Three-dimensional fluorescence microscope device capable of simultaneously acquiring contrast images

Abstract

The invention discloses a three-dimensional fluorescence microscope device capable of simultaneously obtaining contrast images, and relates to the technical field of microscope imaging, a laser emission mechanism comprises a light emitting part and a first slit for converting light output by the light emitting part into a point light source to be output; the object carrying platform comprises a placement table and a three-dimensional driving mechanism for driving the placement table to realize three-dimensional motion, and an object to be detected is arranged on the placement table; the first light conduction mechanism comprises a dichroscope, a first objective lens and a second slit which are sequentially arranged along an illumination light path, the first slit and the second slit are conjugated, and the exciting light is focused on a to-be-detected object arranged on the object carrying platform through the second slit to excite a fluorescence signal; the second light conduction mechanism comprises an optical filter group and a spectroscope, the excited fluorescence signal is divided into two paths after passing through the spectroscope, one path enters the first receiving mechanism, and the other path enters the second receiving mechanism; according to the fluorescence microscope device, the contrast effect of the image is better, and detail information can be better identified.

Description

Technical Field

[0001] The invention relates to the technical field of microscope imaging, and in particular to a three-dimensional fluorescence microscope device capable of simultaneously acquiring contrast images. Background Art

[0002] As one of the most common tumor types in the world, early detection and timely treatment of digestive system tumors are extremely important for improving the survival rate of patients and reducing the socioeconomic burden. With the continuous advancement of medical technology, improving the diagnosis level of early cancer has become the focus of the current medical community. Many international studies have shown that endoscopic examination is one of the most effective means of detecting digestive system tumors. At present, medical endoscope detection systems have been widely used in clinical practice. These systems include both traditional white light reflection endoscopes (such as fiber endoscopes, electronic endoscopes, etc.) and advanced fluorescent endoscopes. These endoscope detection systems generally use thin-diameter flexible image-transmitting optical fiber bundles to transmit images. This optical fiber bundle is integrated by many closely arranged optical fiber filaments. It is soft and easy to bend, and can easily enter the human body for observation.

[0003] However, due to the tiny gaps between the optical fibers, these gaps will lead to blind spots during the imaging process, resulting in the appearance of an optical fiber grid on the acquired image. This optical fiber grid not only affects the clarity of the image, but may also cover up some important lesion information, thus causing trouble for doctors in their diagnosis.

[0004] In order to eliminate this blind spot, researchers have adopted a variety of methods. Among them, the Gaussian filter algorithm is a commonly used denoising method. By applying the Gaussian filter algorithm to the original fluorescence or white light image, the grid image caused by the fiber spacing can be effectively removed, improving the clarity of the image. In addition, the contrast enhancement of the image using the histogram equalization algorithm can further improve the visual effect of the image and generate a corrected fluorescence or white light image.

[0005] However, although these methods can improve the imaging quality of images to a certain extent, it is difficult to accurately judge the pros and cons of these denoising algorithms due to the lack of real images without fiber grids as comparison standards, that is, real gridless images as references. This makes it difficult for researchers to optimize the de-gridding algorithm in practical applications, and it is impossible to objectively compare different de-gridding algorithms, or judge the difference between the same algorithm before and after optimization. In particular, it is difficult to simultaneously obtain gridded and gridless comparison images of images of different depths of the same tissue sample. When different excitation light sources are used to excite the tissue, the images obtained are also different, so gridded images and gridless images of different excitation light sources also need to be obtained at the same time for comparison. Summary of the invention

[0006] Based on the technical problems existing in the background technology, the present invention proposes a three-dimensional fluorescence microscope device that can simultaneously obtain contrast images, so that the image contrast effect is better and the detail information can be better identified.

[0007] The present invention provides a three-dimensional fluorescence microscope device capable of simultaneously acquiring contrast images, comprising: The laser emitting mechanism comprises a light emitting part and a first slit for converting the light outputted by the light emitting part into a point light source output; The object-carrying platform is used to drive the spatial movement of the object to be detected; A first light transmission mechanism includes a second slit, the first slit and the second slit are conjugate, and the excitation light passes through the second slit and is focused on the object to be detected to excite a fluorescent signal; The second light transmission machine, the excited fluorescent signal is divided into two paths after passing through the first light transmission mechanism and the second light transmission mechanism, one path enters the first receiving mechanism to output an image without the optical fiber grid, and the other path enters the second receiving mechanism to output an image with the optical fiber grid.

[0008] Furthermore, the light emitting unit includes a first rotating wheel and a laser, at least two lasers are arranged on the first rotating wheel, and the lasers with required wavelengths are transferred to the illumination light path by rotating the first rotating wheel.

[0009] Furthermore, the laser emitting mechanism also includes a beam expander lens for expanding the point light source output through the first slit into light with a certain diameter.

[0010] Furthermore, the object-carrying platform includes a placement table and a three-dimensional driving mechanism, and the object to be detected is arranged on the placement table; The three-dimensional driving mechanism includes a Z-axis lifting mechanism, an X-axis motion mechanism and a Y-axis motion mechanism. The Y-axis motion mechanism is arranged at the motion end of the X-axis motion mechanism, the X-axis motion mechanism is arranged at the lifting end of the Z-axis lifting mechanism, and the placement table is arranged at the motion end of the Y-axis motion mechanism, thereby realizing the three-dimensional motion of the placement table.

[0011] Furthermore, the first light transmission mechanism further comprises a dichroic mirror and a first objective lens sequentially arranged along the illumination light path; The object to be detected is set at the focal point of the first objective lens. Through the lifting and lowering movement of the Z-axis lifting mechanism, the focusing point of the first objective lens at different thicknesses of the object to be detected is adjusted to achieve three-dimensional fluorescence imaging of the object to be detected.

[0012] Furthermore, the second light transmission mechanism includes a filter group and a beam splitter, and the filter group is arranged between the dichroic mirror and the beam splitter; The filter group includes a second rotating wheel and filters, and the number of filters corresponds to the number of lasers, so that the excitation light signals emitted by the corresponding lasers are filtered out by the filters.

[0013] Further, the first receiving mechanism includes a first focusing mirror, a third slit and a first photomultiplier tube; The second slit is conjugate with the third slit; After passing through the beam splitter, the fluorescence signal is sequentially projected onto the detection chip of the first photomultiplier tube through the first focusing mirror and the third slit, and the first photomultiplier tube outputs an image without the optical fiber grid.

[0014] Furthermore, the second receiving mechanism includes a second objective lens, an image transmission optical fiber, a third objective lens, a second focusing lens, a fourth slit and a second photomultiplier tube; The second slit and the fourth slit are conjugate; The other fluorescent signal after passing through the beam splitter is projected onto the detection chip of the second photomultiplier tube through the second objective lens, image transmission fiber, third objective lens, second focusing lens, and fourth slit in sequence, and the second photomultiplier tube outputs an image of the fiber optic grid.

[0015] Furthermore, the first photomultiplier tube outputs an image without the optical fiber grid based on the conjugation of the second slit and the third slit, which is a partial image of the object to be detected; The second photomultiplier tube is based on the conjugation of the second slit and the fourth slit, and the image of the optical fiber grid outputted is a local image of the object to be detected; The partial images of multiple objects to be detected are stitched together into a complete image of the object to be detected by an image stitching algorithm; The local images are used as the images to be stitched, and the image stitching algorithm is specifically as follows: The image registration method based on feature points is used to preliminarily align the stitched images; A mathematical model is constructed based on the synchronous registration of TESCV similarity to perform pixel registration on the images to be spliced ​​after preliminary alignment, thereby completing the splicing of the images to be spliced.

[0016] Furthermore, the mathematical expression of the mathematical model is as follows: ; in, For the The local image obtained is For the The local image obtained is for and The displacement field between them, TESCV is the TESCV similarity, and They are and At the pixels in the overlapping area, is the thin plate spline energy function, is the displacement field, which indicates the distance between the local image and the reference image. is the penalty factor of the constraint term in the registration, for the entire image area.

[0017] The advantage of a three-dimensional fluorescence microscope device capable of simultaneously acquiring contrast images provided by the present invention is that: a three-dimensional fluorescence microscope device capable of simultaneously acquiring contrast images provided in the structure of the present invention is provided with four slits, and is cooperated with a loading platform capable of realizing three-dimensional motion, so that two photomultiplier tubes can perform local imaging and three-dimensional imaging, thereby obtaining images of different depths of the same object to be detected, so that the image contrast effect is better and detail information can be better identified. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a structural schematic diagram of the present invention; Figure 2 It is a structural schematic diagram of the image transmission optical fiber; Among them, 1-laser emitting mechanism, 2-object carrier platform, 3-first light transmission mechanism, 4-second light transmission mechanism, 5-first receiving mechanism, 6-second receiving mechanism, 11-light emitting part, 12-first slit, 13-beam expanding lens, 21-placing table, 22-three-dimensional driving mechanism, 31-dichroic mirror, 32-first objective lens, 33-second slit, 41-filter group, 42-beam splitter, 51-first focusing lens, 52-third slit, 53-first photomultiplier tube, 61-second objective lens, 62-image transmission optical fiber, 63-third objective lens, 64-second focusing lens, 65-fourth slit, 66-second photomultiplier tube, 111-first rotating wheel, 112-laser, 411-filter, 412-second rotating wheel, 621-cladding part, 622-optical fiber part, 623-gap part. DETAILED DESCRIPTION

[0019] Below, the technical solution of the present invention is described in detail through specific embodiments. Many specific details are set forth in the following description to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific implementation disclosed below.

[0020] like Figure 1 and 2As shown, the present invention proposes a three-dimensional fluorescence microscope device capable of simultaneously acquiring contrast images, comprising a laser emitting mechanism 1, a loading platform 2, a first light transmission mechanism 3, a second light transmission mechanism 4, a first receiving mechanism 5 and a second receiving mechanism 6; The laser emitting mechanism 1 comprises a light emitting part 11 and a first slit 12 which converts the light outputted by the light emitting part 11 into a point light source output; The object loading platform 2 includes a placement table 21 and a three-dimensional driving mechanism 22 for driving the placement table 21 to achieve three-dimensional movement, and the object to be detected is placed on the placement table 21; The first light transmission mechanism 3 includes a dichroic mirror 31, a first objective lens 32 and a second slit 33 arranged in sequence along the illumination light path. The first slit 12 and the second slit 33 are conjugate. The excitation light passes through the second slit 33 and is focused on the object to be detected disposed on the object platform 2 to excite the fluorescent signal. The second light transmission mechanism 4 includes a filter set 41 and a beam splitter 42. The excited fluorescence signal passes through the first objective lens 32, the dichroic mirror 31, the filter set 41 and the beam splitter 42 and is divided into two paths, one path entering the first receiving mechanism 5 and the other path entering the second receiving mechanism 6. The first receiving mechanism 5 is used to receive a fluorescent signal after passing through the beam splitter 42 to output an image without an optical fiber grid; The second receiving mechanism 6 is used to receive another path of fluorescent signal after passing through the beam splitter 42 to output an image with an optical fiber grid.

[0021] This embodiment can simultaneously obtain images with and without fiber grids, especially contrast images of different depths of the same thick tissue. When different fluorescence contrast images are needed, different excitation light sources can be conveniently used in conjunction with specific filters to obtain fluorescence images with or without grids under different excitation light sources at the same position. In addition, this embodiment achieves the generation of three-dimensional images by setting a liftable loading platform 2. In addition, this device system can ensure that the images obtained are at the same position. When developing and designing a de-gridding algorithm, the gridless image is used as the final real image to judge the pros and cons of the de-gridding algorithm, and plays a certain guiding role in the algorithm optimization route. It can also be used for comparative analysis of different de-gridding algorithms, etc.

[0022] In this embodiment, the structure of the image transmission optical fiber 62 is as follows: Figure 2 As shown, the cladding portion 621 cannot transmit light, the optical fiber portion 622 is used to transmit optical information, and the gap portion 623 cannot transmit light. Due to the structural characteristics of the image transmission optical fiber 62 itself, the image signal after passing through the image transmission optical fiber 62 has a grid.

[0023] 1. Laser emitting mechanism 1; The laser emitting mechanism 1 includes a light-emitting part 11, a first slit 12 and a beam expander lens 13; specifically, the light-emitting part 11 includes a first rotating wheel 111 and a laser 112, at least two lasers 112 are arranged on the first rotating wheel 111, and by rotating the first rotating wheel 111, the laser 112 of the required wavelength is transferred to the illumination light path.

[0024] Laser is a special light source with the characteristics of strong directionality, high brightness, good monochromaticity and coherence, etc. This embodiment uses laser as the light source, which can effectively reduce the system power and improve the fluorescence signal excitation efficiency.

[0025] Different dyes are adapted to different excitation wavelengths. Therefore, the light-emitting portion 11 of this embodiment uses a rotating wheel to arrange several lasers as an excitation light source. According to the selected dye, the wheel is rotated to allow the laser with a specific output wavelength to enter the optical path system. The laser outputs light signals of different wavelengths and is adapted to different emission light filters and dyes. Different lasers and corresponding emission light filters and dyes can be selectively replaced as needed.

[0026] 2. Carrying platform 2; The loading platform 2 includes a placement table 21 and a three-dimensional driving mechanism 22 for driving the placement table 21 to achieve three-dimensional motion, and the object to be detected is set on the placement table 21. The three-dimensional driving mechanism 22 includes a Z-axis lifting mechanism, an X-axis motion mechanism and a Y-axis motion mechanism, the Y-axis motion mechanism is arranged at the motion end of the X-axis motion mechanism, the X-axis motion mechanism is arranged at the lifting end of the Z-axis lifting mechanism, and the placement table 21 is arranged at the motion end of the Y-axis motion mechanism, so as to achieve three-dimensional motion of the placement table 21.

[0027] The object to be detected is set at the focal point of the first objective lens 32. Through the lifting movement of the Z-axis lifting mechanism, the focal point of the first objective lens 32 at different thicknesses of the object to be detected is adjusted, thereby realizing three-dimensional fluorescence imaging of the object to be detected.

[0028] Normally, the moving range of the placement table 21 is a relatively large rectangle. The X-axis motion mechanism is adjusted to move. After moving one row, the Y-axis motion mechanism is moved one grid. After the X-axis motion mechanism moves one row, the Y-axis motion mechanism is moved one grid until the entire matrix range is scanned. Then, the Z-axis lifting mechanism is lifted and lowered to scan the rectangular range image of the next layer of the object to be detected, and finally a three-dimensional image is synthesized.

[0029] When setting the scanning range of the object to be detected, according to the size of the end face of the imaging optical fiber 62, only the circular size range of the corresponding optical fiber portion 622 is scanned to avoid scanning other information images that cannot be matched one by one, thereby improving the scanning efficiency.

[0030] 3. A first light transmission mechanism 3; The first light transmission mechanism 3 includes a dichroic mirror 31, a first objective lens 32 and a second slit 33 arranged in sequence along the illumination light path. The first light transmission mechanism 3 transmits the excitation light to the detection area on the placement table 21, where there is an object to be detected, and the collection unit can collect the excited fluorescence signal from the detection area. The light emitted by the laser light source passes through the first slit 12, and is expanded by the beam expansion lens 13, the dichroic mirror 31, the first objective lens 32 and the second slit 33 in sequence, and is irradiated onto the tissue surface of the object to be detected.

[0031] The tissue surface is coated with a specific dye, and fluorescence is excited by the irradiation of excitation light. The wavelength of the fluorescence signal is usually greater than the excitation wavelength, and the fluorescence signal is received by the first receiving mechanism 5 and the second receiving mechanism 6 at the rear end respectively.

[0032] 4. A second light transmission mechanism 4; The second light transmission mechanism 4 includes a filter set 41 and a spectroscope 42 ; the filter set 41 includes a second rotating wheel 412 and filters 411 , and the number of filters 411 corresponds to the number of lasers 112 , so as to filter out the excitation light signal emitted by the corresponding laser 112 through the filters 411 .

[0033] Regarding the relationship between the light-emitting part 11 and the filter group 41: the first rotating wheel 111 is provided with an encoder component to obtain the rotation position information of the first rotating wheel 111, and the second rotating wheel 412 is also provided with an encoder component to facilitate obtaining the rotation position information of the second rotating wheel 412. The two encoder components preferably use absolute encoders, which can directly output digital signals to determine the position information. The laser 112 can output a specific wavelength, and the corresponding transmission wavelength of the filter 411 needs to correspond one by one, otherwise the excited fluorescence signal will not be able to pass through the filter 411. In order to avoid the laser and the filter 411 not corresponding accurately due to the wear between the gears during long-term use, the spectral signal range can be fed back at the first photomultiplier tube 53 or the second photomultiplier tube 66.

[0034] Assuming that the three positions of the laser 112 correspond to three different wavelength laser output signals, the digital signals output by the encoder component encoded by the first rotating wheel 111 are 100, 200, and 300 respectively, and the three spectral ranges fed back at the first photomultiplier tube 53 or the second photomultiplier tube 66 are defined as 520~540, 550~570, and 580~610, which can be defined as codes 10, 20, and 30 respectively.

[0035] Use the software control program to read the digital signal output by the encoder component encoded by the first rotating wheel 111, and read the code of the wavelength range output by the first photomultiplier tube 53 or the second photomultiplier tube 66. Only when the result of dividing the digital signal range by the wavelength range code is 10, it means that the laser 112 and the filter 411 are one-to-one corresponding in the optical path, indicating that there is no problem with the optical path. Otherwise, the second rotating wheel 412 of the filter 411 is driven, and the value of the division between the two is calculated in real time until the result is 10, indicating that the optical path has been debugged, and the rotation of the second rotating wheel 412 is stopped. If the second rotating wheel 412 rotates one circle, the calculation result is not equal to 10, indicating that the position of the laser 112 or the filter 411 may have shifted significantly, and the output signal is fed back to the display, and after-sales adjustment is contacted.

[0036] 5. A first receiving mechanism 5 and a second receiving mechanism 6; The first receiving mechanism 5 includes a first focusing mirror 51, a third slit 52 and a first photomultiplier tube 53; the second slit 33 and the third slit 52 are conjugate; the fluorescence signal after passing through the beam splitter 42 is projected onto the detection chip of the first photomultiplier tube 53 through the first focusing mirror 51 and the third slit 52 in sequence, and the first photomultiplier tube 53 outputs an image without a fiber optic grid.

[0037] The second receiving mechanism 6 includes a second objective lens 61, an image transmission optical fiber 62, a third objective lens 63, a second focusing lens 64, a fourth slit 65 and a second photomultiplier tube 66; the second slit 33 and the fourth slit 65 are conjugate; the other fluorescent signal after passing through the beam splitter 42 is sequentially projected onto the detection chip of the second photomultiplier tube 66 through the second objective lens 61, the image transmission optical fiber 62, the third objective lens 63, the second focusing lens 64 and the fourth slit 65, and the second photomultiplier tube 66 outputs an image of the optical fiber grid.

[0038] The objective lenses of this embodiment (the first objective lens 32, the second objective lens 61 and the third objective lens 63) are finite-field plan achromatic objective lenses with RMS external threads. The magnification can generally be 4X, 10X, 20X, 40X. The objective lenses are parfocal and can amplify the fluorescent image and form an enlarged real image at the conjugate point of the magnifying objective lens. That is, the fluorescent signal excited by a single point is divided into two paths after passing through the beam splitter 42, and the two paths are received by the first photomultiplier tube 53 and the second photomultiplier tube 66 respectively.

[0039] The second slit 33 is conjugate with the third slit 52 , and the second slit 33 is conjugate with the fourth slit 65 , so that the out-of-focus signal (fluorescence signal far away from the focal length of the second objective lens 33 ) cannot enter the first photomultiplier tube 53 and the second photomultiplier tube 66 .

[0040] In addition, in the present embodiment, by setting four slits (a first slit 12, a second slit 33, a third slit 52 and a fourth slit 65) and cooperating with a loading platform 2 that can realize three-dimensional motion, the first photomultiplier tube 53 and the second photomultiplier tube 66 can perform local imaging and three-dimensional imaging, thereby obtaining images of different depths of the same object to be detected, splicing images of the same depth into a two-dimensional image, and splicing two-dimensional images of different layers into a three-dimensional image, thereby realizing three-dimensional imaging.

[0041] That is, the four slits can filter out the fluorescence signal outside the focal plane of the second objective lens 33. By moving the carrier platform 2 in the vertical and horizontal directions, the image within the range of the optical fiber end face can be acquired. By image stitching, the images of different positions of the tissue to be detected can be stitched into one image, for example, the tissue image of a circle with a diameter of 600um can be stitched into an image of a circle with a diameter of 3000um.

[0042] Specifically, the fluorescence signal excited at the object to be detected is collimated by the first objective lens 32, and then passes through the dichroic mirror 31 and the filter group 41 to filter out stray light signals. After passing through the beam splitter 42, since the beam splitter 42 does not have wavelength selectivity, the fluorescence is divided into two paths according to the intensity. One path of fluorescence passes through the first focusing mirror 51. The first focusing mirror 51 adopts a double-glued focusing mirror to effectively eliminate chromatic aberration, and projects the point signal through the third slit 52 onto the first photomultiplier tube 53. After the two-dimensional movement of the three-dimensional driving mechanism, the point signal on the first photomultiplier tube 53 is converted into a two-dimensional image signal (i.e., a planar array image) through Blender software processing. This path obtains an image without a fiber optic grid. After passing through the beam splitter 42, another channel of fluorescence is converged to the head end of the image transmission fiber 62 through the second objective lens 61, and the point signal is coupled into the image transmission fiber 62. After being transmitted through the image transmission fiber 62, it is emitted from the other end, passes through the third objective lens 63, and then passes through the second focusing lens 64, and then the information with the grid image is projected to the second photomultiplier tube 66. After the two-dimensional movement of the three-dimensional driving mechanism, the point signal on the second photomultiplier tube 66 can be converted into a two-dimensional image signal (i.e., a planar array image) through Blender software processing. This path obtains an image with a fiber grid. If a three-dimensional contrast image needs to be obtained, after the movement in the XY direction is completed, the Z-axis lifting mechanism is adjusted to image different layers of the object to be detected, thereby obtaining a three-dimensional contrast image.

[0043] The image without the optical fiber grid output by the first photomultiplier tube 53 is used as the standard image in the comparison image, and the image with the optical fiber grid output by the second photomultiplier tube 66 is used as the image to be compared in the comparison image.

[0044] In this embodiment, the first photomultiplier tube 53 is based on the conjugation of the second slit 33 and the third slit 52, and the image without the optical fiber grid outputted is a local image of the object to be detected; the second photomultiplier tube 66 is based on the conjugation of the second slit 33 and the fourth slit 65, and the image with the optical fiber grid outputted is a local image of the object to be detected.

[0045] The local images corresponding to the image without the optical fiber grid and the image with the optical fiber grid are respectively stitched into a complete image of the object to be detected by an image stitching algorithm, and the image stitching algorithms of the two comparison images are consistent.

[0046] In terms of image stitching algorithms, on the one hand, in view of the coupling effect of rigid motion and elastic deformation between different image frames, the image registration method based on feature points (Harris_PIIFD feature point matching) is used to achieve preliminary alignment of adjacent frames, and then the synchronous registration algorithm based on the synchronous transformation model is used to achieve simultaneous tuning of rigid and elastic transformation parameters. On the other hand, the image texture association and optimal grayscale mapping theory are combined to construct more robust texture encoding conditional variance and TESCV similarity, where Harris_PIIFD specifically refers to Harris Partial Intensity Invariant FeatureDescriptor, that is, the feature descriptor used in the image registration task.

[0047] Synchronous Transformation Model: Assumptions represents the entire image area, is the number of pixels of width, is the number of pixels of height, is one pixel in the width direction, is a pixel in the height direction, It is the overlapping area after matching by the image registration method through feature points. For the The local image obtained is For the The local image obtained is for and The displacement field between the two, the mathematical model of synchronous registration based on TESCV similarity is as follows: ; Among them, TESCV is TESCV similarity, and They are and At the pixels in the overlapping area, is the thin plate spline energy function, which is used to ensure the smoothness of the displacement field. is the displacement field, which indicates the distance between the local image and the reference image. is the penalty factor for the constraint term in the registration.

[0048] For two microscopic images and , SCV similarity is a set of mapping relationships. Image registration based on SCV similarity assumes that when two images to be registered are completely aligned, there is a one-to-one corresponding function mapping relationship between the grayscale values ​​of all pixels. The conditional variance and SCV similarity between the two can be expressed as: ; Among them, SCV is SCV similarity, is the number of pixels in the overlapping area; is the posterior expectation of the current pixel.

[0049] In this embodiment, TESCV is an algorithm that calculates and analyzes variances in different regions or conditions in an image, and encodes or represents these variances in the form of textures; TESCV (Texture Encoded Sum of Conditional Variance) similarity and SCV (Sum of Conditional Variance) similarity are methods for calculating image similarity in OpenCV, which are existing methods and will not be described in detail in this embodiment.

[0050] That is, the image stitching algorithm is specifically as follows: read the images to be stitched, perform preliminary alignment of adjacent frames of local images based on the image registration method of feature points (feature point matching of Harris_PIIFD), and then build a mathematical model based on the synchronous registration of TESCV similarity to perform pixel registration on the images to be stitched after the preliminary alignment, thereby completing the stitching of the images to be stitched.

[0051] This embodiment can simultaneously obtain the presence or absence of the optical fiber grid contrast image within the same position stereoscopic range. By setting each slit, the image contrast effect is better and the detail information can be better identified. In addition, based on the confocal principle, this embodiment is equipped with slits near the laser emission mechanism 1 (illumination), the first light transmission mechanism 3 (detection), and the two photomultiplier tubes (focusing), so that the collected signal is the signal at the focus point, that is, the three points of illumination, focusing, and detection are conjugated in pairs, and the filter group 41 filters out the excitation light signal and other stray light of non-fluorescent signals, so that the image contrast is better, especially for thicker tissue imaging, the imaging quality is significantly better than the wide field imaging effect.

[0052] Existing wide-field imaging is also an illumination, focusing and detection setup. The excitation light of wide-field imaging is directly shone on the object to be detected, and the fluorescent signal is focused into two cameras for imaging. In this process, the two cameras directly image the entire plane of the object to be detected, and obtain images without fiber grids and images with fiber grids. This imaging process has the following defects: wide-field illumination receives all signals near the focal plane of the tissue, the obtained image has poor contrast, the contours of tissues, glands, etc. are not clearly identified, and it is impossible to obtain a three-dimensional image by moving tissues, etc.

[0053] The device proposed in this embodiment can simultaneously obtain contrast images of the presence or absence of optical fiber grids within a three-dimensional range at the same position. Based on the confocal principle, slits are installed at the light source, the object to be detected, and the end of the photomultiplier tube, so that the collected signal can be the signal at the focus point, that is, the three points of illumination, focusing, and detection are conjugated in pairs, and the stray light outside the focus point is filtered out at the detection point, so that the image contrast obtained is better, especially for imaging of thicker tissues, the imaging quality is significantly better than the wide-field imaging effect, and imaging within the range of the optical fiber end face can be achieved by moving the loading platform 2, and images in a larger tissue area can be synthesized by using image stitching technology.

[0054] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A three-dimensional fluorescence microscope device capable of simultaneously acquiring contrast images, characterized in that: include: A laser emitting mechanism (1) comprises a light emitting portion (11) and a first slit (12) for converting light outputted by the light emitting portion (11) into a point light source output; The object-carrying platform (2) is used to drive the spatial movement of the object to be detected; The first light transmission mechanism (3) comprises a second slit (33), the first slit (12) and the second slit (33) being conjugate, and the excitation light passes through the second slit (33) and is focused on the object to be detected to excite a fluorescent signal; The second light transmission mechanism (4) is configured such that the excited fluorescent signal passes through the first light transmission mechanism (3) and the second light transmission mechanism (4) and is divided into two paths, one path entering the first receiving mechanism (5) to output an image without the optical fiber grid, and the other path entering the second receiving mechanism (6) to output an image with the optical fiber grid.

2. The three-dimensional fluorescence microscope device capable of simultaneously acquiring contrast images according to claim 1, characterized in that: The light emitting unit (11) comprises a first rotating wheel (111) and a laser (112), wherein at least two lasers (112) are arranged on the first rotating wheel (111), and the lasers (112) of a required wavelength are transferred to the illumination light path by rotating the first rotating wheel (111).

3. The three-dimensional fluorescence microscope device capable of simultaneously acquiring contrast images according to claim 2, characterized in that: The laser emitting mechanism (1) also includes a beam expanding lens (13) for expanding the point light source output through the first slit (12) into light with a certain diameter.

4. The three-dimensional fluorescence microscope device capable of simultaneously acquiring contrast images according to claim 1, characterized in that: The object loading platform (2) comprises a placement table (21) and a three-dimensional driving mechanism (22), and the object to be detected is arranged on the placement table (21); The three-dimensional driving mechanism (22) comprises a Z-axis lifting mechanism, an X-axis motion mechanism and a Y-axis motion mechanism, the Y-axis motion mechanism is arranged at the motion end of the X-axis motion mechanism, the X-axis motion mechanism is arranged at the lifting end of the Z-axis lifting mechanism, and the placement table (21) is arranged at the motion end of the Y-axis motion mechanism, thereby realizing three-dimensional motion of the placement table (21).

5. The three-dimensional fluorescence microscope device capable of simultaneously acquiring contrast images according to claim 4, characterized in that: The first light transmission mechanism (3) further comprises a dichroic mirror (31) and a first objective lens (32) which are sequentially arranged along the illumination light path; The object to be detected is arranged at the focal point of the first objective lens (32), and the focal points of the first objective lens (32) at different thicknesses of the object to be detected are adjusted by the lifting and lowering movement of the Z-axis lifting mechanism, thereby achieving three-dimensional fluorescent imaging of the object to be detected.

6. The three-dimensional fluorescence microscope device capable of simultaneously acquiring contrast images according to claim 2, characterized in that: The second light transmission mechanism (4) comprises a filter group (41) and a beam splitter (42), wherein the filter group (41) is arranged between the dichroic mirror (31) and the beam splitter (42); The filter set (41) comprises a second rotating wheel (412) and filters (411), and the number of filters (411) corresponds to the number of lasers (112), so that the filters (411) can filter out the excitation light signal emitted by the corresponding laser (112).

7. The three-dimensional fluorescence microscope device capable of simultaneously acquiring contrast images according to claim 1, characterized in that: The first receiving mechanism (5) comprises a first focusing mirror (51), a third slit (52) and a first photomultiplier tube (53); The second slit (33) and the third slit (52) are conjugate; A fluorescent signal passing through the beam splitter (42) is sequentially projected onto a detection chip of a first photomultiplier tube (53) through a first focusing mirror (51) and a third slit (52), and the first photomultiplier tube (53) outputs an image without an optical fiber grid.

8. The three-dimensional fluorescence microscope device capable of simultaneously acquiring contrast images according to claim 7, characterized in that: The second receiving mechanism (6) comprises a second objective lens (61), an image transmission optical fiber (62), a third objective lens (63), a second focusing lens (64), a fourth slit (65) and a second photomultiplier tube (66); The second slit (33) and the fourth slit (65) are conjugate; The other fluorescent signal after passing through the beam splitter (42) is projected onto the detection chip of the second photomultiplier tube (66) through the second objective lens (61), the image transmission optical fiber (62), the third objective lens (63), the second focusing lens (64), and the fourth slit (65) in sequence, and the second photomultiplier tube (66) outputs an image of the optical fiber grid.

9. The three-dimensional fluorescence microscope device capable of simultaneously acquiring contrast images according to claim 8, characterized in that: The first photomultiplier tube (53) is based on the conjugation of the second slit (33) and the third slit (52), and the image without the optical fiber grid outputted is a local image of the object to be detected; The second photomultiplier tube (66) is based on the conjugation of the second slit (33) and the fourth slit (65), and the image outputted by the optical fiber grid is a local image of the object to be detected; The partial images of multiple objects to be detected are stitched together into a complete image of the object to be detected by an image stitching algorithm; The local images are used as the images to be stitched, and the image stitching algorithm is specifically as follows: The image registration method based on feature points is used to preliminarily align the stitched images; A mathematical model is constructed based on the synchronous registration of TESCV similarity to perform pixel registration on the images to be spliced ​​after preliminary alignment, thereby completing the splicing of the images to be spliced.

10. The three-dimensional fluorescence microscope device capable of simultaneously acquiring contrast images according to claim 9, characterized in that: The mathematical expression of the mathematical model is as follows: in, For the The local image obtained is For the The local image obtained is for and The displacement field between them, TESCV is the TESCV similarity, and They are and At the pixels in the overlapping area, is the thin plate spline energy function, is the displacement field, which indicates the distance between the local image and the reference image. is the penalty factor of the constraint term in the registration, for the entire image area.

Images