A vortex structure light field transmission focusing regulation method and device based on large-scale scattering medium
By combining vortex structured light fields with machine learning and genetic algorithms, the focusing problem in large-scale scattering media was solved, achieving stable and low-cost focusing control, which is suitable for extreme environments.
Patent Information
- Application Number
- CN202411810148.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-10
- Publication Date
- 2026-02-13
- Estimated Expiration
- 2044-12-10
AI Technical Summary
Existing technologies struggle to effectively focus and image through large-scale scattering media, especially under extreme conditions such as rain and fog. Existing methods have limitations and complexities, making it impossible to effectively reconstruct target information.
By combining vortex structured light field with machine learning and genetic algorithms, a vortex structured light field is generated by wavefront phase modulation of laser light through a spatial light modulator. A speckle image dataset is collected, a phase mask is trained, and iterative optimization is performed using a single-layer neural network and genetic algorithm to achieve focusing control.
It achieves stable focusing on large-scale scattering media, has good anti-interference ability, simple structure, low cost, strong adaptability, and can maintain good focusing effect under light source fluctuations and external environmental interference.
Smart Images

Figure CN119355960B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of wavefront shaping laser control technology, and in particular to a method and apparatus for focusing and controlling a vortex structure light field through a large-scale scattering medium. Background Technology
[0002] Scattering phenomena frequently occur in scenarios such as heavy fog, rain, snow, turbid water, and smoke, severely impacting fields such as traffic safety, underwater rescue, fire rescue, and satellite remote sensing. Due to the uneven distribution of refractive index within the scattering medium, light waves undergo strong multiple scattering during propagation, resulting in blurred images and complete loss of target information. Furthermore, current imaging methods using static scattering media are unable to address macroscopic, large-scale imaging problems, causing significant negative impacts on fields such as traffic safety, underwater exploration, and biomedical imaging.
[0003] The aim of this technology is to enable light beams to be focused through scattering media such as deep biological tissue, turbid liquids, and atmospheric turbulence. Existing iterative optimization techniques and their variants are only used to focus light through thin scattering media such as white paint layers, eggshells, frosted glass, and biological tissues, which has very limited applicability and cannot effectively improve focusing and imaging problems in extreme rain and fog conditions. The invention patent with application number CN201910563496.6 is based on a phase retrieval algorithm and applied to imaging small-scale dynamic scattering media such as rotating frosted glass, skim milk, a mixture of partially skim milk and water, and zinc oxide solution. This method is effective for processing small-scale scattering media, but it cannot reconstruct target information through large-scale scattering media. The invention patent with application number CN202210891701.3 is based on an emerging deep learning method and applied to imaging fat emulsion solutions, artificial fog, and milk solutions. This method shows good descattering ability, but it is more effective in thin fog scenes and the processing process is relatively complex. The invention patent with application number CN 202310711350.8 is based on a speckle autocorrelation method applied to focusing and imaging on rotating frosted glass. Due to the long iteration time, its versatility is still insufficient, especially in extreme rain and fog conditions, where the effective feature information of the target is very limited. Therefore, developing focusing control methods applicable to macroscopic large-scale scattering media still faces significant challenges. Summary of the Invention
[0004] In view of this, the purpose of the present invention is to provide a method and device for focusing and controlling the light field through a large-scale scattering medium based on a vortex structure light field. The method uses a spatial light modulator to modulate the wavefront phase of the laser light input to the scattering medium, so that the speckle image formed through the scattering medium changes with the change of the wavefront phase.
[0005] To achieve the above object, the application adopts the following technical scheme: a vortex structured light field transmission focusing regulation method based on large-scale scattering medium, comprising the following steps:
[0006] Step 1, generation of vortex structured light field; vortex phase is loaded into a spatial light modulator, and combined with circularly polarized light to generate a vortex structured light field;
[0007] Step 2, collection of speckle image dataset; a sCMOS camera collects speckle images passing through the scattering medium and self-builds an image dataset;
[0008] Step 3, training of single-layer neural network; the mapping relationship between phase mask and speckle image is trained as a training model;
[0009] Step 4, pre-focusing of single-layer neural network; the phase mask obtained by the training model is loaded into the spatial light modulator to test the effectiveness of the training model;
[0010] Step 5, iterative optimization of genetic neural network under vortex structured light field; the predicted phase mask is combined with genetic algorithm to realize speckle focusing under vortex structured light field through multiple iterations.
[0011] In a preferred embodiment, the spatial light modulator is a reflective phase spatial light modulator.
[0012] In a preferred embodiment, the phase modulation of the spatial light modulator is represented as:
[0013]
[0014] wherein, x n is the x-direction position of the nth pixel area of the spatial light modulator, y n is the y-direction position of the nth pixel area of the spatial light modulator, E in (x n ,y n ) is the input light field of the spatial light modulator, φ n is the phase delay added by the nth pixel area of the spatial light modulator, E out (x n ,y n ) is the modulated light field, and i is the imaginary unit.
[0015] In a preferred embodiment, the phase of the vortex structured light field is a spiral phase, and the electric field expression of the vortex structured light field is:
[0016]
[0017] wherein, E0 is the amplitude, θ is the azimuth angle, k is the wave number, is the phase factor, q is the topological charge, and z is the propagation distance.
[0018] In a preferred embodiment, the water mist concentration is characterized by an optical thickness range, and the relationship is expressed as:
[0019]
[0020] Where I0 is the initial incident light intensity, L is the thickness of the scattering medium, τ is the optical thickness, and μ is the extinction coefficient.
[0021] In a preferred embodiment, the step 2 data set includes a mapping relationship between the speckle image and the phase mask.
[0022] In a preferred embodiment, the light field expression of the mapping relationship is:
[0023]
[0024] Where Y is the light field of the scattered speckle, and X is the light field before the scattering medium.
[0025] In a preferred embodiment, in step 4, based on the trained model, the predicted phase mask is output and loaded into the spatial light modulator to achieve pre-focusing.
[0026] In a preferred embodiment, in step 5, the genetic neural network is a hybrid algorithm of a single-layer neural network and a genetic algorithm; after the optimal phase mask is predicted by the single-layer neural network in step 5, the genetic algorithm is used for iterative optimization.
[0027] The application also provides a vortex structure light field focusing device based on large-scale scattering medium, and a vortex structure light field focusing and regulating method based on large-scale scattering medium; including: a laser, an attenuator, an aperture, a beam expander, an aperture, a lens, a spatial filter, an aperture, a reflective spatial light modulator, a mirror, a polarizer, a reflective spatial light modulator, a quarter-wave plate, a lens, a lens, a scattering medium, and an sCMOS camera arranged in sequence along the light path direction;
[0028] The laser, attenuator, aperture, beam expander, aperture, lens, and spatial filter are used to generate a coherent light beam.
[0029] The aperture and reflective spatial light modulator are used for phase regulation of the coherent light beam.
[0030] The polarizer, reflective spatial light modulator, and quarter-wave plate are used to generate a circularly polarized vortex structure light field.
[0031] The lenses are used for spot compression of the light beam.
[0032] The scattering medium and the sCMOS camera are used for imaging speckles to form a speckle image.
[0033] Compared with the prior art, the present application has the following beneficial effects:
[0034] 1) The present application uses the correlation between speckle images and the focusing control wavefront phase, uses the current speckle image to predict the wavefront phase required for the current speckle focusing, combines the excellent characteristics of vortex structured light field transmission through large-scale scattering media, and completes the learning process of genetic neural network through multiple iterations and obtains the best focusing wavefront phase of the speckle image.
[0035] The traditional focusing technology through scattering media is still limited to thin scattering media, and it is basically difficult to achieve focusing for large-scale scattering media (cloud layer, turbid water body, thick fog, etc.) in real scenes. Through the means of searching for the wavefront phase based on the vortex structured light field in the above scheme, the focusing difficulty problem in the speckle regulation task through large-scale scattering media is solved.
[0036] And the existing feedback iteration method in this field converges quickly but is easy to fall into local optimum; based on this, the learnability of machine learning can provide a pre-focused wavefront phase in the regulation task, which facilitates the rapid convergence of genetic algorithm and is not easy to fall into local optimum, and the method can work normally under ordinary parameters.
[0037] 2) The present application uses different speckle images and corresponding learned wavefront phases to update the method, when the light source fluctuation in the collection light path and other external environmental interference cause the change of the illumination light field, the method still has good focusing effect on the existing speckle image, and has good anti-interference ability.
[0038] 3) The scattering medium focusing regulation device based on reinforcement learning of the present application has the advantages of simple structure, high stability and low cost. DETAILED DESCRIPTION
[0039] Figure 1 Fig. 1 is a structural schematic diagram of a vortex structured light field-based focusing regulation device through large-scale scattering media provided in an embodiment of the present application.
[0040] Figure 2 Fig. 2 is a phase grating with a topological charge of 1 loaded on a reflective spatial light modulator.
[0041] Figure 3 Fig. 3 is a diagram of focusing regulation of (a) a target corresponding speckle image, (b) a Gaussian beam (topological charge 0), and (c) a vortex structured light field (topological charge 1).
[0042] Figure 4For the topological charge of 0 and 1, the corresponding focal point center section expansion curve is compared. DETAILED DESCRIPTION
[0043] The application will be further described below in conjunction with the accompanying drawings and examples.
[0044] It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the application. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
[0045] It should be noted that the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application; as used herein, the singular form is intended to include the plural form unless the context clearly indicates otherwise, and furthermore, it should be understood that when the terms "comprise" and / or "include" are used in the specification, there is a feature, step, operation, device, component and / or combination thereof.
[0046] Reference Figures 1-4 The embodiment provides a vortex structured light field transmission focusing regulation method based on large-scale scattering medium, which comprises the following steps:
[0047] Step 1, generation of vortex structured light field: load the required vortex phase into a spatial light modulator, and combine it with circular polarization to form a vortex structured light field;
[0048] As shown in Figure 2 The vortex phase is a helical phase pattern with a topological charge of 1; the change of the polarization state of the vortex structured light field, i.e. rotating the quarter-wave plate 13, can obtain the required polarization state distribution of the light beam by changing the included angle.
[0049] Step 2, collecting speckle image dataset: sCMOS camera collects speckle images through the scattering medium and builds an image dataset;
[0050] It should be noted that the speckle image dataset acquisition link does not load the helical phase.
[0051] Step 3, training of single-layer neural network: training the mapping relationship between phase mask and speckle image as a training model;
[0052] It should be noted that the training of the single-layer neural network aims to learn the mapping function between the speckle pattern and the phase mask, rather than directly modeling the scattering medium. Therefore, no test set is established for testing.
[0053] Step 4, iterative learning of each speckle image based on the feedback regulation method of machine learning;
[0054] The method for regulating described in step 2 includes a machine learning algorithm and controlling a spatial light modulator to load a phase segment.
[0055] Specifically, the feedback regulation method based on machine learning obtains a real-time speckle image through speckle image acquisition, uses the current speckle image to search for a focused wavefront phase through a machine learning algorithm, loads the searched focused wavefront phase through a spatial light modulator, and finally calculates a reward value required for updating a reinforcement learning algorithm according to a focusing condition combined with the speckle image focusing condition of the vortex structured light field.
[0056] It should be noted that the speckle image acquisition segment is not loaded with a spiral phase.
[0057] The current method searches for a focused regulation optimal wavefront phase each time an iteration is output;
[0058] The feedback regulation method based on reinforcement learning outputs a focused wavefront phase required for focusing the current speckle image each time an iteration is performed.
[0059] Step 5: The searched wavefront phase is loaded through a spatial light modulator, and the speckle image acquired by an sCMOS camera is updated, and after repeating multiple iterations, the optimal wavefront phase region searched by the method is stable and speckle focusing is achieved;
[0060] In an optional embodiment, the processor is used to transmit the focused wavefront phase searched by the feedback regulation method based on reinforcement learning to the spatial light modulator, and the speckle image after loading the new wavefront phase is acquired in real time through the sCMOS camera, the feedback regulation method stores historical records in the iteration process, and the historical records are used for method updating, and finally the optimal wavefront phase required for real-time speckle focusing is obtained by inputting a real-time speckle image.
[0061] Specifically, the reward value required for calculation in the feedback regulation method based on machine learning is a background signal-to-noise ratio, which is expressed as:
[0062]
[0063] Wherein, f is the reward value, I n is the light intensity value of a pixel point in a target region, I m is the light intensity value of a pixel point in a background region, n is the total number of pixel points in the target region, and m is the total number of pixel points in the background region.
[0064] It should be noted that the definitions of the target region and the background region are as follows:
[0065]
[0066] Wherein, (x0, y0) is the center coordinate of the target point, r1 is the radius of the focused target region, and r2 is the radius of the focused background region.
[0067] The vortex structured light field based focusing regulation method through large-scale scattering medium of the embodiment of the application utilizes the correlation between the wavefront phase and the speckle image, learns and searches the phase of different speckle images through a machine learning algorithm under the vortex structured light field, and continuously updates the algorithm to obtain a regulation method suitable for the focusing of the speckle image in the experimental light path and a focusing wavefront phase.
[0068] The vortex structured light field based focusing regulation method through large-scale scattering medium of the embodiment of the application utilizes the machine learning algorithm to learn the relationship between the speckle image and the focusing wavefront phase under the vortex structured light field to realize the focusing regulation of the speckle, greatly improves the local optimal problem of the wavefront phase search, and greatly improves the focusing effect through the real scene (thick fog, etc.), has better universality, and can quickly adapt to the new speckle image when the light source in the experimental light path slightly fluctuates.
[0069] As shown in Figure 1 The vortex structured light field based focusing regulation device through large-scale scattering medium of the embodiment of the application comprises, in sequence along the light path direction, a solid laser 1, an attenuation sheet 2, an aperture 3, a spatial filter 4, a lens 5, an aperture 6, a beam expander 7, an aperture 8, a spatial light modulator 9, a mirror 10, a polarizer 11, a spatial light modulator 12, a quarter-wave plate 13, a lens 14, a lens 15, a scattering medium 16, an sCMOS camera 17, and a processor 18.
[0070] The solid laser 1, the attenuation sheet 2, the spatial filter 4, the lens 5, and the beam expander 7 are used to generate a uniform coherent light beam.
[0071] In the embodiment, the solid laser 1 is used, and the laser wavelength is 635 nm; the attenuation sheet 2 is used to weaken the laser intensity to avoid the light intensity being too bright when passing through the scattering medium; the spatial filter 4 and the lens 5 are used to eliminate the stray light around the laser; and the beam expander 7 and the aperture 8 are used to obtain a coherent light beam that can cover the entire center screen of the spatial light modulator.
[0072] The reflective spatial light modulator 9 is used to regulate the phase of the coherent light beam.
[0073] The polarizer 11, the spatial light modulator 12, and the quarter-wave plate 13 are used to generate a vortex structured light field with a required polarization state.
[0074] In the embodiment, the polarizer 11 and the quarter-wave plate 13 are used to obtain a circularly polarized laser beam; and the spatial light modulator 12 is loaded with a spiral phase to generate a vortex structured light field. The angle difference between the polarizer 11 and the quarter-wave plate 13 is 45 degrees.
[0075] The lens 14, the lens 15, the scattering medium 16 and the sCMOS camera 17 are used for imaging the speckle to form a speckle image.
[0076] In the embodiment, the lens is a convex lens, the focal length of the lens 14 is 150mm, the focal length of the lens 15 is 27.5mm, and the lens 14 and the lens 15 can make the light spot transmitted through the scattering medium smaller. The sCMOS camera adjusts the imaging focal plane to be at 150mm behind the lens, so as to obtain a clearer speckle image.
[0077] The processor is used for controlling the implementation of the method and the control of the spatial light modulator and the sCMOS camera.
[0078] In the embodiment, the processor is connected with the spatial light modulator and the sCMOS camera simultaneously, and allows the spatial light modulator to be operated to load the focusing wavefront phase and the sCMOS camera to be operated to collect the corresponding speckle image based on the vortex structured light field transmitted through the large-scale scattering medium.
[0079] The vortex structured light field transmission focusing regulation device based on the large-scale scattering medium according to the embodiment of the application has the advantages of simple structure, high stability and low cost.
[0080] The above description is only the preferred embodiment of the application, and does not limit the application in other forms. Any person skilled in the art can modify or change the above disclosed technical content into equivalent embodiments. However, any simple modification, equivalent change and modification made according to the technical essence of the application without departing from the technical solution of the application still belongs to the protection scope of the application.
Claims
1. A method for focusing and controlling the light field through a large-scale scattering medium based on a vortex structure, characterized in that, Includes the following steps: Step 1: Generation of vortex structured light field; The vortex phase is loaded onto the spatial light modulator and combined with circularly polarized light to generate a vortex structured light field; Step 2: Acquire speckle image dataset; sCMOS camera acquires speckle images through the scattering medium and builds its own image dataset; Step 3: Training a single-layer neural network; The mapping relationship between the phase mask and the speckle image is trained as a training model; Step 4: Pre-focusing of a single-layer neural network; The phase mask obtained from the training model is loaded onto the spatial light modulator to test the effectiveness of the training model; Step 5: Iterative optimization of genetic neural network under vortex structured light field: The predicted phase mask is combined with the genetic algorithm to achieve speckle focusing through multiple iterations under vortex structured light field; The spatial light modulator is a reflective phase spatial light modulator; The phase modulation of the spatial light modulator is represented as follows: Where, x n Let x be the x-direction position of the nth pixel region of the spatial light modulator, and y be the x-direction position of the pixel region. n E represents the y-direction position of the nth pixel region of the spatial light modulator. in (x n ,y n φ represents the optical field input to the spatial light modulator. n The amount of phase delay added to the nth pixel region of the spatial light modulator, E out (x n ,y n () represents the modulated optical field, and i is the imaginary unit; The phase of the vortex structured light field is a spiral phase, and the electric field expression of the vortex structured light field is: Where E0 is the amplitude, θ is the azimuth angle, and k is the wave number. Let q be the phase factor, q be the topological charge number, and z be the propagation distance; The concentration of the scattering medium, i.e., water mist, is characterized by the optical thickness range, and the relationship expression is as follows: Where I0 is the initial incident light intensity, L is the thickness of the scattering medium, τ is the optical thickness, and μ is the extinction coefficient.
2. The method for focusing and controlling the light field through a large-scale scattering medium based on a vortex structure, as described in claim 1, is characterized in that... The dataset in step 2 contains the mapping relationship between speckle images and phase masks.
3. The method for focusing and controlling the light field through a large-scale scattering medium based on a vortex structure, as described in claim 2, is characterized in that... The light field expression for the mapping relationship is: Where Y is the light field of the scattered spot, and X is the light field in front of the scattering medium.
4. The method for focusing and controlling the light field through a large-scale scattering medium based on a vortex structure, as described in claim 1, is characterized in that... In step 4, based on the trained model, a predicted phase mask is output and loaded into the spatial light modulator to achieve pre-focusing.
5. The method for focusing and controlling the light field through a large-scale scattering medium based on a vortex structure, as described in claim 1, is characterized in that... In step 5, the genetic neural network is a hybrid algorithm of a single-layer neural network and a genetic algorithm; after the optimal phase mask is predicted by the single-layer neural network in step 5, it is iteratively optimized by the genetic algorithm.
6. A focusing device for light fields transmitted through a large-scale scattering medium based on a vortex structure, characterized in that, The above-described method for focusing and controlling the optical field through a large-scale scattering medium based on a vortex structure, as described in any one of claims 1-5, includes: a laser, an attenuator, an aperture, a beam expander, an aperture, a lens, a spatial filter, an aperture, a reflective spatial light modulator, a mirror, a polarizer, a quarter-wave plate, a lens, a scattering medium, and an sCMOS camera arranged sequentially along the optical path. The laser, attenuator, aperture, beam expander, aperture, lens, and spatial filter are used to generate a coherent beam. The aperture and reflective spatial light modulator are used to phase-modulate the coherent beam. The polarizer, reflective spatial light modulator, and quarter-wave plate are used to generate a circularly polarized vortex structured light field. The lens is used to compress the beam spot. The scattering medium and sCMOS camera are used to image the speckle pattern and form a speckle image.
Citation Information
Patent Citations
Scattering imaging device and method penetrating through dynamic liquid media
CN110333205A
Experimental data acquisition and deep learning method and system for imaging through scattering medium
CN115237365A
Transmitting scattering medium imaging method based on improved HIO-ER algorithm
CN116630202A
Vortex light information compressed sensing device and method
CN115327771A
Method and device for realizing light field vector regulation and control through scattering medium
CN117406463A