved with so-called coherence filtering, since the multiple scattered light components lose their coherence with the light source. For this purpose, the object to be examined is integrated into an interferometer and equipped with short-coherent light, i. H. spectrally broadband light illuminated, for example, from a light emitting diode.

The object represents the first interferometer arm, the second forms a movable reference mirror. Only the light components returning from the object, which have traveled an optical path length which corresponds within the coherence length to that in the reference arm, can interfere with this reference light and deliver an interference signal at the output of the interferometer. Since the reference arm is movable, different optical path lengths can be set and thus different depths in the object can be addressed. The setting accuracy for this depth addressing is determined by the coherence length. The smaller the coherence length, the better the depth addressing or depth resolution.

If the reference arm is moved continuously, an interferogram results which, for the examined point, provides a depth-resolved structural information, the so-called z-scan of the object. A completely three-dimensional image is created when the object in the plane is scanned point by point perpendicular to the incidence of light and a z-scan is taken at each location. This method provides excellent high resolution images of biological objects and is e.g. B. in ophthalmology or dermatology used clinically. The main disadvantage of the OCT, however, is that each individual pixel of the object must be mechanically addressed. This is extremely time consuming and requires a high mechanical susceptibility to failure and susceptibility to interference from external influences, such as a temperature dependence.

From the European patent EP 0 626 079 B1 and the scientific paper "Optics Comm. 122, pp. 111-116, 1996, SCW Hyde et al., The so-called time-resolved holography is known, which realizes a temporally sequential depth addressing as in TD-OCT and also provides coherent filtering for scattered light suppression. With the help of temporally short light pulses, a depth in the object is sequentially temporally addressed by a movable mirror in the reference arm. The interference pattern resulting from the superposition of the light coming from the object and the light from the reference arm is recorded in a holographic medium placed at this location of the superimposition. By holographic method can then be reconstructed from this two-dimensional interference pattern after recording again the locally laterally resolved original structure of a certain depth. With this method, it is thus possible to simultaneously record lateral surfaces at a specific axial depth at the same time. After each recording, however, the holographic medium must first be read out and erased before another depth is addressed.

Also in this method, temporally successive recordings of the object to be examined are recorded in individual planes perpendicular to the light incident direction, which is extremely time-consuming. Through a correspondingly fast-working holographic medium (write and erase times), so-called real-time imaging can be achieved with low depths and low depth resolution, as described in the European patent EP 0 626 079 B1 is mentioned. Nevertheless, it should be noted that the method relies on a time-serial recording of the individual depth layers. Thus, the so-called real-time imaging with appropriate depth of the object and depth resolution finds its limits.

As in the latter method, the need for temporal serial recording per se is also given in the aforementioned TD-OCT method: a simultaneous generation of the information of all three object dimensions is not possible, since there is always a possibility, a respective depth of the object due to runtime differences and with short coherent light to select.

From Optical Coherence Tomography, Journal of Biomed. Optics 1, pp. 157-173, AF Fercher, 1996, is known as the so-called spectral radar, which is counted among the OCT methods but achieves depth addressing in a different way than the TD-OCT. In this case, the depth information is obtained by registering the spectrally resolved intensity distribution of the backscattered light with a spectrometer for a point of the object. The depth structure, in particular the variation of the refractive index contrast over the depth, results from this spectral data by a Fourier inverse transformation of the registered values of the backscattered light. This method belongs to the class of the so-called Fourier domain OCT or short FD-OCT. A moving reference mirror is thus no longer necessary, however, a lateral scanning of the sample, ie in the plane perpendicular to the light, is still necessary.

An alternative method of the FD-OCT class is the so-called swept-source OCT, which can be found, for example, in "Optical coherence tomography using a frequency-tunable optical source", Optics Letters 22, pp. 340-342, 1997, SR Chinn, EA Swanson, JG Fujimoto , is known. In this case, the spectrally resolved information is not realized via a spectral resolution of the signal reflected by the object but by means of a fast spectral tuning of the light source, which supplies data equivalent to the spectral radar in combination with a spectrally integrating detector. Both concepts, spectral radar and swept-source OCT, reduce the effort for a z-scan. In both methods, this is done for one point at a time, so that they represent time-consuming raster methods.

Accordingly, a decisive disadvantage of the mentioned known OCT method is the need to approach each point individually in the lateral direction, that is to perform a 2-dimensional scanning of the surface. Simultaneous, parallel recording of several lateral points is possible by operating many detectors in combination (eg as a CCD array), but at great expense and high costs. The following examples illustrate some ideas of implementation.

From "Nonmechanical grating-generated scanning coherence microscopy" Optics Letters 23, pp. 1797-1799, 1998, I. Zeylikovich, A. Gilerson, RR Alfano discloses a method in which a two-dimensional CCD array is used for detection whose first dimension the spectrum is plotted and on the second dimension of a local lateral dimension of the object. However, a simultaneous recording of the data for all 3 dimensions is not possible.

From the publication "Parallel Image Acquisition in Frequency Domain OCT", Proc. SPIE-OSA Biomedical Optics, SPIE Vol. 5861, 586112-1-586112-3, 2005, B. Povazay et al., Discloses a method that can also accommodate data for three dimensions. Here, the detector used is not a single photodiode, but a two-dimensional detector array, in particular a CCD camera. The two lateral dimensions of the detector are used to register the two lateral dimensions of the object. The depth dimension of the object, which is coded in the spectrum, is placed in the time dimension by taking the intensity distribution of the different wavelengths in succession.

The scientific contributions "Wavelength-scanning digital interference holography for optical section imaging", Optics Letters 24, 1693-1695, 1999 and "Tomographic three-dimensional imaging of a biological specimen using wavelength-scanning digital interference holography", Optics Express 7, 305-310 , 2000, teach a similar but slightly modified process that could be called swept-source digital holography. With the aid of a CCD camera, digital holograms of an object with different wavelengths are successively recorded one after the other and then combined in a computer-aided manner in retrospect in order to obtain the depth information.

Common to the methods is that they generate the information of the depth dimension from the spectral intensity distribution. In order to determine the intensity spectrally resolved, one dimension is required, either local or temporal. The last two methods use the temporal dimension. Such methods come quite close to a so-called "no-motion" OCT. However, recordings for different spectral positions of the light source must be carried out in chronological succession and the tuning of the light source may still involve mechanical movement.

In the "spectral radar" method, the temporal aspect is indeed shifted out of the light source, since the spectrum carrying the depth information is in principle simultaneously available. For its registration, however, another dimension (spatial or temporal) is inevitable. Either the spectrum is placed in a spatial dimension during registration, but then lacks a recording dimension for the lateral object dimensions, so it must be rasterized for this purpose, or the registration of the spectrum is placed in the temporal dimension. This corresponds to the methods swept-source OCT or swept-source digital holography, in which the light source is spectrally driven through, so that the temporal component in the sequential generation of the laser radiation to be used.

The above methods of FD-OCT thus illustrate a problem of a fundamental nature. Although the depth information is coded in the spectrum and thus can be made available in principle simultaneously, the spectral detection requires the use of a local dimension (eg a dimension of a CCD array) or the temporal dimension (serial detection of different wavelengths one after the other ). Since there is no truly three-dimensional detector but only a maximum of two-dimensional (eg, two-dimensional photodiode array, two-dimensional CCD array), the three spatial dimensions of the object to be imaged can be in the maximum two spatial dimensions of the detector and once in time Dimension can be mapped.

Although splitting up or clustering the two-dimensional detector array into quasi an array of many two-dimensional detector arrays is possible in principle. However, this has the disadvantage that the possible lateral resolution of the recording deteriorates in accordance with the then reduced resolution of the detector array.

Incidentally, from the publication Saxba, Graham, Practical Holography, Prentice Hall International 1988, Chapters 5 and 18, a hologram recording method is known in which an object is stored in an optical storage medium having a spectrally broadband light source in the form of an argon or krypton ion laser in an exposure step becomes. From the publication Kleen, Müller, Springer Verlag, 1969, Chapter 6.5.2, it is known that the light sources mentioned can operate in pulsed mode. Finally, from Hariharan, Cambridge University Press, 1984, chapters 9.3 and 9.4, the recording of multicolor holograms using optically thick recording media is known. The latter publication describes a lighting device with at least one illumination light source, at least one reference light source and a holographic storage medium, wherein the object can be illuminated with the beam of the illumination light source and the beam of the illumination light source and reference light source is guided in such a way that the object transmitted and / or reflected Light with the beam of the reference light source in the storage medium superimposed, the illumination light source and reference light source are spectrally broadband and the storage medium is an optically thick holographic storage medium, in which the belonging to the different spectral components of the light sources holograms with a single shot of the illumination and reference light source defined Duration are writable.

The assessment of the prior art described above shows that none of the described optical methods is suitable for simultaneously or simultaneously recording the three spatial dimensions in or by scattering media in an exposure, which is understood below as a so-called single-shot recording. However, this is an increasing need, especially for medical use.

It is therefore an object of the present invention to realize a high-resolution three-dimensional imaging method and to provide a corresponding device which overcomes the disadvantages of the prior art methods and devices and with a single-shot recording, ie temporally parallel, the receives complete three-dimensional information, in particular, a three-dimensional single-shot image acquisition is realized in or by diffusely scattering materials.

The object is achieved by means of a holography method according to claims 1 to 15 and a corresponding device according to claims 16 to 28, in which or in which in an optically thick storage medium, the entire three-dimensional image information of the object with a single shot of a spectrally broadband illumination - And reference light source defined duration is buffered by superposition of information, the stored information is optically readable and the readout to reconstruct a three-dimensional image of the object using at least one transformation.

Decisive here is that the holographic storage medium for the complete storage of all three-dimensional information only once, namely for the duration of the single shot, exposed or used, without the need for successive operations or even a deletion of stored information to record more information. The superimposition of the information in the optical storage medium takes place in the form of a wavelength division multiplexing method.

With this single shot in the sense of the present invention, a complete acquisition of all three-dimensional information of the particular diffuse scattering object by a single shot done by exposing the object and holographic buffer once for a fixed duration, with an intermediate deletion stored in the optical storage medium Information is not required.

Advantageously, the exposure time is limited only by integration properties, and is thus in a range between a few microseconds to a few seconds. For individual applications, however, a much higher exposure time, for example, of many minutes or even several hours may be advantageous. The holography method according to the invention can be carried out entirely without moving parts ("no-motion" imaging), so that no mechanical wear, mechanical inaccuracies or external influences such as temperature fluctuations influence the imaging or require maintenance, correction or adjustment work.

The method according to the invention makes use of holography with many different wavelengths, which can be termed polychromatic holography. This is a special form of holography commonly known as monochromatic. By holography can be known by fixing the coherent in beam superposition, ie generally monochrome light in the intersection resulting interference pattern, which is referred to as hologram and here and later diffraction of a light beam at this hologram emanating from the object and the intersection of the rays to be reconstructed.

In the holographic method according to the invention and the holographic arrangement according to the invention, the object, which is generally three-dimensional, is illuminated with at least one spectrally broadband illumination light source and the light transmitted through the object and / or reflected therefrom for storage of the object is encoded in many different holograms in one Holographic storage medium with at least one reference beam superimposed on the storage medium. In this case, the reference beam can preferably be diverted from the illumination light source. This has the advantage that no additional reference light source must be provided.

According to the invention, the light transmitted and reflected by the object and the reference beam are guided in such a way that they intersect in the storage medium. As a result of the superimposition of the light beams, the individual monochromatic light components form static interference patterns, ie. H. the light of a given wavelength coming from the object forms an interference pattern only with the monochrome part of the reference light which has the same wavelength. This happens for all monochrome light components of the two beams. The structure of the wavefront emanating from the object is thereby retained by interfering with a reference wave and fixing the interference structure in the storage medium as a hologram.

In the method according to the invention and the corresponding device, the object to be examined is illuminated by a spectrally broadband illumination light source. This spectrally broadband light may, for example, come from a lamp, a light emitting diode, a superluminescent diode, a multi-mode laser diode, or a combination thereof, wherein the illumination light source may have a continuous spectral distribution.

Alternatively, in a preferred embodiment of the invention, a laser unit or a laser diode unit can be used which has a discrete spectrum and continuously emitted, the emission of several individual wavelengths simultaneously or sequentially, preferably without interruption possible. At the same time means in the context of the invention, both a completely simultaneous emission of all wavelengths as well as any temporal sequence within the exposure time, d. H. the duration of the single shot. The spectral width of the light source is determined in this case by the smallest and the largest wavelength. Spectral broadband in the context of the invention is to be understood in such a way that the depth resolution achievable according to the invention is determined by the coherence length of the light used, as in OCT. The coherence length is inversely proportional to the bandwidth of the light source which according to the invention is preferably at least a few tens of nanometers.

The spectrum of the illumination light source, continuous or discrete, may have a spectral width of at least 30 nm, preferably a spectral width of between 30 nm and 120 nm, in particular of approximately 100 nm.

The reflected or transmitted by the object light is superimposed with a reference beam on the holographic storage medium, which is written in the storage medium, a hologram. If appropriate, this can be done using optical images and / or optical transformations, for example by means of a suitable lens arrangement. The reference beam can be emitted by an independent reference light source, whereby it is also broadband spectrally and may have the same structure as the illumination light source. Alternatively, the reference beam can preferably be branched off from the beam originating from the light source by means of a beam splitter. In this case, the illumination light source and the reference light source are identical.

As already mentioned, the fast imaging, ie the complete acquisition of all three-dimensional information of the object by a single shot in the sense of a single shot, ie a single exposure done. In this case, the illumination light source or the reference light source can emit a continuous spectrum or a discrete spectrum, ie, at least several, preferably a plurality of different wavelengths emit simultaneously. The depth information about the object, ie information about its internal structure, is thereby encoded in the spectrally resolved intensity distribution of the interference pattern stored in the storage medium.

In order to be able to buffer both the lateral object dimensions and the depth dimension of the object encoded in the spectral intensity distribution in one shot, it is particularly advantageous according to the invention to use a thick holographic storage medium which, due to its wavelength-selective function, is the Bragg applicable to thick holographic storage media Condition is appropriate. In the two-dimensional case, d. H. When using thin holographic storage media, only monochrome holograms are possible because no time-stable interference pattern is possible with variation of the wavelength. For storing the wavelength-specific holograms or interference patterns, therefore, a storage medium is used according to the invention that is known in holography as optically thick holographic medium and at least several wavelengths thick, so that the spectrally different holograms can be superimposed without disturbing each other ,

The spectrally different holograms do not influence each other if the already mentioned Bragg condition applies to the diffraction. This is only the case if one also expresses the interference pattern of the hologram on the third spatial dimension and uses, in which case so-called thick holograms arise. This is only possible in a visually thick storage medium. This way given way spectrally different holograms do not affect one another superimposed, which is referred to here as "Wavelength Multiplexing", is a fundamental part of the device and the inventive method. On the one hand, the use according to the invention of this wavelength multiplexing makes it possible to simultaneously record the information about the three object dimensions coded in the many spectrally different interference patterns. On the other hand, there is the possibility of selectively reading them out in the read-out step in the manner according to the invention, namely spectrally selective, so that the three object dimensions can be reconstructed from the information.

According to the invention, therefore, a plurality of differently colored holograms can be superimposed in the holographic storage medium, which are individually selectable during readout.

In addition, in this way only a small local extent for the data registration is necessary. According to the invention, the storage medium may have a thickness of approximately at least 10 μm, preferably between 50 μm and approximately 200 μm. It should be noted that the selection of the holograms necessary for readout is all the better, the thicker the storage medium is. In addition, the angle of the writing beams to each other plays a role here. The larger the angle, the better spectral selectivity prevails during the diffraction process. The angle can be according to the invention between 20 ° and 180 °, preferably 150 °.

By the use according to the invention of the wavelength multiplexing method and an optically thick medium, it is possible for the spectrally different holograms necessary for the generation of the depth information to be parallel in time or simultaneously, ie. H. take in a single exposure process or a single shot, without an intermediate deletion of the storage medium is necessary. The storage medium is only once, i. H. described during the single shot. The decisive time constant for this single shot is the exposure time of the holographic memory, which can be between one microsecond and one second depending on the storage medium. In special applications, longer exposure times can be used. Instead of a permanent simultaneous irradiation with all wavelengths during the exposure time, according to the invention it is also possible to carry out arbitrary temporal sequences of the wavelengths during the single-shot exposure.

The two-dimensional extent of the holograms entails the lateral image information of the object. With the method according to the invention, it is therefore possible to simultaneously record all three dimensions of the object with one shot or with only one exposure.

By using the method according to the invention with an illumination light source or reference light source with a continuous spectrum, a large number of monochrome holograms are simultaneously written into the optically thick storage medium and superimposed accordingly. This can create an effect known as crosstalk. This is understood to mean that when the holographic medium is illuminated with a discrete wavelength in which a plurality of monochrome holograms superimposed by wavelength multiplexing are stored (inter alia a hologram written with the read-out wavelength), not only Diffraction occurs at the readout wavelength written hologram, as would be idealized by the Bragg condition. Instead, the light is also diffracted by monochrome holograms written at a slightly different wavelength. At which spectral deviation from the write wavelength such diffraction still occurs has to do with the wavelength of the light, the angle between the writing beams and the thickness of the storage medium. Advantageous values have already been indicated for the latter parameters.

In order to prevent the disturbing crosstalk between the monochrome holograms, the use of an illumination light source with a discrete spectrum is recommended, with the individual wavelengths forming support wavelengths. Here, it is technically better the more wavelengths are used as nodes and the wider the spectral range over which the wavelengths extend. Finally, it should be noted that the thicker the optical storage medium is, the tighter the spectral support points may be, without the crosstalk interfering. Thus, it is possible to preferably use a particularly thick optical storage medium.

For reading the entire stored hologram, the storage medium is exposed by a readout beam of a readout light source, which may have the same structure as the illumination light source, wherein the readout light source may have the same time-averaged spectrum. For the three-dimensional reconstruction of the object structure, it is necessary to be able to read out the many spectrally different or multiplexed holograms again selectively and independently of one another. This is achieved by reading out the many holograms with only one of the wavelengths used for writing. When the crosstalk is sufficiently suppressed, the light is selectively diffracted only at the hologram that has been written at the appropriate wavelength. According to the invention, therefore, the read-out is preferably carried out with the same discrete, different wavelengths, possibly distributed over the entire spectral range of the illumination light source used for writing. The emission of the different wavelengths takes place in good time successively. The readout is done to save time during the recording process usually after the single shot.

Alternatively, in special applications, for. B. with long exposure times but also a start of the readout already be favorable during the recording. This has the advantage that with this method also applications are possible that register changes in the object. This possibility of reading during the saving is given because it can be read independently of the writing process. For this purpose, the polarization of the readout light is chosen to be orthogonal to the polarization of the light of the write beams. Thus, an influence on the storage process by the reading is largely avoided.

The diffracted light from the hologram of a certain wavelength can, optionally with reference to optical images and / or optical transformations, for example, by a lens arrangement, with a sensor, preferably with a two-dimensional CCD array, taken and processed immediately or initially stored and later processed , In this way, the lateral structure information of the object, i. H. that information of the object which lies within a plane perpendicular to the illumination beam is recorded on the two dimensions of the CCD array, which consists of individual pixels. Preferably, this is done successively for all wavelengths used, so that an iterative method is particularly advantageous for the reconstruction of the image information buffered in the holographic medium, in which the wavelengths are passed through successively during readout.

The three-dimensional reconstruction of the object can be carried out according to the invention by applying a mathematical transformation rule to the sensor signal. In this case, the image data for an object dimension can be calculated from the sensor signal, in particular from the signal of the two-dimensional CCD array, by the intensity read out with the mathematical transformation of the readout beam as a function of the wavelength. For this purpose, the device according to the invention can have a computing unit, for example a computer, with which the mathematical transformation can be applied to the sensor signal. Particularly advantageous here is the inclusion of a Fourier inverse transformation.

The entire two-dimensional CCD array is subject to a lateral light distribution which represents a two-dimensional projection or image of the wave front coming from the object and falling through the holographic medium and reconstructed therefrom. This reconstructs the two lateral object dimensions. The reconstruction of the depth dimension results from the fact that the interference signal of two coherent light waves superimposed at one point depends on the wavelength of the light in its intensity: whether constructive or destructive interference results depends on the transit time difference or is equivalent to the path difference, and is dependent on the wavelength. Locally, z. For example, constructive interference for each different wavelength at a different location, ie at a fixed location within the illuminated area of the holographic storage medium results in a different intensity for different wavelengths. By inverse Fourier transformation and optionally by resorting to further mathematical and / or optical transformations, the depth structure of the object can be determined from these at a location of different intensities depending on the wavelength. Because it is feasible due to the Bragg condition to selectively diffract only light of a specific wavelength, it is possible to read out the spectral intensity spectrally selectively stored as interference pattern again. Together with the above-described possibility of object reconstruction in a lateral respect, it is thus possible with the device according to the invention or the method according to the invention to read out the information about the three local object dimensions in an assignable manner.

As an alternative to the prescribed iterative readout method, a so-called single-shot readout can also be used. The hologram can be illuminated simultaneously with all readout wavelengths. The diffracted holograms of the individual wavelengths are then spatially separated and recorded in parallel with many sensors, in particular CCD arrays, wherein a two-dimensional CCD array can be provided for each readout wavelength. To separate the light beams diffracted by the holograms, it is possible, for example, to use spectrally selective optics, in particular dichroic mirrors and / or beam splitters, or prisms or optical gratings.

Alternatively, it is also possible to use a two-dimensional CCD array in such a way that the spectrally different portions of the light fall onto different two-dimensional areas of the CCD array, ie each wavelength onto a different subarea of the CCD array.

In a particularly advantageous embodiment of the present invention can be used as an optically thick storage medium, a medium having a so-called photorefractive effect, in particular a photorefractive crystal, with which due to its property to record only changes in light intensity, a uniformly bright background can be hidden which is generated for example by the useless multiply scattered photons, which have lost the coherence property. In alternative embodiments, a photorefractive polymer or a holographic film may also be used.

As a particularly advantageous application of the present invention, the use in the biomedical in-vivo imaging can be called because the risk of camera shake is reduced due to the very short recording times (critical time results from the patient's heart rate), the patient with it in addition to avoiding inconveniences such as long silence, and because the method is safe for human applications because non-ionizing radiation is used. In contrast, ionizing radiation such. B. X-rays, as is known, cancer. For the application of the method according to the invention, the object can be illuminated by a strongly light-scattering medium. Furthermore, the light transmitted or reflected by the object can also be conducted through a medium that scatters strongly light. In biomedical research and / or diagnostics, the object may also be illuminated, for example, by a microscope or endoscope, and / or the light transmitted and / or reflected by the object may be passed through a microscope or endoscope.

An alternative advantageous application example results from the alternative method according to the invention already mentioned above, in which the exposure time is selected to be longer, and during individual exposure and independently of which selectively individual information is read out, for example spectrally selected holograms. Such an advantageous application is, for example, a monitoring of longer-lasting processes. As an example, the analysis of long-term change processes of cell cultures should be mentioned.

The invention will be explained in more detail with reference to embodiments shown in FIGS. Further features and advantages of the invention can be found in the description of the embodiments and the figures.

Show it:

1 A device according to the invention for realizing the inventive method of three-dimensional single shot imaging in a general schematic form, in which the object is addressed in reflection.

2 A device according to the invention for implementing the method according to the invention of the three-dimensional single shot image in a detailed schematic form, in which the object is addressed in reflection.

3 A device according to the invention for implementing the method of the three-dimensional single shot imaging according to the invention, in which the object is passed through in transmission.

4 A device according to the invention for realizing the method according to the invention of the three-dimensional single-shot imaging, in which the object beam and the reference beam fall from the same side onto the holographic medium.

5 : A flow chart illustrating the method.

An advantageous embodiment of the invention is shown in the schematic representation 1 shown as well as in 2 a little more detailed design. As illumination light source 1 For example, a laser unit employs a laser that can emit many different wavelengths, for example, 1024 different wavelengths over a spectral range of about 70 nm at a central wavelength of about 830 nm both simultaneously and sequentially.

For this purpose, a semiconductor laser 1a used with external resonator, as in 2 is shown. The resonator includes a diffraction grating 1b and a lens 1c for the spatial separation of the spectral components of the semiconductor laser 1a and a switching element 1d to the reflection of the spectral components at the end mirror 1e to control individually. In the embodiment according to 2 has the diffraction grating 1b 800 lines per millimeter and the focal length of the lens 1c is preferably about 150 mm. Furthermore, the switching element 1d designed as a liquid crystal display, in particular as a line array with 1024 pixels.

Such an embodiment of the laser unit is known from the German patent with file number DE 199 09 497 C1 known. The collimated output beam 2 the illumination light source 1 gets onto a beam splitter 3 directed and split there in two parts. The first ray, the object ray 4 runs to the object to be imaged 5 , The light reflected by the object 6 runs on the thick holographic storage medium 7 , which is designed as a photorefractive polymer with a thickness of about 200 microns. The storage medium 7 is sensitive to the spectral range of about 830 nm central wavelength and is characterized by high efficiency (over 90% at best) and suppressed by the photorefractive effect of the scattered light background.

As in 3 a further realization variant is shown in such a way that the object beam 4 the object 5 passes through in transmission and then via a deflection mirror 15 as a transmitted light beam 6 on the holographic storage medium 7 meets. In both variants mentioned, the second runs from the beam splitter 3 generated reference beam 8th via another beam splitter 9 on the thick holographic storage medium 7 where he is on the object beam 6 meets. The angle θ between object beam 6 and reference beam 8th is chosen to give a broad interference pattern and a good condition for high spectral selectivity by the Bragg condition in the storage medium 7 given is. This is the case for the angles of the beams that are as large as possible relative to one another and in the exemplary embodiments according to the figures is approximately θ = 150 °.

The angle θ _O between object beam 6 and surface of the storage medium 7 becomes substantially equal to the angle θ _R between the reference beam 8th and surface of the storage medium 7 chosen and thus results from the angle θ. It can be between a so-called transmission geometry, where the object beam 6 and the reference beam 8th on the same surface in the storage medium 7 to enter, see 4 , and a so-called reflection geometry, in which the object beam 6 and the reference beam 8th on different, in particular opposite surfaces enter the storage medium, see 1 - 3 to be distinguished. These two alternative geometries correspond to the transmission holography known in classical holography and the reflection holography, which refer in particular to the diffraction direction of the light diffracted at the hologram. The reconstructed diffracted light is transmitted through or reflected by the holographic storage medium. It should be noted that this is not to be confused with the aforementioned alternative possibilities transmitted by the object (in contrast to the holographic storage medium just described) ( 3 ) or reflected light ( 2 ) for recording.

The readout process begins in the exemplary embodiment presented after the end of the single shot exposure. To read the image information, the object beam 4 with the help of a mechanical shutter 11 blocked. The laser unit 1 is switched to a temporally successive emission of all wavelengths.

The reference beam 8th now serves as a readout beam and it creates a beam bent by the hologram 12 which runs in the direction of the unblocked object beam as through the storage medium 7 transmitted beam 10 when recording the hologram, and that corresponds to a reconstructed object beam. This bowed selection beam 12 contains the image information for each individual wavelength. He falls on a detector 13 with the ability to capture the incident light two-dimensionally, preferably on a two-dimensional CCD array 13a , please refer 2 , Alternatively, a CCD camera can be used.

A previously from the selection beam 8th at a beam splitter 9 diverted another beam 16 , here referred to as "interference beam", is through a mirror 17 on the CCD array 13a directed and superimposed at the location of the CCD array 13a with the bent selection beam 12 , As a result, the phase information of the read-out beam is obtained at the location of the CCD array for each selected wavelength 8th which corresponds to the reconstructed object beam. This makes it possible to obtain the depth dimension of the object from the constructive or destructive interference which results in each case at the different wavelengths by means of mathematical transformations, in the exemplary embodiment with recourse to a Fourier inverse transformation. Together with the lateral image of the object on the CCD array given by the holographic image 13a can with the help of a computer 14 the three object dimensions are reconstructed using mathematical transformations.

A favorable variant of the method steps of the three-dimensional single-shot holography is shown in the flow chart according to FIG 5 , and is explained below.

For the writing procedure of the multi-colored holograms, the laser unit 1 brought into a multi-mode operating state in which as many, desirably 1024, discrete wavelengths are emitted during the single shot exposure so that spectrally broadband light is generated, step S1: generation of spectrally broadband light 2 in which the condition Δλ / λ> 0.1 preferably applies. This emission can be exactly the same time, but also in any temporal order, such. A fast wavelength sweep, within the single shot exposure time.

In a subsequent step S2, the generated light is collimated and in the two writing beam reference beam 8th and object beam 4 split. The object beam 4 meets the object 5 at which it reflects or through which it is transmitted, with the reflected or transmitted light continuing in the direction of the holographic storage medium 6 , Step S3: object beam 4 meets object 5 , then continue 6 in the direction of storage medium 7 , The reference beam 8th In contrast, it runs over the same distance as the object beam 4 respectively. 6 towards holographic storage medium 7 , Step S4. In the holographic storage medium, the superposition of object beam takes place 6 and reference beam 8th wherein for each given wavelength, a hologram is simultaneously generated and all holograms are superimposed in one exposure, step S5.

The reading procedure now becomes a readout beam 8th with the same time-averaged spectrum as that of the write beams used and falls from the direction of the reference beam on the holographic storage medium 7 so that the holograms can be read out again, see step S6: Exposure of the storage medium 7 with readout beam of the same wavelengths as the write beams. The different wavelengths are emitted temporally one after the other. From the selection beam 8th becomes another interference beam before hitting the storage medium 16 branched off, step 57 , The reading beam bent at the hologram 12 superimposed with the interference beam 16 on a surface detector, e.g. B. a two-dimensional CCD array 13 whose signal is stored by a computer: see step S8: reading beam diffracted at the hologram 12 falls on CCD array 13 , superimposed there with interference beam 16 ; Detector signal is stored. This is repeated serially for all wavelengths.

For each pixel of the CCD array, the intensity values resulting at the different wavelengths are converted by means of a Fourier inverse transformation into a depth information for that location representing the pixel, so that a representative depth profile results for all the pixels of the CCD array. Step S9: Determine the depth information by applying a Fourier inverse transformation per CCD array pixel. The depth profiles of the individual pixels are combined in a further step S10 such that a reconstruction of the three-dimensional image of the object results.

Claims (28)

Holographic method for the high-resolution three-dimensional imaging of an object ( 5 ), in which the object ( 5 ) with at least one illumination light source ( 1 ) and the transmitted and / or reflected light ( 6 ) for storing holograms in a holographic storage medium ( 7 ) with a reference beam ( 8th ) of a reference light source on the storage medium ( 7 ), wherein the readout of the stored holograms takes place with a read-out light source, characterized in that the illumination, reference and read-out light source are spectrally broadband light sources, that as storage medium ( 7 ) an optically thick holographic storage medium is used and that all three-dimensional information of the object ( 5 ) defined by a single shot duration of the illumination and reference light source ( 1 ) in the holographic storage medium ( 7 ), the stored information for reconstructing a three-dimensional image of the object ( 5 ) are read out optically and the reading is carried out using at least one transformation and wherein the depth information about the internal structure of the object ( 5 ) is reconstructed from the spectrally resolved intensity distribution of the interference pattern stored in the storage medium. Method according to claim 1, characterized in that in the holographic storage medium ( 7 ) are superimposed on a variety of different holograms, which are individually selectable when reading. A method according to claim 1 or 2, characterized in that the duration of the single shot is in the range between one microsecond and several seconds. Method according to claim 1, 2 or 3, characterized in that the illumination light source ( 1 ) during the single shot on a plurality of individual wavelengths emitted simultaneously or sequentially. Method according to one of the preceding claims, characterized in that for reading the stored hologram, the storage medium ( 7 ) is illuminated by the readout light source with a readout beam, wherein the readout light source simultaneously or temporally successively emits light of the same wavelengths as the illumination source ( 1 ). Method according to one of the preceding claims, characterized in that the diffracted by the stored hologram light of the readout light source with at least one two-dimensional sensor ( 13 ) is detected. A method according to claim 6, characterized in that for detecting each of the two lateral object dimensions each have a dimension of the two-dimensional sensor ( 13 ) is completely available. A method according to claim 6 or 7, characterized in that the sensor signal is subjected to a Fourier inverse transformation. Method according to claim 8, characterized in that the inverse Fourier transform per pixel of the sensor ( 13 ) is made. Method according to one of claims 8 or 9, characterized in that from the sensor signal, the image data to an object dimension by the with the Fourier inverse transformation of the with the readout beam ( 8th ) are calculated as a function of the wavelength. Method according to one of the preceding claims, characterized in that it is adapted to at least one light scattering object ( 5 ) or at least one object ( 5 ) through a light scattering medium. Method according to one or more of the preceding claims, characterized in that the reading is carried out by a single shot. Method according to one or more of the preceding claims, characterized in that the reading takes place during the storage. A method according to claim 13, characterized in that the reading during the storage process with the light of the illumination light source ( 1 ) orthogonally polarized light. Use of the method according to one or more of the preceding claims in biomedical research and diagnostics. Apparatus for carrying out the method according to one of claims 1 to 14 with at least one illumination light source ( 1 ), at least one reference light source, a holographic storage medium ( 7 ) and at least one readout light source, wherein with the beam of the illumination light source ( 1 ) the object ( 5 ) and the beam of the illumination and reference light source ( 4 . 8th ) is guided in such a way that the object ( 5 ) transmitted and / or reflected light ( 6 ) with the beam of the reference light source ( 8th ) in the storage medium ( 7 superposed, and wherein the illumination and reference light source ( 1 ) are spectrally broadband and the storage medium ( 7 ) is an optically thick holographic storage medium, characterized in that the device is adapted to all three-dimensional information of the object ( 5 ) defined by a single shot duration of the illumination and reference light source ( 1 ) in the holographic storage medium ( 7 ) stored information for reconstructing a three-dimensional image of the object ( 5 ) optically read and carry out the reading using at least one transformation and the depth information on the internal structure of the object ( 5 ) to reconstruct from the spectrally resolved intensity distribution of the interference pattern stored in the storage medium. Apparatus according to claim 16, characterized in that the illumination light source ( 1 ) is emitted at a plurality of individual wavelengths simultaneously or during the single shot in a particular time sequence. Device according to one of claims 16 or 17, characterized in that the Readout light source sequentially or simultaneously emits the same wavelengths as the illuminating light source ( 1 ). Device according to one of claims 16 to 18, characterized in that the reference light source with the illumination light source ( 1 ) and the reference light beam ( 8th ) of the illumination light beam ( 2 ) is split off. Device according to one of claims 16 to 19, characterized in that at least one two-dimensional sensor ( 13 ) is provided, with which the from the storage medium ( 7 ) diffracted light ( 12 ) of the readout light source is detectable. Apparatus according to claim 20, characterized in that a computing unit ( 14 ) is provided with the for determining the depth dimension of the object ( 5 ) to the measuring signal of the sensor ( 13 ) a Fourier inverse transformation is applicable. Device according to one of the preceding claims 16 to 21, characterized in that the illumination light source ( 1 ) and / or reference light source and / or read-out light source is a light emitting diode unit of a light emitting diode or an arrangement of a plurality of light emitting diodes and / or a laser unit. Device according to one of claims 16 to 22, characterized in that the illumination light source ( 1 ) and / or the reference light source light ( 2 ) emits several different wavelengths simultaneously. Device according to one of the preceding claims 16 to 23, characterized in that the optically thick holographic storage medium ( 7 ) is a photorefractive-effective medium, a photorefractive crystal, a photorefractive polymer, or a holographic film. Device according to one of claims 16 to 24, characterized in that the spectrum of the illumination light source ( 1 ) has a spectral width of at least 30 nanometers. Device according to one of claims 16 to 25, characterized in that the read-out light source is identical to the reference light source and the read-out light beam from the reference light beam ( 8th ) is split off. Device according to one of claims 16 to 26, characterized in that the storage medium ( 7 ) has a thickness of at least 10 microns. Device according to one of claims 16 to 27, characterized in that the angle (θ) between the on the storage medium ( 7 ) incident reference light beam ( 8th ) and the object ( 5 ) transmitted or reflected illumination light beam ( 6 ) is between 20 ° and 180 °.

Images