A method for focusing a multimode fiber exit end based on a metasurface

Abstract

The application discloses a kind of based on super surface multimode fiber exit end point focusing method, using super surface to carry out wavefront regulation and control to incident light field, make the light spot of multimode fiber exit change from speckle into focused light spot.The specific steps are: according to theory, build multimode fiber point focusing light path based on spatial light modulator (SLM), calculate the phase corresponding to focusing light spot at a certain target position by sequential iteration optimization method;Then simulate and calculate the structure of the corresponding super surface based on this phase, the wavefront phase modulation of incident light wave can realize the point focusing of multimode fiber exit light spot at the corresponding position.The application carries out point focusing by super surface, compared with other methods has higher integration, is conducive to the development of entire device to smaller, light weight, has broad application prospect in medical, industrial and other fields.

Description

Technical Field

[0001] This invention belongs to the field of endoscopic imaging, specifically relating to a focusing method for the output end of a multimode fiber based on metasurfaces. Background Technology

[0002] Endoscopes are widely used in industry and medicine due to their high flexibility, small size, and ability to penetrate deep into workpieces or the human body. They primarily use single-mode optical fiber for light transmission. Since single-mode fiber can only transmit one mode at a time and cannot transmit image information in parallel, endoscopes typically use fiber bundles composed of multiple single-mode fibers, or single-mode fibers combined with complex robotic arms for imaging. However, both multiple fibers and complex robotic arms limit the further miniaturization of endoscopes, significantly hindering their development in many fields. Unlike single-mode fiber, multimode fiber can transmit multiple guided wave modes in parallel, and its pixel density is one to two orders of magnitude higher than that of single-mode fiber. Its application in endoscopes offers advantages such as small size and high resolution. However, light transmission in multimode fiber exhibits intermodal dispersion and intermodal coupling, resulting in speckled light that loses its spatial phase relationship, limiting its imaging capabilities. To address this issue, the mainstream methods currently include the transfer matrix method and the phase conjugate compensation method. However, these methods suffer from complex optical paths and low integration, hindering their development towards miniaturized and lightweight instruments. Therefore, there is an urgent need for a method with relatively high integration to achieve point focusing of multimode optical fibers. Summary of the Invention

[0003] To address the problem of difficult imaging due to speckle patterns in multimode fiber output light that have lost phase relation, this invention proposes a multimode fiber output endpoint focusing method based on metasurfaces. This method reduces the optical field distortion caused by intermodal dispersion and mode coupling in multimode fibers, and achieves focused optical field output from multimode fibers. It has advantages such as relatively high integration, small size, and simple structure.

[0004] The technical solution to achieve the objective of this invention is as follows: a multimode fiber optic endpoint focusing method based on metasurfaces, comprising the following steps:

[0005] S1. Construct a multimode fiber point-focusing optical path based on SLM:

[0006] The multimode fiber point focusing optical path of the SLM includes a laser, a collimating beam expander, a polarizing beam splitter, an SLM, a focusing objective, a multimode fiber, an objective lens, a convex lens, and a CCD camera arranged sequentially along the optical path; the two ends of the multimode fiber are connected to the focusing objective lens and the objective lens, respectively; the SLM and the CCD camera are respectively connected to a computer.

[0007] The laser emits 532nm green light, which is collimated and expanded into parallel light by a collimating beam expander. The parallel light passes through a polarizing beam splitter and then P-polarized light is incident on the SLM. The light, which is phase-modulated by the SLM, is then refracted by the polarizing beam splitter and enters the focusing objective. The focusing objective couples the refracted light into a 50μm core diameter multimode fiber. The light emitted from the multimode fiber is received by the objective. The objective and the convex lens work together to perform a microscopic function, and finally the light emitted from the convex lens is imaged onto a CCD camera.

[0008] S2. The computer repeatedly accesses the SLM and CCD camera using a sequential iterative optimization method to solve for the loading phase of the SLM that makes the output end of the multimode fiber a focused light field.

[0009] S3. Design the metasurface based on the loading phase of the SLM obtained from the solution.

[0010] S4. Replace the SLM in the multimode fiber point focusing optical path based on SLM with a metasurface to realize that the output end of the multimode fiber is a focused optical field.

[0011] Compared with the prior art, the significant advantages of this invention are:

[0012] (1) In this invention, the metasurface is a single-layer subwavelength size. Compared with other modulation elements such as digital micromirror arrays (DMD), there is no multi-level diffraction order crosstalk, and the focused spot imaging quality is better.

[0013] (2) This invention modulates the phase of the incident light field of a multimode fiber by using a metasurface. Combined with other simple optical components, the output end of the multimode fiber can be transformed from a speckle to a focused light spot. It has the advantages of simple optical path, high integration and lightweight. Attached Figure Description

[0014] Figure 1 It is a multimode fiber point-focusing optical path diagram based on SLM.

[0015] Figure 2 This is a flowchart of the phase solution algorithm for a multimode fiber point focusing optical path based on SLM.

[0016] Figure 3 It is the target phase diagram of the designed metasurface obtained by solving.

[0017] Figure 4 This is a diagram of the focused light field emitted from a multimode fiber.

[0018] Figure 5 It is a simulated diagram of a single superatomic structure.

[0019] Figure 6 It is a graph showing the change in the phase modulation capability of metaatoms as a function of their diameter, where the position of the black triangle indicates the phase selected by the metaatoms. Detailed Implementation

[0020] The present invention will now be described in detail with reference to the accompanying drawings.

[0021] This invention discloses a multimode fiber optic endpoint focusing method based on a metasurface. Addressing the problem of phase-dissociated speckle patterns in multimode fiber optic outputs, which hinder imaging, this invention proposes a metasurface with phase-modulated nanocylinders as metaatoms. Modulating the incident light field with this metasurface significantly reduces optical field distortion caused by the multimode fiber, further forming a focused light field at the output end of the multimode fiber. Combined with a few other simple optical components, scanning imaging of the target object can be achieved. The specific steps of this multimode fiber optic endpoint focusing method based on a metasurface are as follows:

[0022] S1. Construct a multimode fiber point-focusing optical path based on SLM:

[0023] Figure 1 The multimode fiber point focusing optical path based on SLM described in S1 is shown, including a laser 1, a collimating beam expander 2, a polarizing beam splitter 3, an SLM 4, a focusing objective lens 6, a multimode fiber 7, an objective lens 8, a convex lens 9, and a CCD camera 10 arranged sequentially along the optical path; the two ends of the multimode fiber 7 are respectively connected to the focusing objective lens 6 and the objective lens 8; the SLM 4 and the CCD camera 10 are respectively connected to a computer.

[0024] Laser 1 emits 532nm green light, which is collimated and expanded into parallel light by collimating and beam expander 2. The parallel light, after passing through polarizing beam splitter 3, becomes P-polarized light and is incident on SLM4. The phase-modulated light from SLM4 is then refracted by polarizing beam splitter 3 and enters focusing objective 6. Polarizing beam splitter 3 satisfies the specific linear polarization angle of SLM4 while also reducing the intensity of unmodulated reflected light. Focusing objective 6 couples the refracted light into a 50μm core diameter multimode fiber 7. The light emanating from multimode fiber 7 is received by objective 8. Objective 8 and convex lens 9 together function as a microscope, ultimately imaged onto CCD camera 10 from the light emanating from convex lens 9.

[0025] S2. The computer repeatedly accesses SLM4 and CCD camera 10 using a sequential iterative optimization method to solve for the loading phase of SLM4 that makes the output end of multimode fiber 7 a focused light field. Figure 2 The flowchart for the sequential iterative optimization method is as follows:

[0026] S21. Calibrate the CCD camera's 10-pixel equivalent:

[0027] Adjust the position of the output end of the multimode fiber 7 so that the output port is clearly imaged onto the CCD camera 10. Simultaneously, the center of the field of view of the CCD camera 10 is the center of the multimode fiber 7. The CCD camera 10 acquires images, and by calculating the number of pixels e occupied by the radius of the output port of the multimode fiber 7, the actual physical size d corresponding to a single pixel of the CCD camera 10 is:

[0028]

[0029] The unit of d is μm.

[0030] S22. Determine the focusable area: The light intensity distribution at the output end of the multimode fiber 7 is within a circular area centered on the fiber's center and with a diameter equal to the fiber's core diameter of 50 μm. Assume the image captured by the CCD camera 10 is a square with a side length of (1 + 2 * e) pixels. The center position of the multimode fiber 7 is then (e + 1, e + 1). Let the target focus point be (x, y), satisfying:

[0031]

[0032] S23. Select the exposure time of CCD camera 10: Before SLM4 phase optimization, adjust the exposure time of CCD camera 10 to prevent overexposure of the image, and set the initial exposure time to t.

[0033] S24. Partition SLM4: Take the central square region of SLM4 with a side length of l. SLM Divide it into M*N sub-regions according to rows and columns, and load a 0-phase in each sub-region. The length of each sub-region is l. M and width l N They are respectively:

[0034]

[0035] S25. Modulation optimization of the phase of the sub-region: Select 8 phases: 0, π / 4, π / 2, 3π / 4, π, 5π / 4, 3π / 2, and 7π / 4. Select the optimal phase from among them. The specific steps are as follows:

[0036] S251. Select the first modulation sub-region, load phase 0, and acquire an image after waiting for a period of time.

[0037] S252. Determine if there is a pixel with a grayscale value greater than 240. If so, halve the exposure time; otherwise, keep the exposure time unchanged and record the number of times the exposure time is halved, b. The exposure time T at this point is:

[0038]

[0039] S253. Record the gray value of the target point as g(x,y), and calculate the value of the target fitness function G.

[0040]

[0041] Where g(i,j) is the gray value at position (i,j).

[0042] S254. Increase the phase by π / 4, wait for a period of time, and then acquire an image. Repeat S252 to S253 to obtain the G value corresponding to the current phase.

[0043] S255. Determine if all phases have been loaded: If yes, select the phase corresponding to the maximum G value as the optimal phase for this sub-region. Where m and n represent the row and column of the current optimized sub-region, respectively. Otherwise, repeat S254 until all phases are loaded.

[0044] S26. Determine if all sub-regions have been loaded: If yes, save the optimal phase of the entire SLM4; otherwise, repeat S25 until all sub-regions are optimized. The final loading phase of SLM4 is as follows:

[0045] ,

[0046] Figure 3 That is, take the side length of the middle square of SLM4 as l. SLM A region of 9600μm (total pixels 1200*1200) is used as the total modulation region. This region is then divided into 30*30 sub-regions (M=30, N=30) with equal rows and columns. The target focal point is the image center. The loading phase of the SLM4 is obtained using a sequential iterative optimization method. After loading this phase onto the SLM4, the image received by the CCD camera 10 is... Figure 4 This enables the output end of the multimode fiber 7 to be a focused light field.

[0047] S3. Design metasurface 5 based on the loading phase of the solved SLM4.

[0048] S31. Set the loading phase of SLM4 to the phase distribution of metasurface 5.

[0049] S32, metasurface 5 is composed of several metaatoms arranged and combined. The computer scans the size of the metaatoms using the finite-difference time-domain method to obtain the metaatoms corresponding to the phase of metasurface 5. The specific steps are as follows:

[0050] S321. Determine the shape and material of the meta-atoms: Titanium dioxide, which has a high refractive index and low absorption loss in the visible light band, is selected as the material for both the meta-atoms and the substrate to improve the phase modulation capability. Nanocylinders are used as the shape of the meta-atoms; by changing the diameter of the nanocylinders at different positions, phase control is achieved. Figure 5 This is a simulated diagram of a single superatomic structure.

[0051] S322. Set up the simulation space: Set the XY direction as periodic boundary conditions to simulate the case of metaatoms in the metasurface, and set the perfect matching layer (PML) as the boundary condition in the Z direction to simulate the real space of infinite propagation.

[0052] S323. Set the light source: Determine the propagation direction of the light source as positive along the Z-axis, with a wavelength of 532nm.

[0053] S324. Scanning to obtain the phase modulation capability of metaatoms: Scanning the diameter d1 of the titanium dioxide nanopillars to obtain the phase modulation capability of metaatoms in the X direction at l M For period (l) M =320μm), Y direction with l N For period (l) N =320μm), the curve showing the change in phase modulation capability as a function of its diameter, as shown in the figure. Figure 6 As shown. The specific correspondence between the simulated metaatom diameter and its modulation phase is as follows:

[0054] Table 1. Correspondence between the diameter of superatoms and their modulation phase

[0055]

[0056] S33. Determine the final structure of the metasurface 5 that meets the requirements: The size of the metasurface 5 is consistent with the size of the SLM4 modulation region, i.e., the side length is l. SLM square (l) SLM =9600μm), and then the diameter of the nanopillars at the corresponding positions of the metasurface is selected according to the phase of each sub-region of SLM4. After arranging them, the final metasurface 5 is obtained.

[0057] S4. Replace SLM4 in the multimode fiber point focusing optical path based on SLM with metasurface 5 to realize that the output end of the multimode fiber is a focused optical field.

[0058] In summary, this invention uses metasurface 4 to phase modulate the incident light field of multimode fiber 7, so that the output end of multimode fiber 7 becomes a focused light field, providing preliminary conditions for subsequent point scanning imaging. When used in endoscopes, it can effectively reduce the probe size and has broad development prospects in the field of endoscopic imaging.

Claims

1. A method for focusing the output end of a multimode optical fiber based on metasurfaces, characterized in that, The steps are as follows: S1. Construct a multimode fiber point-focusing optical path based on SLM: The multimode fiber point focusing optical path of the SLM includes a laser (1), a collimating beam expander (2), a polarizing beam splitter (3), an SLM (4), a focusing objective (6), a multimode fiber (7), an objective (8), a convex lens (9), and a CCD camera (10) arranged sequentially along the optical path; the two ends of the multimode fiber (7) are respectively connected to the focusing objective (6) and the objective (8); the SLM (4) and the CCD camera (10) are respectively connected to a computer; The laser (1) emits 532nm green light, which is collimated and expanded into parallel light by the collimating beam expander (2). The P-polarized light after passing through the polarizing beam splitter (3) is incident on the SLM (4). The light phase-modulated by the SLM (4) is then refracted by the polarizing beam splitter (3) and enters the focusing objective (6). The focusing objective (6) couples the refracted light into a multimode fiber (7) with a core diameter of 50µm. The light emitted from the multimode fiber (7) is received by the objective (8). The objective (8) and the convex lens (9) together play a microscopic role, and finally the light emitted from the convex lens (9) is imaged onto the CCD camera (10). S2. The computer repeatedly accesses the SLM(4) and CCD camera(10) using a sequential iterative optimization method to solve for the loading phase of the SLM(4) that makes the output end of the multimode fiber (7) a focused light field, as follows: S21. Calibrate the pixel equivalent of the CCD camera (10): Adjust the position of the output end of the multimode fiber (7) so that the output port is clearly imaged onto the CCD camera (10). At the same time, the center of the field of view of the CCD camera (10) is the center of the multimode fiber (7). The CCD camera (10) acquires the image. By calculating the number of pixels e occupied by the radius of the output port of the multimode fiber (7), the actual physical size d corresponding to a single pixel of the CCD camera (10) is: ； The unit of d is µm; S22. Determine the focusable area: The light intensity distribution at the output end of the multimode fiber (7) is within a circular area centered on the fiber's center and with a diameter equal to the core diameter of the multimode fiber (7) of 50µm. Assume the image acquired by the CCD camera (10) is a square with a side length of (1+2*e) pixels. At this time, the center position of the multimode fiber (7) is (e+1, e+1). Let the target focus point be (x, y), satisfying: ； S23. Select the exposure time of the CCD camera (10): Before optimizing the phase of the SLM (4), adjust the exposure time of the CCD camera (10) to prevent overexposure of the image, and set the initial exposure time to t. S24. Divide SLM(4) into regions: Take the central square region of SLM(4) with a side length of . Divide it into M*N sub-regions according to rows and columns, and load 0 phase into each sub-region. The length of each sub-region is... Hekuan They are respectively: ； S25. Modulation optimization of the phase of the sub-region: Select 8 phases: 0, π / 4, π / 2, 3π / 4, π, 5π / 4, 3π / 2, and 7π / 4. Select the optimal phase from among them. The specific steps are as follows: S251. Select the first modulation sub-region, load phase 0, and acquire an image after waiting for a period of time. S252. Determine if there is a pixel with a grayscale value greater than 240. If so, halve the exposure time; otherwise, keep the exposure time unchanged and record the number of times the exposure time is halved (b). for: ； S253. Record the grayscale value of the target point as g(x, y), and calculate the value of the target fitness function G; ； Where g(i, j) is the gray value at position (i, j); S254. Increase the phase by π / 4, wait for a period of time, and then acquire an image. Repeat S252~S253 to obtain the G value corresponding to the current phase. S255. Determine if all phases have been loaded: If yes, select the phase corresponding to the maximum G value as the optimal phase for this sub-region. , where m and n represent the row and column of the current optimized sub-region, respectively. If not, repeat S254 until all phases are loaded. S26. Determine if all sub-regions have been loaded: If yes, save the optimal phase of the entire SLM(4); otherwise, repeat S25 until all sub-regions are optimized. The final loading phase of SLM(4) is as follows: ； S3. Design the metasurface (5) based on the loading phase of the solved SLM (4); S4. Replace the SLM (4) in the multimode fiber point focusing optical path based on SLM with a metasurface (5) to realize that the output end of the multimode fiber is a focused optical field.

2. The multimode fiber optic terminal focusing method based on metasurfaces according to claim 1, characterized in that, In step 3, the metasurface (5) is designed based on the loading phase obtained from the SLM (4). The specific steps are as follows: S31. Set the loading phase of SLM(4) to the phase distribution of metasurface(5); S32. The metasurface (5) is composed of several metaatoms arranged and combined. The computer scans the size of the metaatoms using the finite-difference time-domain method to obtain the metaatoms corresponding to the phase of the metasurface (5). The specific steps are as follows: S321. Determine the shape and material of the meta-atom: Select titanium dioxide as the material for the meta-atom and the substrate; use nanocylinders as the shape of the meta-atom, and achieve phase control by changing the diameter of the nanocylinders at different positions. S322. Set up the simulation space: Set the XY direction as periodic boundary conditions to simulate the case of metaatoms in the metasurface, and set the perfect matching layer as the boundary condition in the Z direction to simulate the real space of infinite propagation. S323. Set the light source: Determine the propagation direction of the light source as positive along the Z-axis, with a wavelength of 532nm; S324. Scanning to obtain the phase modulation capability of metaatoms: Scanning the diameter d1 of the titanium dioxide nanopillars to obtain the phase modulation capability of metaatoms in the X direction. For period, Y direction with The curves showing the phase modulation capability as a function of the diameter are given, with a fixed period and height h. The simulated relationship between the diameter of the metaatom and its modulation phase is as follows: When the metaatom's phase modulation capability is 0, its diameter is 0.5 μm; When the metaatom's phase modulation capability is π / 4, its diameter is 1.41 μm; When the metaatom's phase modulation capability is π / 2, its diameter is 1.79 μm; When the metaatom's phase modulation capability is 3π / 4, its diameter is 1.94 μm; When the metaatom's phase modulation capability is π, its diameter is 2.22 μm; When the meta-atom's phase modulation capability is 5π / 4, its diameter is 2.33 μm; When the metaatom's phase modulation capability is 3π / 2, its diameter is 2.41 μm; S33. Determine the structure of the final metasurface (5) that meets the requirements: The size of the metasurface (5) is consistent with the size of the modulation region of the SLM (4), that is, the side length is The square shape is then used to select the diameter of the nanopillars corresponding to the position of the metasurface according to the phase of each sub-region of SLM(4). After arranging them, the final metasurface(5) is obtained.

Images