WO2022027336A1 - 一种滤光器、滤光方法及多光谱成像系统 - Google Patents

一种滤光器、滤光方法及多光谱成像系统 Download PDF

Info

Publication number
WO2022027336A1
WO2022027336A1 PCT/CN2020/107160 CN2020107160W WO2022027336A1 WO 2022027336 A1 WO2022027336 A1 WO 2022027336A1 CN 2020107160 W CN2020107160 W CN 2020107160W WO 2022027336 A1 WO2022027336 A1 WO 2022027336A1
Authority
WO
WIPO (PCT)
Prior art keywords
light signal
pattern
unit
optical
filter
Prior art date
Application number
PCT/CN2020/107160
Other languages
English (en)
French (fr)
Inventor
黄晨明
余佳
廖九零
章辰
郑炜
Original Assignee
中国科学院深圳先进技术研究院
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中国科学院深圳先进技术研究院 filed Critical 中国科学院深圳先进技术研究院
Priority to PCT/CN2020/107160 priority Critical patent/WO2022027336A1/zh
Publication of WO2022027336A1 publication Critical patent/WO2022027336A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
    • G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
    • G01N21/64—Fluorescence; Phosphorescence
    • G—PHYSICS
    • G02—OPTICS
    • G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
    • G02B27/00—Optical systems or apparatus not provided for by any of the groups G02B1/00 - G02B26/00, G02B30/00
    • G02B27/10—Beam splitting or combining systems

Definitions

  • the present invention relates to the field of optical technology, in particular to an optical filter, a corresponding optical filtering method, and a multispectral imaging system using the optical filter.
  • Confocal microscopy is a microscopic imaging technique that uses a laser beam to scan the excitation sample point by point, and uses a pinhole to spatially filter the fluorescence signal to remove stray light from the non-focal plane. It is also a common multispectral imaging system.
  • the meaning of the term "confocal" is that the aperture at the light source, the focal point in the sample and the aperture in front of the detector are conjugated to each other. Compared with widefield microscopy, confocal microscopy has good optical resolution and signal-to-noise ratio. Generally speaking, the central wavelength of fluorescence is different from the wavelength of the laser.
  • multicolor or multispectral imaging is not a unique requirement of confocal microscopes.
  • many imaging systems such as other types of fluorescence microscopes and multispectral imagers require time-sharing or simultaneous processing of multiple different wavelength bands. light signal.
  • the conventional method for realizing multispectral imaging of such equipment including confocal microscopes is to install filter wheels, filter sets, etc., that is, filters need to be used to obtain fluorescence signals in corresponding wavelength bands, but optical filters
  • the filter band of the filter is fixed.
  • multiple filters or a multi-bandwidth filter must be installed to match different fluorescence. This method is not only expensive, slow, but also spectrally resolved. The frequency is low, the arbitrary multi-band tuning function is poor, and the imaging bandwidth also depends entirely on the filter manufacturer.
  • the commercial confocal systems that can realize the adjustable filter function include Leica's TCS SP5, Nikon (Nikon). ) of C1si et al. Referring to Figure 1 of the description, it is a representative prior art.
  • the system mainly includes excitation light module A, confocal imaging module B and multi-channel spectral imaging module C.
  • the excitation light module A includes lasers with different wavelengths, dichroic mirrors 1-5, mirrors 6 and acousto-optic tunable filters 7;
  • confocal imaging module B includes a dichroic mirror 8, an X-Y galvanometer mirror 9 and 10, scanning lens 11, tube lens 12, objective lens 13, sample 14, focusing lens 15 and aperture 16;
  • multi-channel spectral imaging module C consists of collimating lens 17, beam splitting prism 18, first, second and third Focusing lenses 19 , 22 and 25 , first, second and third variable slits 20 , 23 and 26 and first, second and third detectors 21 , 24 and 27 are composed.
  • the working principle of the tunable filter-confocal system is as follows: in the excitation light module A, lasers with different wavelengths are integrated into an optical path after passing through the corresponding dichroic mirrors, and then injected into the acousto-optic tunable filter. . Adjusting the corresponding parameters of the acousto-optic tunable filter can change the wavelength and intensity of the outgoing light.
  • the laser light emitted from A is first reflected by the dichroic mirror 8 into the two X-Y galvanometers, and then passes through the scanning lens 11, the tube lens 12 and the objective lens 13 in sequence, and the beam is focused on the sample 14. superior.
  • the two X-Y galvanometers can change the position of the excitation light on the sample, an image of an area on the sample can be obtained by scanning the focused beam with that area.
  • the fluorescence returns along the same optical path and is filtered by the dichroic mirror 8 to remove the excitation light, and finally focused by the lens 15 onto the aperture 16 .
  • the fluorescence filtered by the aperture directly enters the multi-channel spectral imaging module C. First, it is collimated by the collimating lens 17 and then enters the beam splitting prism 18.
  • the beam splitting prism disperses the fluorescence to expand its spectrum in space, and is then focused by the first focusing lens 19 into a fringe containing spectral information.
  • a first variable slit 20 is installed at the stripe, and the position and spacing of the slits are variable. Only after the relationship between the position and wavelength of the slits is determined, the system can realize tunable filtering. The filtered fluorescence can also be reflected by the mirror on the first variable slit 20 to the remaining two slits 23 and 26 , and filtered again to enter the second and third detectors 24 and 27 .
  • the movement of the slit is driven by a motor, so the speed is limited.
  • the movement of the mechanical device will inevitably bring about accuracy problems, which will lead to a certain error between the actual value and the calibration value.
  • the number of slits is limited, so usually only three-channel filtering can be achieved at most. Therefore, the above-mentioned prior art cannot actually achieve arbitrary tunable filtering (arbitrary center wavelength, arbitrary bandwidth), which also affects the multispectral imaging capability of many imaging systems including confocal microscopes.
  • the present invention proposes an optical filter with an arbitrarily tunable working band, a filtering method and a corresponding multi-spectral imaging system.
  • the unit collimates the dispersed optical signal, and combines the control of the digital micromirror device DMD to realize the tuning of arbitrary central wavelengths and (multiple) bandwidths in the filtering technology. Therefore, compared with the prior art, the optical filter and the optical filtering method of the present invention can significantly improve the imaging capability of various multispectral imaging systems such as confocal microscopes.
  • the optical filter of the present invention includes a dispersion unit for dispersing an incident light signal containing multiple wavelength bands; and a collimation unit for collimating the incident light signal dispersed by the dispersion unit; a digital micromirror device , which receives the incident light signal collimated by the collimating unit; the digital micromirror device reflects part of the light signal in the incident light signal as a reflected light signal back to the collimating unit under the control of the control unit; the mirror , the reflected light signal is incident on the reflector after passing through the collimation unit and the dispersion unit, and is reflected by the reflector into the detection light path; wherein, the incident light signal collimated by the collimation unit
  • the optical axis of the DMD does not coincide with the optical axis of the reflected light signal reflected by the digital micromirror device.
  • control unit includes a pattern generating unit and a pattern loading unit; the pattern generating unit can generate a loading pattern, and the loading pattern is composed of one or more basic patterns; the pattern loading unit loads the loading pattern Loaded into the digital micromirror device, so that part of the light signal in the incident light signal is reflected back to the collimating unit as a reflected light signal.
  • the pattern generating unit can generate a plurality of the basic patterns, the basic patterns are determined by the angle between the hypotenuse in the pattern and the edge on the target surface of the digital micromirror device, and the pattern on the digital micromirror device. The length of the upper edge and/or the lower edge of the target surface of the micromirror device is determined; the plurality of basic patterns have the same angle and have different lengths.
  • the dispersion unit is a first dispersion prism
  • the collimation unit is a second dispersion prism; wherein, the light exit surface of the first dispersion prism is parallel to the light incident surface of the second dispersion prism.
  • the structure of the first dispersion prism is the same as that of the second dispersion prism. Therefore, the functions of the dispersion unit and the collimation unit are realized by the double prism structure.
  • the present invention also provides a light filtering method, which includes the following steps: dispersing an incident light signal containing multiple wavelength bands; collimating the dispersive incident light signal; using a digital micromirror device to receive the collimated light signal direct incident light signal; control the digital micromirror device to reflect part of the light signal in the incident light signal as a reflected light signal; use a mirror to reflect the reflected light signal into the detection light path; wherein, the The optical axis of the collimated incident optical signal does not coincide with the optical axis of the reflected optical signal reflected by the digital micromirror device.
  • the operation of controlling the digital micromirror device to reflect part of the incident optical signal as a reflected optical signal is specifically implemented by loading a loading pattern into the digital micromirror device; wherein the The loading pattern is generated by the following steps:
  • the basic pattern is determined by the angle between the hypotenuse in the pattern and the edge on the target surface of the digital micromirror device, and the length of the edge and/or the lower edge of the pattern on the target surface of the digital micromirror device. Determine; the plurality of basic patterns in the step S1 have the same angle and have different lengths.
  • a first dispersion prism is used to disperse the incident light signal including multiple wavelength bands
  • a second dispersion prism is used to collimate the dispersed incident light signal.
  • the structures of the first dispersive prism and the second dispersive prism may be the same.
  • the functions of dispersion and collimation are achieved by using a double prism structure.
  • the present invention also provides a multispectral imaging system, which includes the above-mentioned optical filter, and an optical signal generating unit.
  • the light signal generating unit may be, for example, a confocal microscope.
  • the optical signal is transmitted between the confocal microscope and the optical filter by means of optical fiber transmission or spatial transmission.
  • the optical filter, the optical filtering method and the corresponding multispectral imaging system of the present invention can easily adjust the filtering performance of the optical filter by manipulating the pattern loaded by the digital micromirror device DMD, which can not only realize the filtering of a single waveband but also Can achieve arbitrary multi-band multi-channel filtering. Since the tuning speed of the digital micromirror device DMD is extremely fast, which can reach 10 kHz or even several tens of kHz, the speed of the optical filter of the present invention will not affect the original imaging system such as confocal, so the above can be achieved. Imaging Systems Ultra-fast multispectral imaging.
  • the digital micromirror device DMD in the system is fixed during the imaging process and does not require any mechanical movement, so there is no problem that the actual wavelength is different from the calibrated wavelength due to insufficient mechanical movement precision in the prior art. Therefore, the technical solution of the present invention can greatly improve the multispectral imaging performance of the multispectral imaging system.
  • FIG. 1 is a schematic structural diagram of a confocal microscope with adjustable filter function in the prior art
  • FIG. 2 is a schematic structural diagram of a confocal microscope system of the present invention
  • Fig. 3 is the structure schematic diagram of the optical filter of the present invention.
  • Fig. 4 is the principle schematic diagram of the first and second dispersive prism collimation dispersive beam of the present invention.
  • 5a is a schematic diagram of the separation of fluorescent light beams of an optical filter of the present invention.
  • 5b is a schematic diagram of the DMD control unit of the present invention.
  • Fig. 7 is the schematic diagram of dispersion fringe of the present invention.
  • Fig. 8 is the DMD use state diagram of the present invention.
  • FIG. 10 is a schematic diagram of a series of basic patterns of the present invention.
  • FIG. 11 is a schematic diagram of another series of basic patterns of the present invention.
  • FIG. 13 is the experimental result one of filtering the spectrum of the LED light source according to the present invention.
  • Fig. 14 is the experiment result 2 of filtering the spectrum of the LED light source according to the present invention.
  • FIG. 15 is the third experiment result of filtering the spectrum of the LED light source according to the present invention.
  • FIG. 16 is the experiment result four of filtering the spectrum of the LED light source according to the present invention.
  • Fig. 17 is the experiment result five of filtering the spectrum of the LED light source according to the present invention.
  • Fig. 18 is the experiment result six of filtering the spectrum of the LED light source according to the present invention.
  • Fig. 19 is the resolution test chart of filtering the spectrum of the LED light source according to the present invention.
  • Figure 21 is a schematic diagram of another series of basic patterns in different directions of the present invention.
  • FIG. 22 is a schematic structural diagram of another embodiment of the confocal microscope system of the present invention.
  • Prior art drawings 1-5- dichroic mirror, 6-reflector, 7- acousto-optic tunable filter, 8- dichroic mirror, 9- XY galvanometer, 10- XY galvanometer Mirror, 11-scanning lens, 12-tube lens, 13-objective lens, 14-sample, 15-focusing lens, 16-small hole, 17-collimating lens, 18-beam splitting prism, 19-first focusing lens, 20- First variable slit, 21-first detector, 22-second focusing lens, 23-second variable slit, 24-second detector, 25-third focusing lens, 26-third variable Slit, 27 - Third detector.
  • Example drawings 201-laser, 202-collimating lens, 203-dichroic mirror, 204-galvanometer, 205-scanning lens, 206-tube lens, 207-objective lens, 208-sample, 209-focusing lens, 210-small hole, 211-fiber, 212-filter, 301-filter collimating lens, 302-reflector, 303-first dispersion prism, 304-second dispersion prism, 305-digital micromirror device, 306 - filter slit, 307 - filter focusing lens, 308 - detector.
  • a confocal microscope is taken as an example to illustrate the optical filter, the optical filtering method and the corresponding imaging system of the present invention.
  • the confocal microscope is a conventional multispectral imaging system, which generally includes a laser, a pinhole, a galvanometer, an objective lens, a dichroic mirror and a collimating lens.
  • a laser a laser
  • a pinhole a galvanometer
  • an objective lens a dichroic mirror
  • collimating lens a collimating lens
  • the light beam from the laser is expanded to reach the dichroic mirror, the galvanometer, the scanning lens, the tube lens and the objective lens respectively.
  • the fluorescence generated by excitation returns to the original way through the objective lens, tube lens, scanning lens, and galvanometer, and then is filtered by the dichroic mirror to remove the laser light and some stray light, and then is filtered by the pinhole and then coupled into the fiber.
  • the fluorescent signal is transmitted in an optical fiber, and the other end of the optical fiber is a filter.
  • the optical filter of the present invention includes, for example, an optical fiber entrance port, a filter collimating lens, a dispersive unit (for example, composed of a first dispersive prism), a collimation unit (for example, composed of a second dispersive prism), a digital micromirror device ( DMD), mirrors, slits, focusing lenses and detectors.
  • the filter collimating lens such as a fiber collimating lens, collimates the fluorescence signal at the fiber port and sends it to the dispersion unit.
  • the fluorescence signals of different wavelengths are expanded in parallel in space, and the dispersion expanded fluorescence signal passes through the collimation unit and is irradiated on the DMD.
  • the fluorescent signal of a specific wavelength can be reflected off the original optical path and received by the detector.
  • a confocal microscope is used as an example to describe what kind of light signal the light signal generating unit can generate.
  • FIG 2 it is a schematic structural diagram of the confocal microscope system of the present invention.
  • the laser 201 , the collimating lens 202 , the dichroic mirror 203 , the galvanometer 204 , the scanning lens 205 , the tube lens 206 , the objective lens 207 , the focusing lens 209 and the small hole 210 form a laser scanning confocal microscope.
  • the confocal microscope is connected to the filter 212 through the optical fiber 211 .
  • the optical path structure of the laser scanning confocal microscope is as follows: the laser light is emitted from the laser 201 and collimated into a laser beam with a suitable spot size after passing through the collimating lens 202. After selecting the corresponding dichroic mirror 203, the light beam can be transmitted through 203. The collimated light beam transmitted from the dichroic mirror 203 passes through the galvanometer mirror 204 , the scanning lens 205 , the tube lens 206 and the objective lens 207 in sequence, and finally the light beam is focused on the sample 208 . The fluorescence generated by exciting the sample 208 returns to the original path and is reflected by the dichroic mirror 203 to the focusing lens 209. The focusing lens 209 focuses the light beam on the small hole 210. Stray light.
  • the above-mentioned stray light filtered fluorescence is transmitted to the optical filter 212 through the optical fiber 211 .
  • the structure of the optical filter 212 of the present invention is shown in FIG. 3 of the specification.
  • the filter 212 includes a filter collimating lens 301 , a mirror 302 , a dispersive unit 303 , a collimating unit 304 , a digital micromirror device (DMD) 305 , a filter slit 306 , a focusing lens 307 and a detector 308 .
  • DMD digital micromirror device
  • the function of the dispersion unit 303 is to expand the optical signals of different wavelength bands in space
  • the function of the collimation unit 304 is to collimate the dispersed optical signal to ensure the realization of the subsequent filtering process.
  • the dispersion unit 303 is, for example, a first dispersion prism
  • the collimation unit 304 is, for example, a second dispersion prism, wherein the first dispersion prism and the second dispersion prism may be the same dispersion prism.
  • those skilled in the art can know that as long as the modification can realize the above functions, it belongs to the scope of the present invention.
  • the basic optical path structure of the optical filter 212 is as follows: the fluorescent signal filtered by the small hole 210 is coupled into the optical fiber 211 for transmission, and when the fluorescent light comes out from the other end of the optical fiber 211, it is collimated by the collimating lens 301 of the optical filter, and then A specific incident angle is incident on the first dispersing prism, namely the dispersing unit 303, and the first dispersing prism will disperse the light signal and spatially expand its spectrum. The spectrally expanded dispersive light is then incident on the second dispersive prism, namely the collimating unit 304, at a specific incident angle.
  • the second dispersive prism will collimate the dispersive light, and the collimated dispersive light will finally be irradiated on the digital micromirror. device (DMD) 305.
  • DMD digital micromirror. device
  • the light beam slightly deviates from the original optical path and returns, that is, after passing through the second dispersive prism and the first dispersive prism again, it is irradiated on the reflecting mirror 302, and the reflecting mirror 302 changes the beam direction It is focused by the filter focusing lens 307 .
  • a fixed slit 306 can also be placed in front of the filter focusing lens 307 for filtering out stray light in the optical path.
  • the desired fluorescent signal is received by detector 308 .
  • FIG. 4 of the description shows a schematic diagram of the principle of collimating the dispersed light beam by the dispersion unit 303 (the first dispersion prism) and the collimation unit 304 (the second dispersion prism). Since the dispersive prism can generally achieve the largest clear aperture when used at the smallest deflection angle, it is preferable that the fluorescence collimated by the filter collimating lens 301 is incident into the first dispersive prism with the smallest deflection angle. The placement of the prisms is shown in Figure 4 of the description.
  • the first dispersive prism and the second dispersive prism need to be placed in parallel, that is, the light exit surface of the first dispersive prism and the light entrance surface of the second dispersive prism are both parallel.
  • the collimated beam is spectrally expanded after passing through the first dispersive prism and forms a dispersive fringe.
  • the dispersive fringes become a beam of collimated dispersive light after passing through the second dispersive prism, that is, the length of the fringes in the unfolding direction does not change due to the observation distance, so that the beams with different wavelength components are parallel to each other although the spatial positions are different. .
  • the collimated scattered light can ensure that the beam is also a beam of collimated light after returning to the original path, and will not be expanded by dispersion.
  • the digital micromirror device (DMD) 305 can reflect the light beam corresponding to the wavelength band that needs to be retained by loading a specific pattern through the control unit shown in FIG. 5b of the specification. The generation and loading of the specific pattern are described later.
  • the incident light is perpendicular to each micromirror in the DMD, the reflected light will coincide with the incident light, so it is not easy to accurately select the wavelength band selected by the DMD, so it is necessary to make some small adjustments to some devices to make the reflection Light can be separated from incident light.
  • the method adopted in this embodiment is to fine-tune the angle of the DMD305.
  • the reflection direction of the micromirror corresponding to a specific wavelength band (selected and reserved wavelength band) in the DMD305 can be fine-tuned in the vertical direction and/or the horizontal direction, so that the selected part of the DMD is reserved.
  • a small reflector 302 can be placed between the filter collimating lens 301 and the dispersion unit 303, so as to reflect the selected and retained reflected light returned from the original path to other directions for subsequent detection.
  • the selection of each component is described by taking the application of the filter in the confocal microscope as an example. Fluorescence is emitted from the small hole of the confocal microscope. In general, the size of the small hole is between a dozen micrometers to several tens of micrometers. In order to achieve better matching, the diameter of the fiber 211 can preferably be Within the above range, preferably the fiber diameter of the optical fiber 211 is slightly larger than the diameter of the small hole, so as to ensure that the light from the small hole can be received by the optical fiber 211 . In a specific embodiment, the optical fiber 211 may be a jumper MM62.5/125 of HannStar Optical Fiber Communication.
  • the filter collimating lens 301 is the preferred objective lens, and the NA of the objective lens must be greater than the NA of the fiber, so that the objective lens can well collect the beam emitted from the fiber.
  • the filter collimating lens 301 is a Nikon lens.
  • the present application uses a dispersion prism.
  • the first two parameters to be determined are the wavelength range and the Abbe number.
  • the applicable range of wavelength should match the fluorescence signal, and the Abbe number reflects the dispersion ability of the prism.
  • the smaller the Abbe number the greater the dispersion ability, that is, the more obvious the spectral effect.
  • the size of the prism also needs to be considered in the actual system design.
  • the collimating unit 304 can also select the same prism, that is, the first dispersing prism and the second dispersing prism of the dispersing unit 303 and the collimating unit 304 are Thorlabs' N-SF11 equilateral dispersing prism, the specific model for PS853.
  • DMD target size (diagonal length), resolution (corresponding to the number of micromirrors), spectral range, and efficiency should be considered.
  • DMDs with a diagonal length of 0.65 to 0.9 inches are acceptable.
  • the minimum standard of resolution is 1024*768.
  • the spectral range should match the optical signal to be filtered, for example, for a confocal microscope imaging in the visible range, the spectral range is 400nm to 700nm.
  • the digital micromirror device 305 may be TI's DLP discovery 4100 development kit (resolution 1024*768).
  • the reflector 302 is selected to have a higher reflectivity in the wavelength band of the signal to be filtered, and the shape may be, for example, a rectangle to facilitate the reception of the returned light.
  • GMEH-30*30-AL of Hengyang Optics can be selected.
  • the filter slit 306 may be a self-made slit, and may specifically be a rectangular filter slit whose shape matches the shape of the returned light.
  • the filter focusing lens 307 selects a double cemented achromatic lens with high transmittance corresponding to the waveband of the signal to be filtered.
  • the diameter of the lens is preferably slightly larger, such as 30 mm, which is beneficial to the reception of the back detector.
  • Hengyang Optical GLH31-030-050-VIS a spectrometer with a large detection range and high resolution can be selected, for example, a spectrometer from Princeton Instruments can be used as the detector 308 to check the filtering performance of the filter.
  • the detector 308 is the corresponding wavelength band PMT with high detection efficiency, such as the PMT of Beijing Hamamatsu Photonics (model CH345).
  • Fig. 6 it is the spectrum diagram of the experimental LED light source, wherein, Fig. 6a is the ideal spectrum of the LED light source, but the spectrogram of the LED light source in the actual experiment is shown in Fig. 6b, which is also the object of subsequent experiment filtering .
  • Optical signals such as 1: 400 ⁇ 500nm; 2: 500 ⁇ 600nm; 3: 600 ⁇ 700nm; Expanded.
  • the length of the dispersion fringes in the spreading direction will vary with the observation distance.
  • a biprism structure is formed. Therefore, when the composite light passes through the biprism, the stripes shown in FIG. 7 will be formed, and Since the dispersed optical signal is also collimated, the dispersion fringes will not change in the spreading direction.
  • the dispersion prism is used in this embodiment, the dispersion of the prism is non-linear. Therefore, the degree of dispersion of the optical signals in different wavelength bands is different, which is reflected in the figure, that is, the optical signals in different wavelength bands have along the direction of dispersion expansion. different lengths.
  • the grating is used as the dispersion unit, since the grating is linearly dispersed, the optical signals of different wavelength bands will have the same degree of dispersion, that is to say, the optical signals of different wavelength bands will have the same length along the direction of dispersion expansion.
  • dashed line 1, dashed line 2 and dashed line 3 respectively represent the demarcation line between two states in which the micromirror in the DMD can reflect the optical signal back to the detection optical path and cannot reflect the optical signal back to the detection optical path.
  • the above-mentioned dividing line preferably can reach the state of dotted line 2 in the figure, that is, the dividing line is perpendicular to the dispersion fringes, so that the optical signals of different wavelength bands can be accurately separated.
  • the dotted line 1 or dotted line 3 because it is not perpendicular to the stripes, in these two cases, the filtered light will be mixed with the light of other wavelength bands or the light of the desired wavelength band will be lost, which will cause impure filtering or filtering.
  • a series of problems such as low efficiency will eventually lead to the reduction of spectral resolution. Therefore, the pattern loaded into the DMD described later needs to be such that the demarcation line between the two states of the micromirror that can reflect the optical signal back to the detection optical path and cannot reflect the optical signal back to the detection optical path is perpendicular to the direction in which the optical signal is unfolded ( That is, the length direction of the dispersion fringes, or the spreading direction of the dispersion fringes).
  • the dotted line represents the horizontal plane
  • the dotted line represents the diagonal line of the DMD target surface.
  • the DMD needs to comply with the requirement that a certain diagonal of the target surface is parallel to the horizontal plane, so that the target surface of the DMD can be fully used. If the optical path is adjusted correctly, the dispersion fringes from the prism are also parallel to the horizontal plane. Therefore, the dispersion fringes are also parallel to the diagonal of the target. In general, it is sufficient to make the dispersion fringes coincide with a certain diagonal of the DMD target surface. In this way, the diagonal part with the longest dimension is used to correspond to the dispersion fringes, which can reduce the requirements for DMD. In order to form the state represented by the dotted line 2 in FIG.
  • a loading pattern hereinafter. Once a suitable loading pattern is found, it can be loaded into the DMD by the pattern loading unit in the control unit to meet specific filtering needs.
  • the DMD has a large number of pixels, so there will be many patterns that can be loaded on the DMD. In order to mark each pattern well, certain parameters are required to define these patterns. In the present invention, the parameters used are The angle and top length of the hypotenuse of the pattern. Referring to Fig. 9a, it is a pattern that can be loaded on the DMD, and the dotted line represents the diagonal line of the target surface. In this embodiment, the pattern can be understood as the black part. Black and white represent the two states of the DMD micromirror respectively. Black indicates that the DMD micromirror is deflected to -12 degrees, and white indicates that the DMD micromirror is deflected to 12 degrees.
  • the black area on the DMD target is set here to reflect the spectrum of the area back and be detected by the detector 308, while the white represents the spectrum of this area on the DMD target. is not reflected back to the detector 308, ie is filtered out.
  • the black part of the pattern has a hypotenuse, and the acute angle between this hypotenuse and the edge of the entire target surface is the parameter of the angle of the hypotenuse, and the length of the edge of the black pattern part on the target surface is the length of the upper edge. Two parameters can determine each pattern.
  • Figure 9b is a diagram illustrating the effect of loading the pattern of Figure 9a onto a DMD.
  • the rendering after loading to the DMD is equivalent to the original image in Figure 9a rotated clockwise by a certain angle.
  • the hypotenuse of the black part will be perpendicular to the diagonal of the target, that is, also perpendicular to the dispersion fringes, so the effect of the dotted line 2 in FIG. 7 can be achieved.
  • the angle of the hypotenuse is optimally between 40 and 50 degrees, different angles will have different spectral resolutions, and different upper edge lengths will correspond to different filtering bands.
  • 10a to 10c are examples of patterns with the same upper edge length of 510 pixels, but with different angles of the hypotenuse.
  • the angles of the hypotenuse are 40 degrees, 45 degrees, and 50 degrees, respectively.
  • 11a to 11c are examples of patterns in which the angle of the hypotenuse is the same as that of 42 degrees, but the lengths of the upper edge are 100 pixels, 200 pixels and 600 pixels, respectively. These patterns all correspond to different spectral resolutions and different filter bands, which means that these patterns have different filter performances after being loaded on the DMD.
  • the appropriate angle of the hypotenuse is determined according to the requirement of spectral resolution. Then, in the case of fixing the angle of the hypotenuse, change the parameter of the length of the upper edge to obtain a series of patterns, use these patterns as the basic patterns, and analyze the filtering bands of these basic patterns.
  • the above-mentioned changing of the length of the upper edge may be performed in steps of a fixed interval.
  • the step size of the interval can be 1 DMD pixel, so the basic pattern will include a fixed hypotenuse angle, the upper edge length is 1 DMD pixel, 2 DMD pixels, 3 DMD pixels, 4 DMD pixels... .
  • the step size of the interval can also be 10 DMD pixels, so the basic pattern will include the angle of the fixed hypotenuse, the upper edge length is 10 DMD pixels, 20 DMD pixels, 30 DMD pixels, 40 DMD pixels... . As the length of the upper edge increases, the black pattern will fill the entire DMD target surface.
  • the black pattern may take on two shapes, one is a triangle as shown in Figures 10 and 11, and the other is a trapezoid as shown in Figure 12, where the 12 is the case where the angle of the hypotenuse is 46 degrees and the length of the upper edge is 1000 pixels.
  • the length of some sides may be greater than the length of the target surface.
  • the left vertical side of the black triangle pattern in Figure 12 exceeds the left side of the target surface. long, so it has a trapezoid shape.
  • the following processing method is set in the pattern generation program: when the angle is determined, when the side length of the black area of the basic pattern is not greater than the side length of the target surface, the basic The growth rule of the black area of the pattern is that the above edge length is used as a reference to increase the fixed step size; but when a certain side length of the black area of the basic pattern is greater than a certain side length of the target surface, it is necessary to use the lower edge and the upper edge. Together as a reference to meet the pattern's growth law.
  • the length of the pattern on the lower edge of the DMD target surface is used as a reference to increase the fixed step size (for example, 10 pixels) .
  • the above basic patterns are generated by the pattern generation unit in the control unit through a pattern generation program.
  • the pattern generation program will generate many patterns. For example, the angle of the hypotenuse is 44 degrees, and the length of the upper edge is increased by 1 DMD pixel each time.
  • 1801 basic patterns of triangles or trapezoids can be generated. These 1801 patterns have the same angle, but the lengths of the upper and lower edges are different. It is incremented by 1 DMD pixel value.
  • these patterned pictures are loaded into the DMD in turn, and the spectral curve corresponding to each pattern can be measured, so that the collected spectral data corresponding to the basic pattern can be used as pre-collected data.
  • the above basic pattern is used to obtain a loading pattern that meets the specific filtering requirements, and the generation of the loading pattern is also realized by the pattern generation unit.
  • the generated loading pattern is loaded onto the DMD by the pattern loading unit of the control unit and tested with a spectrometer to confirm that the expected filtering effect can be achieved.
  • FIGs 13-18 in the description are schematic diagrams of experimental results of filtering the spectrum shown in Figure 6b using the optical filter of the present invention.
  • the figure on the left is a picture loaded on the digital micromirror device 305
  • the figure on the right is a corresponding spectrum after filtering by the filter.
  • FIG. 15 shows the band from 524nm to 600nm.
  • the two patterns on the left in Figure 15(c) are loaded on the DMD, and the spectra on the right in Figure 15c can be obtained respectively.
  • the required spectrum similar to that of Fig. 15(b) can be obtained by subtracting the upper spectrum from the lower spectrum on the right side of Fig. 15(c). Therefore, facing the required spectrum of Fig. 15(b), the The pattern on the lower left side of 15(c) is subtracted from the pattern on the upper left side to obtain the pattern corresponding to Fig. 15(a) as the loading pattern.
  • the spectral curve examples corresponding to Figures 13 and 17 are only the results of a single channel.
  • the single-channel patterns need to be merged into one image, and then loaded into the DMD to measure the spectral results. That is, after finding several single channels, select several single channels whose spectra do not overlap, and then combine the patterns of these single channels into one pattern as the loading pattern for the required multi-channel spectral filtering.
  • Figure 18 of the description shows how to easily realize five-channel filter modulation using the optical filter of the present invention.
  • the loading pattern is derived from the fusion of Figure 13, Figure 15, Figure 16 and Figure 17.
  • the picture can be loaded into the DMD.
  • the final loaded picture is a picture with the same pixels generated according to the number of pixels of the DMD.
  • the DMD used in this embodiment is 1024*768 pixels. , so it is necessary to use the software Matlab to generate a picture of 1024*768 pixels.
  • DMD is the DLP discovery 4100 development kit of Texas Instruments (TI). This type of DMD can use the software developed by TI to load images to the DMD. Of course, you can also use the software Labview to load images to the DMD.
  • the principles of the two loading methods are similar, because A complete DMD is composed of an FPGA and a DMD target surface, so the micromirrors of the DMD are controlled by the FPGA.
  • a series of basic patterns need to be generated according to the angle of the stripes and the lengths of multiple upper edges, these basic patterns are loaded onto the DMD in turn, and the spectrum corresponding to each pattern is measured at the same time.
  • the loading patterns corresponding to Fig. 13 to Fig. 17 are obtained, and then the spectrometer is used to obtain a single-channel schematic diagram with non-overlapping intervals.
  • the result of multi-channel is to fuse the loading pattern of single channel together, and then use the spectrometer to measure it.
  • the loading patterns corresponding to the single-channel can be added together to achieve multi-channel.
  • Figure 19 of the specification shows the spectral resolution of the filter in this embodiment.
  • the dispersive unit whether it is a dispersive prism or a grating, its dispersive ability is limited.
  • the Abbe number of the prism and the number of grating lines can reflect the size of their dispersive ability.
  • DMD's micromirrors also have sizes, not infinitely fine. Therefore, the two parameters of the dispersion capability of the dispersion unit and the size of the DMD micromirror limit the spectral resolution of the tunable filter, that is, the minimum wavelength difference that can distinguish two wavelengths.
  • the dispersion prism is non-linear dispersion, that is, the degree of dispersion in different wavelength bands is different in the whole working wavelength, so in order to evaluate the resolution of the optical filter in this embodiment, it is necessary to measure several resolutions in the working wavelength range. rate value.
  • rate value For example, in the embodiment of filtering the above-mentioned LED light source, four graphs a, b, c, and d in FIG. 19 are used to show the spectra of four different working wavelength bands of 440-460 nm, 510-550 nm, 580 nm, and 660 nm, respectively. Resolutions, 3.4131, 7.3065, 9.6063, and 11.6618, have good performance for most scenarios.
  • the beveled edges of the black areas may also be reversible.
  • Figure 20 shows a DMD with a pixel count of 1920*1080. The placement of the DMD is different from the DMD above, so different patterns can be used.
  • 20a to 20c respectively show the cases where the angle of the hypotenuse is 45 degrees, and the length of the upper edge is 300 pixels, 600 pixels and 1200 pixels.
  • the angle of the hypotenuse may also be other angles. In some cases, the angle of the hypotenuse may also be 90 degrees, as shown in Figures 21a-21c.
  • the optical filter of the present invention can easily adjust the filtering performance of the optical filter by manipulating the pattern loaded by the digital micromirror device 305, which can realize not only single-band filtering but also any multi-band filtering. Multi-channel filtering. Since the tuning speed of the digital micromirror device 305 is extremely fast, which can reach 10 kHz, or even several tens of kHz, the speed of the optical filter of the present invention will not affect the original imaging system such as confocal, so the above can be achieved. Imaging Systems Ultra-fast multispectral imaging. Moreover, the digital micromirror device 305 in the system is fixed during the imaging process and does not require any mechanical movement, so the problem that the actual wavelength is different from the calibrated wavelength due to insufficient mechanical movement precision in the prior art will not arise.
  • the optical filter it is also feasible to replace the prism of the dispersive unit with a grating.
  • the spectral energy of the grating is better than that of the prism, but since the grating is only efficient at the center wavelength, the efficiency at other wavelengths is the highest. Lower, so the energy efficiency is not as high as the technical solution using the prism.
  • the optical fiber 211 may not be used, and the fluorescence after passing through the small hole 209 is no longer coupled into the optical fiber. Transmission, but directly using space transmission, so it can also play the same function.
  • the present invention proposes an optical filter with an arbitrarily tunable working band, which can completely replace the optical filter or filter set in the prior art.
  • the key points include that because the digital micromirror device DMD can be easily controlled, the optical filter can realize the tuning of any central wavelength and (multi) bandwidth, and the speed is relatively fast; the design of the dispersion unit and the collimation unit is used to perform the dispersion light. Collimation, beams of different wavelengths travel parallel to each other in space, which can ensure that each wavelength or band can correspond to the pixels of the DMD, so that the distance of the DMD is not limited by space; the light reflected by the DMD will not be dispersed , that is, the beams of all wavelengths are concentrated together. Therefore, compared with the prior art, the optical filter and the optical filtering method of the present invention can significantly improve the imaging capability of various multispectral imaging systems such as confocal microscopes.

Abstract

本发明提出了一种工作波段任意可调谐的滤光器、滤光方法以及相应的多光谱成像系统，所述滤光器包括色散单元、用于将经过色散单元色散的入射光信号准直的准直单元、接收准直单元准直后的入射光信号并将部分光信号作为反射光信号反射回所述准直单元的数字微镜器件、将反射光信号反射至探测光路中的反射镜；其中，所述经过准直单元准直的入射光信号的光轴与所述数字微镜器件反射的反射光信号的光轴不重合。相对于现有技术来说，本发明的滤光器以及滤光方法不仅能够实现任意多波段多通道的滤波，而且滤波速度极快，不会出现由于机械运动精度不够而产生的实际波长与标定波长不同的问题，因此，能够大大提升多光谱成像系统的多光谱成像性能。

Description

一种滤光器、滤光方法及多光谱成像系统 技术领域
本发明涉及光学技术领域，尤其是一种滤光器、相应的滤光方法，以及使用该滤光器的多光谱成像系统。
背景技术
共聚焦显微镜是一种使用激光光束逐点扫描激发样品，并利用针孔对荧光信号进行空间滤波去除非焦平面杂散光的一种显微成像技术，也是一种常见的多光谱成像系统。其中“共聚焦”一词的含义是：光源处的小孔、样品中的焦点和探测器前面的小孔是相互共轭的。相比于宽场显微镜，共聚焦显微镜具有良好的光学分辨率和信噪比。一般而言，荧光的中心波长与激光的波长是不同的，为了可以获得正确、良好的荧光图像，需要利用滤光片去滤除激发光和杂散光。在很多情况下，观测生物组织样品中的不同结构成分是非常有意义的，但是不同结构成分发出的荧光在波长上也是不同的，因此需要采取某些方法保证探测器能接收到不同结构成分发出的正确荧光，即实现多色或多光谱成像。
实际上多色或多光谱成像并不是共聚焦显微镜独有的要求，除了共聚焦显微镜以外，其它类型的荧光显微镜以及多光谱成像仪等许多成像系统都需要分时或者同时处理多个不同波段的光信号。目前，实现包括共聚焦显微镜在内的此类设备多光谱成像的常规办法就是安装滤光轮、滤光片组等，即，需要使用滤光片来得到对应波段的荧光信号，但是光学滤光片的滤光波段都是固定的，为了能实现多光谱成像，必须安装多个滤光片或一个多带宽滤光片来匹配不同的荧光，这种方法不仅成本高、速度慢，而且光谱分辨率低、任意多波段调谐功能差，成像带宽也完全取决于滤光片的生产厂商。
以共聚焦显微镜为例，现有技术中也开发出了一些具有可调滤光功能的设备，商用的能实现可调滤光功能的共聚焦系统有Leica(莱卡)的TCS SP5、 Nikon(尼康)的C1si等。参见说明书附图1所示，是一种具有代表性的现有技术。该系统主要包括激发光模块A、共聚焦成像模块B以及多通道光谱成像模块C。
其中，激发光模块A中有不同波长的激光器、二向色镜1－5、反射镜6和声光可调谐滤波器7；共聚焦成像模块B包括二向色镜8、X－Y振镜9和10、扫描透镜11、筒镜12、物镜13、样品14、聚焦透镜15和小孔16；多通道光谱成像模块C由准直透镜17、分光棱镜18、第一、第二和第三聚焦透镜19、22和25、第一、第二和第三可变狭缝20、23和26以及第一、第二和第三探测器21、24和27组成。
该可调滤光器－共聚焦系统的工作原理如下：在激发光模块A中，波长不同的激光经过对应的二向色镜后整合在一条光路上，然后射入声光可调谐滤波器中。调节声光可调谐滤波器的相应参数可以改变出射光的波长和强度。在共聚焦成像模块B中，从A中射出的激光先被二向色镜8反射进入两个X－Y振镜后依次经过扫描透镜11、筒镜12和物镜13后，光束聚焦在样品14上。因为两个X－Y振镜可以改变激发光在样品上的位置，因此让聚焦光束扫描样品上的某个区域后可获得该区域的图像。荧光沿着相同的光路返回并被二向色镜8滤除激发光，最后被透镜15聚焦到小孔16上。小孔滤波后的荧光直接进入多通道光谱成像模块C。首先被准直透镜17准直后进入分光棱镜18中，分光棱镜色散荧光使其光谱在空间上展开，然后被第一聚焦透镜19聚焦成一个含有光谱信息的条纹。随后在条纹处安装一个第一可变狭缝20，狭缝的位置和间距是可变的，确定了狭缝的位置与波长的关系后系统才可以实现可调滤波。被滤除的荧光还可以被第一可变狭缝20上的反射镜反射到其余两个狭缝23和26上，再次滤波后进入第二和第三探测器24和27。
如上述所提到的具有可调滤光功能的设备，虽然改变狭缝的位置和间距可以改变透射荧光的中心波长和带宽，但是狭缝的移动是由电机驱动的，所以速度受限。而且机械装置的移动必然会带来精度问题，这会导致实际值与标定值之间有一定的误差。更重要的一点是由于空间大小问题，狭缝的数量 是有限的，所以通常最多只能达到三通道滤波。因此，上述现有技术实际上并不能实现任意可调滤光(任意中心波长、任意带宽)，这也影响了包括共聚焦显微镜在内的许多成像系统的多光谱成像能力。
发明内容
有鉴于此，为了克服上述现有技术的缺陷，本发明提出了一个工作波段任意可调谐的滤光器、滤光方法以及相应的多光谱成像系统，通过使用色散单元色散光信号，配合准直单元准直色散后的光信号，并结合控制数字微镜器件DMD来实现滤光技术中任意中心波长、(多)带宽的调谐。从而，相对于现有技术来说，本发明的滤光器以及滤光方法能够显著提高以共聚焦显微镜为例的多种多光谱成像系统的成像能力。
具体地，本发明的滤光器包括用于使包含多个波段的入射光信号进行色散的色散单元；以及准直单元，用于将经过色散单元色散的入射光信号准直；数字微镜器件，其接收准直单元准直后的入射光信号；所述数字微镜器件经过控制单元的控制将所述入射光信号中的部分光信号作为反射光信号反射回所述准直单元；反射镜，所述反射光信号经过所述准直单元和所述色散单元之后入射到所述反射镜上，并被反射镜反射到探测光路中；其中，所述经过准直单元准直的入射光信号的光轴与所述数字微镜器件反射的反射光信号的光轴不重合。
其中，所述控制单元包括图案生成单元和图案加载单元；所述图案生成单元能够生成加载图案，所述加载图案由一个或多个基础图案组合而成；所述图案加载单元将所述加载图案加载到所述数字微镜器件中，从而实现将所述入射光信号中的部分光信号作为反射光信号反射回所述准直单元。
具体地，所述图案生成单元能够生成多个所述基础图案，所述基础图案由该图案中斜边与所述数字微镜器件靶面上边缘之间的角度，以及该图案于所述数字微镜器件靶面上边缘和/或下边缘的长度来确定；所述多个基础图案具有相同的所述角度，并且具有不同的所述长度。
优选地，所述色散单元是第一色散棱镜，所述准直单元是第二色散棱镜；其中，所述第一色散棱镜的光出射面平行于所述第二色散棱镜的光入射面。进一步，所述第一色散棱镜与所述第二色散棱镜结构相同。从而，通过双棱镜结构来实现色散单元和准直单元的功能。
相应地，本发明还提出一种滤光方法，其包括如下步骤：将包含多个波段的入射光信号进行色散；对经过色散的入射光信号进行准直；使用数字微镜器件接收所述准直后的入射光信号；控制所述数字微镜器件以将所述入射光信号中的部分光信号作为反射光信号反射；使用反射镜将所述反射光信号反射到探测光路中；其中，所述经过准直的入射光信号的光轴与所述数字微镜器件反射的反射光信号的光轴不重合。
所述控制所述数字微镜器件以将所述入射光信号中的部分光信号作为反射光信号反射的操作具体是通过将加载图案加载到所述数字微镜器件中来实现的；其中所述加载图案通过如下步骤生成：
S1、构建多个基础图案；
S2、将所述多个基础图案分别加载到所述数字微镜器件中，获得每个基础图案对应的光谱数据，并将这些光谱数据作为预采集数据；
S3、根据滤光的需求和所述预采集数据，选择一个或多个所述基础图案组合形成所述加载图案。
进一步，所述基础图案由该图案中斜边与所述数字微镜器件靶面上边缘之间的角度，以及该图案于所述数字微镜器件靶面上边缘和/或下边缘的长度来确定；所述步骤S1中的多个基础图案具有相同的所述角度，并且具有不同的所述长度。
优选地，使用第一色散棱镜对所述包含多个波段的入射光信号进行色散，使用第二色散棱镜对经过色散的入射光信号进行准直。所述第一色散棱镜与所述第二色散棱镜结构可以相同。从而，通过使用双棱镜结构来实现色散和准直的功能。
本发明还提出了一种多光谱成像系统，其包括上述滤光器，以及光信号生成单元。所述光信号生成单元例如可以是共聚焦显微镜。其中，所述共聚焦显微镜与所述滤光器之间通过光纤传输或者空间传输的方式传输光信号。
本发明的滤光器、滤光方法以及相应的多光谱成像系统可以很方便的通过操控数字微镜器件DMD加载的图案，来调节滤光器的滤光性能，不仅能够实现单一波段的滤波还能实现任意多波段多通道的滤波。由于数字微镜器件DMD的调谐速度极快，可以到达10kHz，甚至几十kHz，因此本发明的滤光器的速度不会对原有的共聚焦等成像系统造成影响，应此可以实现所述成像系统速度极快的多光谱成像。而且，系统中的数字微镜器件DMD在成像过程中是固定的，无需任何机械运动，所以不会产生现有技术中由于机械运动精度不够而产生的实际波长与标定波长不同的问题。从而，本发明的技术方案能够大大提升多光谱成像系统的多光谱成像性能。
附图说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为现有技术中具有可调滤光功能的共聚焦显微镜结构示意图；
图2为本发明共聚焦显微镜系统结构示意图；
图3为本发明滤光器结构示意图；
图4为本发明第一和第二色散棱镜准直色散光束的原理示意图；
图5a为本发明滤光器荧光光束分离示意图；
图5b为本发明DMD控制单元示意图；
图6为本发明实验LED光源的光谱图；
图7为本发明色散条纹示意图；
图8为本发明DMD使用状态图；
图9为本发明DMD图案示意图；
图10为本发明一系列基础图案示意图；
图11为本发明另一系列基础图案示意图；
图12为本发明另一种基础图案示意图；
图13为本发明对LED光源光谱进行滤光的实验结果一；
图14为本发明对LED光源光谱进行滤光的实验结果二；
图15为本发明对LED光源光谱进行滤光的实验结果三；
图16为本发明对LED光源光谱进行滤光的实验结果四；
图17为本发明对LED光源光谱进行滤光的实验结果五；
图18为本发明对LED光源光谱进行滤光的实验结果六；
图19为本发明对LED光源光谱进行滤光的分别率测试图；
图20为本发明方向不同的一系列基础图案示意图；
图21为本发明方向不同的另一系列基础图案示意图；
图22为本发明共聚焦显微镜系统另一实施例结构示意图。
附图标记：
现有技术附图：1－5－二向色镜，6－反射镜，7－声光可调谐滤波器，8－二向色镜，9－X－Y振镜，10－X－Y振镜，11－扫描透镜，12－筒镜，13－物镜，14－样品，15－聚焦透镜，16－小孔，17－准直透镜，18－分光棱镜，19－第一聚焦透镜，20－第一可变狭缝，21－第一探测器，22－第二聚焦透镜，23－第二可变狭缝，24－第二探测器，25－第三聚焦透镜，26－第三可变狭缝，27－第三探测器。
实施例附图：201－激光器，202－准直透镜，203－二向色镜，204－振镜，205－扫描透镜，206－筒镜，207－物镜，208－样品，209－聚焦透镜，210－小孔，211－光纤，212－滤光器，301－滤光器准直透镜，302－反射镜，303－第一色散棱镜，304－第二色散棱镜，305－数字微镜器件，306－滤光狭缝，307－滤光器聚焦透镜，308－探测器。
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
在本具体实施方式中，以下以共聚焦显微镜为例，来说明本发明的滤光器、滤光方法以及相应的成像系统。其中共聚焦显微镜是一种常规的多光谱成像系统，其一般包括激光器、小孔、振镜、物镜、二向色镜以及准直透镜，例如参见说明书附图1所示的系统结构，其能够生成待测样品的荧光信号，在后文中，该部分也被称为光信号生成单元，以和滤光器进行区分。在上述荧光信号生成部分或者光信号生成单元中，从激光器出来的光束经扩束后分别到达二向色镜、振镜、扫描透镜、筒镜和物镜。激发产生的荧光原路返回经过物镜、筒镜、扫描透镜、振镜后被二向色镜滤除激光和一些杂散光，然后被小孔滤波后耦合进入光纤。荧光信号在光纤中传输，而光纤的另一端则是滤光器。本发明的滤光器例如包括光纤入射端口、滤光器准直透镜、色散单元(例如由第一色散棱镜构成)、准直单元(例如由第二色散棱镜构成)、一个数字微镜器件(DMD)、反射镜、狭缝、聚焦透镜以及探测器。滤光器准直透镜例如光纤准直镜把光纤端口处的荧光信号准直后射向色散单元，不同波长的荧光信号在空间中平行展开，色散展开的荧光信号经过准直单元后照射在DMD上，通过调整DMD靶面上镜片的翻转角度，可以将特定波长的荧光信号反射偏离原光路并被探测器接收。
首先，以共聚焦显微镜为例说明光信号生成单元能够生成怎样的光信号。参见说明书附图2所示，为本发明共聚焦显微镜系统结构示意图。其中，激光器201、准直透镜202、二向色镜203、振镜204、扫描透镜205、筒镜206、物镜207、聚焦透镜209和小孔210组成一个激光扫描共聚焦显微镜。共聚焦显微镜经过光纤211和滤光器212相连。
激光扫描共聚焦显微镜的光路结构为：激光从激光器201出射，经过准直透镜202后被准直成光斑大小合适的激光光束，选择对应的二向色镜203 后使得光束可以从203透射过去。从二向色镜203透射过去的准直光束依次经过振镜204、扫描透镜205、筒镜206和物镜207，最后光束聚焦在样品208上。激发样品208产生的荧光原路返回并被二向色镜203反射到聚焦透镜209上，聚焦透镜209将光束聚焦在小孔210上，小孔210对荧光进行空间滤波，滤除焦点之外的杂散光。
上述滤除了杂散光的荧光经过光纤211传输到滤光器212中。本发明的滤光器212的结构参见说明书附图3所示。滤光器212包括滤光器准直透镜301、反射镜302、色散单元303、准直单元304、数字微镜器件(DMD)305、滤光狭缝306、聚焦透镜307和探测器308。其中，色散单元303的作用为将不同波段的光信号在空间中展开，而准直单元304的作用为对色散后的光信号进行准直，保证后续滤光过程的实现，此部分在后续再进行阐述。优选地，在本实施例中，所述色散单元303例如为第一色散棱镜，准直单元304例如为第二色散棱镜，其中第一色散棱镜和第二色散棱镜可以为相同的色散棱镜。但是本领域技术人员可以知晓，只要能实现上述功能的变形都属于本发明的范围。
滤光器212的基本光路结构为：经过小孔210滤波后的荧光信号被耦合进了光纤211中传输，当荧光从光纤211另一端出来后被滤光器准直透镜301准直，然后以一个特定的入射角入射到第一色散棱镜，即色散单元303上，第一色散棱镜会色散光信号并空间展开其光谱。光谱展开的色散光再以一个特定的入射角入射到第二色散棱镜，即准直单元304上，第二色散棱镜会对色散光进行准直，准直后的色散光最终照射在数字微镜器件(DMD)305上。通过设定，需要保留的波段对应的光信号经DMD反射后，光束稍微偏离原光路返回，即再次经过第二色散棱镜、第一色散棱镜后照射在反射镜302上，反射镜302改变光束方向使其被滤光器聚焦透镜307聚焦。在滤光器聚焦透镜307前面还可以放置一个固定的狭缝306用于滤除光路中的杂散光。最终，需要的荧光信号被探测器308接收。
说明书附图4展示了色散单元303(第一色散棱镜)和准直单元304(第二色散棱镜)准直色散光束的原理示意图。因为色散棱镜一般在最小偏向角下使用时可以实现最大的通光孔径，所以优选经过滤光器准直透镜301准直后的荧光以最小偏向角入射到第一色散棱镜中。棱镜的摆放如说明书附图4所示，第一色散棱镜和第二色散棱镜需要平行放置，即第一色散棱镜的出光面与第二色散棱镜的入光面两者平行。准直光束经过第一色散棱镜后被光谱展开并形成一个色散条纹。
色散条纹经过第二色散棱镜后成为一束准直的色散光，即条纹在展开方向上的长度不会因观察距离而变化，实现了不同波长成分的光束在空间位置上虽然不同但却相互平行。准直的色散光可以保证光束原路返回后也是一束准直光，不会再被色散展开。这里要保证经过第二色散棱镜准直后的色散光光斑大小不能超过数字微镜器件(DMD)305的靶面尺寸，这样才能保证色散光的每个波段都可以被数字微镜器件(DMD)305进行调制。
进一步，参见说明书附图5a所示，显示了如何将原路返回的光束与入射光分离。数字微镜器件(DMD)305可以经如说明书附图5b所示的控制单元通过加载特定的图案从而将需要保留的波段所对应的光束反射回去，具体图案的生成和加载参见后文。在正常情况下如果入射光垂直入射DMD中的各个微镜，则反射光会与入射光重合，因而不容易准确挑选出经过DMD选择的波段，所以需要对某些器件做一些微小的调整使得反射光可以与入射光分离。在本实施例中采用的方法是微调DMD305的角度，例如可以在垂直方向和/或水平方向微调DMD305中特定波段(选择保留的波段)所对应的微镜的反射方向，使得DMD部分所选择保留的反射光与入射光有一个很小的偏差，这将导致这部分原路返回的反射光与滤光器准直透镜301的光轴不重合。因此，可以在滤光器准直透镜301与色散单元303之间放置一个很小的反射镜302，从而将所选择保留的原路返回的反射光反射到其他的方向，以便于后续的探测。
在以下的具体实施例中，以滤光器应用在共聚焦显微镜中为例说明各组成部分的选用情况。荧光从共聚焦显微镜的小孔出射，一般情况下，该小孔 的尺寸在十几个微米到几十个微米之间，为了实现更好的匹配，光纤211纤径的大小最好也可以在上述这个范围之内，优选地光纤211纤径略大于小孔的直径，这样可以保证小孔过来的光都可以被光纤211接收。在具体实施例中，光纤211具体可以为瀚宇光纤通信的跳线MM62.5/125。
因为单个透镜并不能准直光纤出射的光束，因此滤光器准直透镜301优先选择物镜，其中还要保证物镜的NA大于光纤的NA，这样物镜才可以很好的收集到光纤出射的光束，在本申请中，滤光器准直透镜301为尼康的透镜。
关于色散单元303，本申请使用色散棱镜。首先要确定的两个参数就是波长适用范围和阿贝数。波长适用范围要与荧光信号匹配，而阿贝数体现的是棱镜的色散能力，阿贝数越小，色散能力越大，即分光效果越明显。对于在可见光范围内成像的共聚焦显微镜来说，需要选取波长范围400nm￣700nm的色散棱镜以匹配相应的荧光信号，然后再选择阿贝数较低的棱镜。此外，在实际系统设计中还需要考虑棱镜的尺寸，尺寸太大则不容易选配合适的棱镜架，尺寸太小则会导致入射光不能完全照在棱镜上。在本实施例中，准直单元304也可以选择相同的棱镜，即，色散单元303和准直单元304的第一色散棱镜和第二色散棱镜为Thorlabs的N－SF11等边色散棱镜，具体型号为PS853。
关于数字微镜器件305的选择，应当考虑DMD靶面大小(对角线长度)、分辨率(对应微镜的数量)、光谱范围以及效率。一般情况下，兼顾价格成本的考虑，对角线长度在0.65￣0.9英寸的DMD都是可以的，靶面更大的DMD当然更能满足要求，不过价格也会相应的增加。分辨率的最基低标准就是1024＊768，当然也可以选择1920＊1080的DMD，只不过后者的价格会比前者贵。光谱范围要与待滤波的光信号匹配，例如对于可见光区间成像的共聚焦显微镜来说，光谱范围为400nm￣700nm。因为DMD是许多微镜组合在一起的，类似于一个二维光栅，会产生多缝衍射效应，所以需要选取衍射效率较高的DMD。在具体实施例中，数字微镜器件305可以为TI的DLP discovery 4100 development kit(分辨率1024＊768)。
此外，反射镜302选择在对应待滤波信号波段具有较高反射率的反射镜，形状例如可以是矩形以有利于返回光的接收。在具体实施例中，可以选择恒洋光学的GMEH－30＊30－AL。滤光狭缝306可以是自制的狭缝，具体可以为形状与返回光形状匹配的矩形滤光狭缝。滤光器聚焦透镜307选择在对应待滤波信号波段具有较高透射率的双胶合消色差透镜，透镜直径优选稍微大一点，如30毫米，如此有利于后面探测器的接收，具体例如为恒洋光学的GLH31－030－050－VIS。在验证或检验模式中可以选择探测范围大、分辨率高的光谱仪，例如Princeton Instruments的光谱仪作为探测器308来查看滤光器的滤光性能，而在具体成像模式中，探测器308为相应波段探测效率高的PMT，例如北京滨松光子的PMT(型号为CH345)。
为了展示本发明滤光器的具体效果，并进一步阐述本发明的滤光器和滤光方法，以对LED光源进行滤光为例做进一步说明。参见说明书附图6所示，为实验LED光源的光谱图，其中，图6a为LED光源的理想光谱，但实际实验中LED光源的光谱图为图6b所示，其也是后续实验滤光的对象。
对于LED这样的复合光源，其所发出的复合光经过色散单元后不同波段的光信号将沿着一个方向展开，如图7所示，其中①、②、③、④示意性地代表不同波段的光信号，例如①：400￣500nm；②：500￣600nm；③：600￣700nm；④：700￣750nm，因此在视觉上可以看到不同波段的光信号占据不同的位置，即在一个方向上实现了展开。
如果复合光只经过色散单元的话，例如只经过单个棱镜，因为不同波段的光信号具有不同的出射角，因此色散条纹在展开方向上的长度会随着观察距离的不同而发生变化。但在本发明的实施例中，色散单元后还有准直单元，当两者都使用棱镜的情况下构成双棱镜结构，因此当复合光经过双棱镜后会形成图7所示的条纹，并且由于色散后的光信号还经过准直，因此色散条纹在展开方向上不会发生变化。
此外，由于本实施例中使用了色散棱镜，棱镜的色散具有非线性的特点，因此，不同波段的光信号的色散程度不同，反应在图中即不同波段的光信号 沿着色散展开的方向具有不同的长度。但如果使用光栅作为色散单元，由于光栅是线性色散的，因此不同波段的光信号将有相同的色散程度，也就是说不同波段的光信号沿着色散展开的方向会具有相同的长度。
对于滤波器来说，最好的滤波效果就是能够精确的分离出不同波段的光信号。参见图7所示，虚线1、虚线2和虚线3分别代表不同情况下，DMD中微镜具有能够将光信号反射回探测光路和不能将光信号反射回探测光路两种状态的分界线。为了得到最优的滤光效果，上述分界线最好可以达到图中虚线2的状态，即分界线与色散条纹是垂直的，如此便可以精确地分离不同波段的光信号。但是虚线1或者虚线3，由于其与条纹不垂直，因此在这两种情况下滤出来的光会夹杂着其他波段的光或者丢失需要的波段的光，这将造成滤光不纯或滤光效率低等一系列的问题，最后还会导致光谱分辨率的降低。因此，在后文描述的加载到DMD中的图案需要使得所述能够将光信号反射回探测光路和不能将光信号反射回探测光路两种微镜状态的分界线垂直于光信号展开的方向(即色散条纹的长度方向，或者色散条纹的展开方向)。
参见图8所示，虚线代表水平面，点线代表DMD靶面的对角线。在本实施例中，DMD是需要遵循靶面的某个对角线与水平面是平行的要求，如此，能够充分地使用DMD的靶面。如果光路调试正确的话，棱镜出来的色散条纹也是和水平面平行的，因此，色散条纹与靶面对角线也是成平行关系的。在一般情况下，使色散条纹与DMD靶面的某条对角线重合即可，这样使用尺寸最长的对角线部分来对应色散条纹，可以降低对DMD的要求。为了形成图7中所述虚线2代表的状态，需要找到特定的加载到DMD上的图案，下文可称为加载图案。一旦找到了合适的加载图案，便可通过控制单元中的图案加载单元将其加载到DMD中以满足特定的滤波需求。
DMD的像素数较多，所以能够加载到DMD上面的图案也会是很多的，为了能很好的标记每一张图案，需要某些参数来限定这些图案，在本发明中，使用的参数是图案斜边的角度和上边长长度。参见图9a所示，为一个能够加载到DMD上的图案，虚点线表示靶面的对角线。本实施例中，图案可以理解为 黑色的部分。黑色和白色分别代表的是DMD微镜的两种状态，黑色表示DMD的微镜偏转到－12度，白色表示DMD的微镜偏转到12度。因为色散条纹是打在DMD上面的，所以这里设定DMD靶面上的黑色区域是可以将该区域的光谱反射回去并被探测器308探测到的，而白色表示DMD靶面上该区域的光谱是无法反射回到探测器308的，即被滤除了。具体地，图案中黑色部分有一个斜边，这个斜边与整个靶面上边缘的锐角夹角就是斜边的角度这个参数，黑色图案部分在靶面上边缘的长度就是上边缘长度，使用这两个参数就可以确定每一个图案。图9b是示意了将图9a的图案加载到DMD上的效果。加载到DMD上后的效果图相当于图9a原图像顺时针旋转了一定的角度。通过改变角度，黑色部分的斜边将可以垂直于靶面对角线，即也垂直于色散条纹，因而可以实现图7中所述虚线2的效果。
通过改变黑色图案斜边的角度和上边缘长度就可以实现不同的滤波效果。因此，要实现特定波段的滤波就需要确定图案的上述两个参数。在本发明中，斜边的角度最优为40￣50度之间，不同的角度将具有不同的光谱分辨率，而不同的上边缘长度将对应不同的滤出波段。图10a至图10c为上边缘长度相同为510个像素，但是斜边的角度不同的图案示例，斜边的角度分别为40度、45度和50度。图11a至图11c为斜边的角度相同都为42度，但是上边缘长度分别为100个像素、200个像素和600个像素的图案示例。这些图案都对应有不同的光谱分辨率和不同的滤光波段，也就是说这些图案加载到DMD上后具有不同的滤波性能。
在本实施例中，通过对光谱分辨率的要求确定合适的斜边的角度。然后，在固定斜边的角度情况下，改变上边缘长度这个参数，获得一系列的图案，将这些图案作为基础图案，并对这些基础图案的滤波波段进行分析。上述改变上边缘长度，可以是以固定间隔的步长进行改变。例如间隔的步长可以为1个DMD像素，因而基础图案会包括固定斜边的角度下，上边缘长度为1个DMD像素、2个DMD像素、3个DMD像素、4个DMD像素……。间隔的步长也可以为10个DMD像素，因而基础图案会包括固定斜边的角度下，上边缘长度为10 个DMD像素、20个DMD像素、30个DMD像素、40个DMD像素……。随着上边缘长度的增加，黑色图案将会充满整个DMD靶面。
可以预见，随着上边缘长度的增长，黑色图案有可能呈现出两种形状，一种就是如图10和11中所示的三角形，另外一种为如图12所示的梯形，其中，图12为斜边的角度为46度，而上边缘长度为1000个像素的情况。出现这种情况是因为在特定的角度和上边缘长度的情况下，可能会出现某些边的长度大于靶面的长度，例如图12中黑色三角形图案的左侧竖边超过了靶面的左边长，因而呈现出梯形。
为了保证黑色区域可以按照规律填充满整个靶面，在图案生成程序中设置如下的处理方法：在角度确定时，当基础图案的黑色区域的边长没有大于靶面的边长时，所述基础图案黑色区域的增长规律是以上边缘长度为参考进行固定步长的增加；但当所述基础图案黑色区域的某个边长大于靶面的某个边长时，就需要用下边缘和上边缘一起作为参考来满足图案的增长规律。在本实施例中，例如当出现黑色三角形图案的左侧竖边超过了靶面的左边长时，以图案在DMD靶面下边缘的长度为参考进行固定步长(例如10个像素)的增加。
上述这些基础图案的生成是由控制单元中的图案生成单元通过图案生成程序来获得的，设置好相关参数后，所述图案生成程序就会生成很多图案。例如以斜边的角度为44度、上边缘长度每次增加1个DMD像素为例，最后可以生成1801个三角形或者梯形的基础图案，这1801个图案角度相同，但上边缘或下边缘长度都是以1个DMD像素值递增的。获得这些基础图案后，依次把这些图案化的图片加载到DMD中，便可以测出每一个图案所对应的光谱曲线，从而将这些采集到的对应基础图案的光谱数据作为预采集数据。随后，通过分析及特定的算法，根据上述预采集数据，使用上述基础图案获得满足特定滤波需求的加载图案，加载图案的生成也是通过图案生成单元来实现的。最后，通过控制单元的图案加载单元将所述生成好的加载图案加载到DMD上并利用光谱仪进行测试，以确认可以达到预期的滤波效果。
说明书附图13－18为使用本发明的滤光器对如图6b所示的光谱进行滤光的实验结果示意图。在上述实验结果示意图中，左边附图为在数字微镜器件305加载的图片，对应地，右边附图为相应的经过滤光器滤光后的光谱图。通过说明书附图13－18可以看出，通过为数字微镜器件305加载不同的图片，能够方便地过滤留下不同波段的光谱信号。以图15为例，图15显示的是524nm￣600nm这个波段。其中，给DMD加载如图15(c)中左侧的两个图案，分别可以得到图15c右边的光谱。经过计算可知，图15(c)右侧下面的光谱减上面的光谱就可以得到所需求的与图15(b)差不多的光谱，所以面对所需求的图15(b)的光谱，将图15(c)左侧下面的图案减去左侧上面的图案，即得到图15(a)对应的图案作为加载图案。
图13￣图17对应的光谱曲线示例只是单通道的结果，多通道的情况则需要把单通道的图案融合在一个图片里，然后加载到DMD并测出其光谱结果。即当找到几个单通道后，选取几个光谱没有重叠的单通道，然后把这些个单通道的图案融合在一幅图案里，作为所需求的多通道光谱滤波的加载图案。说明书附图18展示了，使用本发明的滤光器如何简便地实现五通道的滤波调制，该加载图案来源于图13、图15、图16和图17的融合，当然为了获得较好的结果，会对图案做一些细微的调整，从而可以达到图18(b)的结果，该结果可以从入射光谱图中选择424￣450nm、450￣500nm、525￣600nm、600￣670nm以及660nm￣747nm五个波段。为了防止通道交叠，在做单通道实验时就需要保证每个通道之间是不会有重叠的。此外，单通道的数量和带宽是可以改变的，在可见光区域中，可以设置5个不重叠的单通道。
在获得所需要的图片后，便可以将图片加载到DMD中，所述最终加载的图片是根据DMD的像素数生成的像素相同的图片，例如本实施例中使用的DMD是1024＊768像素的，因此需要用软件Matlab生成1024＊768像素的图片。DMD为德州仪器(TI)的DLP discovery 4100 developmentkit，该型号的DMD可以使用TI自己开发的软件给DMD加载图片，当然也可以使用软件Labview给DMD加载图片，两种加载方式的原理都差不多，因为完整的DMD是由FPGA 和DMD靶面组成的，所以都是通过FPGA来控制DMD的微镜。利用DMD的控制软件给DMD载入图片时需要将预载入的图片放在一个文件夹中，操作该软件并从电脑中找到需要的图片先加载到FPGA的缓存中，然后再运行FPGA，FPGA依据某个传输协议先将图片转化为DMD可以接受的数字信号，最后DMD依据传输过来的数字信号来改变每个微镜的状态。如果需要加载另一张图片的时，需要操作软件先将暂停FPGA和DMD之间的通信，然后再清空FPGA的缓存并加载下一张图片，最后就是运行FPGA控制DMD。
综上，首先需要根据条纹的角度和多个上边缘长度生成一系列的基础图案，将这些基础图案依次加载到DMD上，同时测出每幅图案对应的光谱。根据上述步骤所获得的数据得到图13￣图17对应的加载图案，再利用光谱仪得到区间不重叠的单通道示意图。多通道的结果就是把单通道的加载图案融合在一起后，再利用光谱仪测出来。简单的说，只要得到不重叠的单通道结果，就可以将单通道对应的加载图案加在一起然后实现多通道。
说明书附图19展示了本实施例中滤光器的光谱分辨率。对于色散单元来说，无论是色散棱镜还是光栅，其色散能力都是有极限的，棱镜的阿贝数、光栅的刻线数都能够反映各自色散能力的大小。而且，DMD的微镜也是有尺寸大小的，并不是无限精细。因此，色散单元的色散能力以及DMD微镜的尺寸大小这两个参数就限制了可调滤光器的光谱分辨率，也就是能分辨开两个波长的最小波长差。此外，因为色散棱镜是非线性色散的，即在整个工作波长内的不同波段色散程度是不一样的，所以为了评估本实施例中滤光器的分辨率，需要在工作波段范围内测几个分辨率值。例如，在对上述LED光源进行滤光的实施例中，分别使用图19中的a、b、c、d四幅图来展示440－460nm、510－550nm、580nm、660nm四个不同工作波段的光谱分辨率，分别为3.4131、7.3065、9.6063和11.6618，具有能够适用大多数场景的良好性能。
在其他的实施例中，黑色区域的斜角边也是可以改变方向的。图20展示的就是像素数为1920＊1080的DMD，该DMD的摆放方式与上文中的DMD不一样，所以可以采用不同的图案形式。图20a至图20c分别为斜边的角度为45度， 上边缘长度为300个像素、600个像素和1200个像素的情况。同样的，斜边的角度也可以是其它角度。在一些情况下，该斜边的角度也可以为90度，如图21a至图21c所示。
通过上述实验可见，使用本发明的滤光器可以很方便的通过操控数字微镜器件305加载的图案，来调节滤光器的滤光性能，不仅能够实现单一波段的滤波还能实现任意多波段多通道的滤波。由于数字微镜器件305的调谐速度极快，可以到达10kHz，甚至几十kHz，因此本发明的滤光器的速度不会对原有的共聚焦等成像系统造成影响，应此可以实现所述成像系统速度极快的多光谱成像。而且，系统中的数字微镜器件305在成像过程中是固定的，无需任何机械运动，所以不会产生现有技术中由于机械运动精度不够而产生的实际波长与标定波长不同的问题。
除了以上实施例以外，还可以具有不同的变形例。例如，在滤光器中，将色散单元的棱镜换成光栅也是可行的，光栅的分光能量是要优于棱镜的，不过由于光栅只有在中心波长上效率才是最高的，其他波长下的效率较低，因而能量利用率没有使用棱镜的技术方案高。
此外，在将滤光器用于多光谱成像系统时，例如用于共聚焦显微镜时，参见说明书附图22所示，也可以不使用光纤211，经过小孔209后的荧光不再耦合进光纤中传输，而是直接采用空间传输，如此也可以起到一样的功能。
综上所述，本发明提出了一种工作波段任意可调谐的滤光器，其完全可以替换现有技术中的滤光片或者滤光片组。重点包括由于可以方便地控制数字微镜器件DMD，因此滤光器可以实现任意中心波长、(多)带宽的调谐，并且速度较快；采用色散单元配合准直单元的设计，对色散光进行了准直，不同波长的光束在空间上相互平行传播，从而可以保证每个波长或波段都可以与DMD的像素对应，使DMD的摆放距离不受空间的限制；DMD反射回去的光不会色散，即所有波长的光束都聚集在一起。从而，相对于现有技术来说，本发明的滤光器以及滤光方法能够显著提高以共聚焦显微镜为例的多种多光谱成像系统的成像能力。
以上所述仅为本发明的较佳实施例，并不用以限制本发明，凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。

Claims (13)

  1. 一种滤光器，包括用于使包含多个波段的入射光信号进行色散的色散单元，其特征在于，还包括：
    准直单元，用于将经过色散单元色散的入射光信号准直；
    数字微镜器件，接收所述准直单元准直后的入射光信号；所述数字微镜器件经过控制单元的控制将所述入射光信号中的部分光信号作为反射光信号反射回所述准直单元；
    反射镜，所述反射光信号经过所述准直单元和所述色散单元之后入射到所述反射镜上，并被反射镜反射到探测光路中；
    其中，所述经过准直单元准直的入射光信号的光轴与所述数字微镜器件反射的反射光信号的光轴不重合。
  2. 根据权利要求1所述的滤光器，其特征在于，
    所述控制单元包括图案生成单元和图案加载单元；
    所述图案生成单元能够生成加载图案，所述加载图案由一个或多个基础图案组合而成；
    所述图案加载单元将所述加载图案加载到所述数字微镜器件中，从而实现将所述入射光信号中的部分光信号作为反射光信号反射回所述准直单元。
  3. 根据权利要求2所述的滤波器，其特征在于，
    所述图案生成单元能够生成多个所述基础图案，所述基础图案由该图案中斜边与所述数字微镜器件靶面上边缘之间的角度，以及该图案于所述数字微镜器件靶面上边缘和/或下边缘的长度来确定；
    所述多个基础图案具有相同的所述角度，并且具有不同的所述长度。
  4. 根据权利要求1－3任一项所述的滤光器，其特征在于，
    所述色散单元是第一色散棱镜，所述准直单元是第二色散棱镜；其中，所述第一色散棱镜的光出射面平行于所述第二色散棱镜的光入射面。
  5. 根据权利要求4所述的滤光器，其特征在于，
    所述第一色散棱镜与所述第二色散棱镜结构相同。
  6. 一种滤光方法，将包含多个波段的入射光信号进行色散，其特征在于，还包括如下步骤：
    对经过色散的入射光信号进行准直；
    使用数字微镜器件接收所述准直后的入射光信号；控制所述数字微镜器件以将所述入射光信号中的部分光信号作为反射光信号反射；
    使用反射镜将所述反射光信号反射到探测光路中；
    其中，所述经过准直的入射光信号的光轴与所述数字微镜器件反射的反射光信号的光轴不重合。
  7. 根据权利要求6所述的滤光方法，其特征在于，
    所述控制所述数字微镜器件以将所述入射光信号中的部分光信号作为反射光信号反射的操作具体是通过将加载图案加载到所述数字微镜器件中来实现的；
    其中所述加载图案通过如下步骤生成：
    S1、构建多个基础图案；
    S2、将所述多个基础图案分别加载到所述数字微镜器件中，获得每个基础图案对应的光谱数据，并将这些光谱数据作为预采集数据；
    S3、根据滤光的需求和所述预采集数据，选择一个或多个所述基础图案组合形成所述加载图案。
  8. 根据权利要求7所述的滤波器，其特征在于，
    所述基础图案由该图案中斜边与所述数字微镜器件靶面上边缘之间的角度，以及该图案于所述数字微镜器件靶面上边缘和/或下边缘的长度来确定；
    所述步骤S1中的多个基础图案具有相同的所述角度，并且具有不同的所述长度。
  9. 根据权利要求6－8任一项所述的滤光方法，其特征在于，
    使用第一色散棱镜对所述包含多个波段的入射光信号进行色散，使用第二色散棱镜对经过色散的入射光信号进行准直。
  10. 根据权利要求9所述的滤光方法，其特征在于，
    所述第一色散棱镜与所述第二色散棱镜结构相同。
  11. 一种多光谱成像系统，其特征在于，包括权利要求1－5任一项所述的滤光器，以及光信号生成单元。
  12. 根据权利要求11所述的多光谱成像系统，其特征在于，
    所述光信号生成单元是共聚焦显微镜。
  13. 根据权利要求12所述的多光谱成像系统，其特征在于，
    所述共聚焦显微镜与所述滤光器之间通过光纤传输或者空间传输的方式传输光信号。
PCT/CN2020/107160 2020-08-05 2020-08-05 一种滤光器、滤光方法及多光谱成像系统 WO2022027336A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/107160 WO2022027336A1 (zh) 2020-08-05 2020-08-05 一种滤光器、滤光方法及多光谱成像系统

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/107160 WO2022027336A1 (zh) 2020-08-05 2020-08-05 一种滤光器、滤光方法及多光谱成像系统

Publications (1)

Publication Number Publication Date
WO2022027336A1 true WO2022027336A1 (zh) 2022-02-10

Family

ID=80118677

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/107160 WO2022027336A1 (zh) 2020-08-05 2020-08-05 一种滤光器、滤光方法及多光谱成像系统

Country Status (1)

Country Link
WO (1) WO2022027336A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114441495A (zh) * 2022-02-13 2022-05-06 中国科学院长春光学精密机械与物理研究所 多色荧光显微成像系统
CN116148811A (zh) * 2023-04-18 2023-05-23 锋睿领创(珠海)科技有限公司 基于波长异位分布平行光源的多光谱分段成像视觉系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103454774A (zh) * 2013-09-16 2013-12-18 华中科技大学 一种光的滤波方法及光滤波器
US8810884B1 (en) * 2012-02-14 2014-08-19 Artur Olszak Light sources for spectrally controlled interferometry
CN105628200A (zh) * 2015-12-24 2016-06-01 南京理工大学 计算光谱成像装置
CN107167929A (zh) * 2017-06-12 2017-09-15 华南师范大学 基于dmd的双模式光学超分辨显微成像装置及方法
CN109683307A (zh) * 2018-12-28 2019-04-26 中国科学院深圳先进技术研究院 一种基于dmd的调谐滤光装置及调谐滤光方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8810884B1 (en) * 2012-02-14 2014-08-19 Artur Olszak Light sources for spectrally controlled interferometry
CN103454774A (zh) * 2013-09-16 2013-12-18 华中科技大学 一种光的滤波方法及光滤波器
CN105628200A (zh) * 2015-12-24 2016-06-01 南京理工大学 计算光谱成像装置
CN107167929A (zh) * 2017-06-12 2017-09-15 华南师范大学 基于dmd的双模式光学超分辨显微成像装置及方法
CN109683307A (zh) * 2018-12-28 2019-04-26 中国科学院深圳先进技术研究院 一种基于dmd的调谐滤光装置及调谐滤光方法

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114441495A (zh) * 2022-02-13 2022-05-06 中国科学院长春光学精密机械与物理研究所 多色荧光显微成像系统
CN114441495B (zh) * 2022-02-13 2024-03-22 中国科学院长春光学精密机械与物理研究所 多色荧光显微成像系统
CN116148811A (zh) * 2023-04-18 2023-05-23 锋睿领创(珠海)科技有限公司 基于波长异位分布平行光源的多光谱分段成像视觉系统
CN116148811B (zh) * 2023-04-18 2023-06-20 锋睿领创(珠海)科技有限公司 基于波长异位分布平行光源的多光谱分段成像视觉系统

Similar Documents

Publication Publication Date Title
US10591417B2 (en) Systems and methods for 4-D hyperspectral imaging
JP5953230B2 (ja) 空間的な色分離による可変フィルタリングを用いたスペクトル検出器又はレーザ走査顕微鏡
US6229635B1 (en) Light sensing device
CN109196333B (zh) 用于4-d高光谱成像的系统和方法
JP6096814B2 (ja) スペクトル検出を伴う光走査型顕微鏡
US7605976B1 (en) Method and device for changing light in an adjustable manner
JP6061958B2 (ja) ソフトウェア定義式顕微鏡
US20100314554A1 (en) Device to illuminate an object with a multispectral light source and detect the spectrum of the emitted light
DE19835072A1 (de) Anordnung zur Beleuchtung und/oder Detektion in einem Mikroskop
JP2000199855A (ja) 走査型光学顕微鏡装置
JP2021505940A (ja) 顕微鏡システムおよびこの種の顕微鏡システムを用いた顕微鏡結像のための方法
US11686928B2 (en) Light microscope
WO2022027336A1 (zh) 一种滤光器、滤光方法及多光谱成像系统
US8189191B2 (en) Spectroscopic imaging microscopy
US20140368904A1 (en) Software Defined Microscope
JP3568626B2 (ja) 走査型光学顕微鏡
CN112098378A (zh) 一种滤光器、滤光方法及多光谱成像系统
JP7079509B2 (ja) 計測装置及び照射装置
JP5871149B2 (ja) 顕微分光システム
US11947097B2 (en) Bandpass filter for light having variable lower and upper cut-off wavelengths
JP5787151B2 (ja) 分光ユニット及び走査型顕微鏡
US20220229307A1 (en) Spectrally Shaped Light Source
JP6459070B2 (ja) 分光器及び顕微鏡システム

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20947955

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20947955

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 20947955

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 10/08/2023)