34 results about "Spatial coherence" patented technology

The invention discloses an acoustic emission activity monitoring method and an acoustic emission activity monitoring system for early warning of internal damage of concrete, particularly relates to the technical field of acoustic emission monitoring, and is used for solving the problems of large positioning result deviation and low reliability caused by non-uniformity of a medium in the concrete in an existing acoustic emission positioning method. Spatial coherence characteristics are generated by analyzing spatial coherence of acoustic emission signals among acceleration sensor arrays, preliminary positioning is carried out in combination with time difference of arrival and sensor coordinates, and multi-dimensional characteristic vectors containing the spatial coherence characteristics and preliminary positioning results are extracted; and carrying out clustering analysis on the current damage event feature vector and the historical event feature vector to determine a reliability level, and carrying out weighted correction on a preliminary positioning result based on the reliability level to finally obtain an accurate damage source space coordinate.

The invention discloses a coal mine tunneling roadway advanced geological exploration and ground stress inversion method, particularly relates to the technical field of seismic signal processing in geophysical exploration, and is used for solving the problems of geological exploration signal distortion and insufficient stress inversion precision caused by strong noise interference of tunneling equipment in the prior art. Rock mass vibration signals are collected through an optical fiber sensor arranged on a roadway wall, equipment noise characteristics are identified through spectral analysis, a quantitative mapping relation between working condition parameters and noise characteristics is established, a noise reference time period is determined according to the quantitative mapping relation, a spatial coherence characteristic template is constructed, high-coherence noise components are inhibited through template comparison, and the noise reference time period is determined according to the spatial coherence characteristic template. Finally, the geological structure and stress field distribution in front of the working face are inverted based on the purified seismic wave signals, a stress concentration area and a disaster risk area are identified, the signal-to-noise ratio of the seismic signals in the strong noise environment and the reliability of the geological inversion result are effectively improved, and accurate technical support is provided for safe mining of a coal mine.

The embodiment of the invention provides a light field space coherence length measurement method and system, and relates to the technical field of optical measurement. The method may first obtain a training set and train a measurement model using the training set. And obtaining an interference fringe image through numerical simulation generation or experimental acquisition, and inputting the interference fringe image as an input image into the measurement model so as to output a spatial coherence length prediction value of the input image through the measurement model. Wherein the measurement model is a convolutional neural network model for performing feature extraction by adopting a multi-layer cascaded convolutional pooling structure, and the features can be mapped into a spatial coherence length predicted value through a full connection layer. According to the method, one-step measurement from the original interference image to the coherence length can be realized, a tedious fringe contrast calculation process is avoided, the influence of experimental noise on measurement precision is reduced through depth feature extraction, the whole prediction process can be completed within millisecond-level time, and the measurement efficiency of the light source space coherence length is improved.

The invention relates to the field of speech analysis, and discloses a natural sound interference stripping algorithm and an application thereof in noise monitoring, which are used for efficiently and accurately stripping interference components in a target speech signal in a complex natural sound environment. Comprising the following steps: synchronously acquiring an environmental sound signal by using a spatially distributed microphone array, generating an original mixed voice signal with spatial coherence, performing real-time fractional order differential processing, strengthening resonance characteristics, realizing spatial separation of a voice guided wave mode and an interference radiation mode through real-time phase analysis, and extracting an interference radiation mode component. And reconstructing a time domain interference waveform, and outputting a pure target voice signal according to the original mixed voice signal and the reconstructed interference waveform. The system has the functions of residual interference sound energy recovery, equipment vibration suppression, environment sudden change monitoring, system resetting, bimodal output and the like, the stability, reliability and practicability of the system are effectively improved, and pure voice can be accurately extracted in a complex environment.

The present application relates to the technical field of ultrasonic nondestructive testing and industrial pipeline safety evaluation, in particular to a kind of stainless steel pipeline weld internal defect detection system, comprising: obtaining ultrasonic multi-angle scanning data, extracting waveform features and sound path information;Anisotropic propagation model is constructed, and sound beam deflection correction factor is solved;Generate corrected sound field spatial distribution data;Calculate the multi-view spatial coherence coefficient representing the consistency of signal phase under different sound beam paths;If the multi-view spatial coherence coefficient is greater than the coherence threshold, then determine that the corresponding signal is a defect echo, and reconstruct the defect spatial morphology;Otherwise, it is determined that the corresponding signal is structural scattering noise, and noise suppression processing is performed;Output the quality detection report of target component;The present application solves the missed detection hidden danger caused by geometric constraint and material attenuation, and ensures that the detection result of nuclear power and petrochemical high-risk pipeline under complex physical conditions meets the stringent full-coverage quality standard.

A ray trace operation includes tracing a ray from an origin point in accordance with a ray path into a virtual environment (where the virtual environment comprises one or more virtual objects defined by one or more object components) and determining an intersected object component of the one or more object components that the ray intersects with. Determining the intersected object component comprises accessing (i) ray trace temporal coherence data based upon a preceding ray trace operation that temporally precedes the ray trace operation or (ii) ray trace spatial coherence data based upon a spatially proximate ray trace operation.

The application discloses a light uniforming and speckle eliminating system using an optical waveguide and an inverted-pyramid-shaped reflecting cavity, which comprises a hollow reflecting circular truncated cone, a circular truncated cone-shaped hollow reflecting cavity, an optical waveguide bundle, an inverted-pyramid-shaped hollow reflecting cavity and a light uniforming sheet, the hollow reflecting circular truncated cone is arranged at a light inlet of the circular truncated cone-shaped hollow reflecting cavity, and the optical waveguide bundle is arranged between the circular truncated cone-shaped hollow reflecting cavity and the inverted-pyramid-shaped hollow reflecting cavity or encapsulated in the inverted-pyramid-shaped hollow reflecting cavity.In the application, the light reflected by the circular truncated cone-shaped hollow reflecting cavity is incident into the optical waveguide bundle at a certain angle with the horizontal optical axis, so that the spatial coherence of the laser is reduced, and the speckle is inhibited; the light incident into the optical waveguide bundle is continuously reflected and scattered in the waveguide to further inhibit the speckle noise, and the light uniforming effect is achieved; and the light is uniformly lighted again by the light uniforming sheet, so that the light uniforming and speckle eliminating system can be used for light uniforming and speckle eliminating of a large-aperture collimated light beam, no vibration is generated, the cost is low, and the system is suitable for industrial popularization and use.

The invention relates to the technical field of breast ultrasonic image recognition, in particular to an ultrasonic puncture part anomaly detection method and system based on image processing. Acquiring all similar pixel points of the reference pixel points in the breast ultrasound image of the patient; and calculating a spatial coherence coefficient according to the distance distribution between the reference point and each similar point, the number of the similar points and the regional distribution characteristics. And identifying a high-echo region in combination with gray level distribution of the pixel points. For each high echo region, detecting a mark inflection point and a non-mark inflection point; calculating an abnormal extension coefficient according to the position distribution of the mark points and the non-mark points; and obtaining an abnormal factor, and screening abnormal inflection points. The clustering abnormal inflection point forms an abnormal clustering cluster based on an abnormal extension coefficient and an abnormal factor; and performing contour fitting on the abnormal inflection points in each cluster to obtain a suspected lump area. According to the method, the suspected lump area can be accurately identified, so that the detection efficiency and accuracy of related personnel are improved.

The invention discloses a complex test scene perception and partition construction method based on a multi-modal large model, and belongs to the technical field of digital test verification of high-end equipment. According to the method, a scene description prompt is designed, a multi-modal large model is utilized to generate a 2D test scene image with an equidistant view angle, and images with consistent semantics are screened through similarity calculation. Secondly, extracting spatial positions, semantic information and three-dimensional geometric parameters of objects in the scene by adopting a detection segmentation model and a depth estimation technology, and generating a feature map fusing space and semantics; and finally, geometric and texture description of an object is generated based on a multi-modal large model, a 3D grid is constructed by using a 3D generation model, model parameters are adjusted through dense matching and iterative optimization, and category accuracy and spatial coherence are ensured. According to the method, the full-process automation from 2D scene generation to 3D refined reconstruction is realized, and the precision and efficiency of complex test scene construction are remarkably improved.

This invention discloses an adaptive anti-glare lighting method, aiming to resolve the technical contradiction between measurement efficiency, reconstruction accuracy, and control robustness in dynamic anti-glare control. The method includes: obtaining sparse angular coherence observations through online measurement; rapidly reconstructing the current-cycle angular coherence matrix from the sparse observations based on the inherent low-rank, positive semi-definite structural priors of the angular coherence matrix, and quantifying its angular domain uncertainty; constructing and solving an optimization problem with modal weight increments as decision variables based on the reconstructed matrix and uncertainty, generating coherent modal weights before execution; mapping these weights to an actuator instruction sequence and issuing it, while simultaneously collecting execution records for the next cycle. This invention achieves rapid and accurate sensing and robust closed-loop control of the spatial coherence state of the light field by combining sparse sensing, structured reconstruction, and uncertainty-aware optimization control, thus suppressing dynamic glare.

The invention relates to a method and device for measuring the concurrency of vector light beams based on single speckle spatial correlation, and the method comprises the steps: scattering a to-be-measured vector light beam into a spatial incoherent light field, converting the spatial incoherent light field into a partially coherent light field through Fourier transform, collecting an original speckle intensity distribution map in the central region of a Fourier transform plane, and calculating the concurrency of the vector light beam; the method comprises the following steps of: performing background correction and spatial self-correlation operation on the optical concurrency to obtain a normalized spatial correlation distribution diagram, reading a zero displacement point value as a correlation value, calculating the concurrency according to a linear relation between the correlation value and the optical concurrency, and comprising rotary frosted glass, a lens, an image sensor and a calculation module. The concurrency degree can be output in real time only through single-time speckle shooting and single-point correlation reading, the system complexity is remarkably reduced through linear optical device combination, high precision and high robustness are kept in turbulent flow and other complex medium environments, and an efficient measurement scheme is provided for optical communication and other dynamic scenes.

The invention discloses an imaging method for passing through a random scattering medium in visible light based on a diffractive optical neural network, and the method comprises the following steps: 1, carrying out the optical coding of an input image through a coherent or low-coherent monochromatic plane wave, and enabling a coded light field to pass through a random phase diffuser for scattering; and step 2, continuous wavefront modulation of the scattered light field is obtained through a coherent or incoherent diffractive optical neural network, and finally, high-fidelity original image reconstruction is generated on an output plane. The invention finds that the coherent diffraction neural network obtains dynamic phase modulation by introducing a randomly generated diffuser in the training process, enhances the adaptive capacity of the network to the spatial coherence change of the light source, realizes the high-quality reconstruction of the scattering image under the visible light, has the robustness to the low-coherence light source, and is suitable for large-scale popularization and application. The method can be applied to low-power-consumption and real-time imaging of the dynamic scattering medium under natural light.

This invention discloses a novel multimodal semantic-spatial representation method for embodied perception and spatial reasoning, belonging to the field of embodied perception. Addressing the two major pain points of existing methods—insufficient spatial coherence in semantic reasoning and poor relation consistency and interpretability—this invention designs an optimization mechanism: on the one hand, it designs a transitive relation learning mechanism to correct relation conflicts and missing relations in reasoning under weak supervision, improving the completeness and logical consistency of spatial representation; on the other hand, relying solely on RGB images and depth maps, it integrates monocular depth estimation as a spatial modality into embodied perception, injecting implicit three-dimensional spatial attributes into the semantic structure, constructing a unified multimodal semantic-spatial representation, and enhancing the spatial coherence of semantic reasoning. This invention improves the performance of structured scene modeling and spatial grounding description generation through advanced spatial representation, and can be widely used in embodied spatial perception scenarios.

The invention discloses a confocal optical path system and method for suppressing stray light, and belongs to the technical field of optical precision measurement. According to the invention, the beam expander is moved forwards by 2-5 mm relative to the flat beam splitter prism, so that the light beam incident to the core beam splitter is actively converted into the Gaussian beam with the specific divergence angle from the parallel light, and is coupled with the symmetrically arranged confocal light paths, thereby solving the problem that in the existing heterodyne laser vibration measurement technology, the parallel light generates multi-path reflection on the surface of a planar optical element, so that the measurement precision is greatly improved. In the prior art, a stable parasitic interference cavity is formed, high-order and dynamic nonlinear errors and phase drift are introduced, and the technical problems of system complexity, high cost and difficult installation and adjustment caused by traditional suppression means (such as off-axis design and an ultrahigh antireflection film) are solved, the spatial coherence of stray light and main signal light is destroyed, the stray light intensity in the system is suppressed by about two orders of magnitude, and the system stability is improved. The non-linear distortion of a measurement signal is significantly reduced, and the adjustment tolerance and long-term stability of an optical path are improved at the same time.

The invention provides a liquid die forging defect detection method and system, and the method comprises the steps: transmitting an ultrasonic wave to the interior of a liquid die forging through an ultrasonic phased array probe, and collecting a plurality of ultrasonic signals of the ultrasonic wave transmitted in the liquid die forging; dividing all the ultrasonic signals into direct waves at the initial stage of ultrasonic propagation and scattered waves after complete ultrasonic diffusion according to time points of change of an ultrasonic propagation mechanism in the liquid die forging during each ultrasonic emission; the method comprises the following steps: performing near field effect suppression on each scattered wave through spatial coherence among all the scattered waves so as to obtain a reconstructed signal matrix which retains defect scattering characteristics, fusing the reconstructed signal matrix and all direct waves, performing full-focusing imaging, and identifying and positioning defects in the liquid die forging according to an imaging result. By adopting the scheme of the invention, the near-field effect of ultrasonic waves in nondestructive defect detection can be inhibited under the condition that the material of the liquid die forging is not uniform.

The application provides a stratum absorption compensation method and system based on a seismic signal space, and the compensation method comprises the following steps: acquiring a seismic signal; determining a coherent direction of the seismic signal by using a dip angle scanning method, and designing a spatial prediction error filter with signal identification capability; constructing a non-steady filter representing an absorption effect of the seismic signal; establishing a target functional for non-steady inversion of the seismic signal after absorption compensation, and introducing spatial coherence of the seismic signal into a regularization condition of an inversion system; and performing inversion on the target functional by using a conjugate gradient method to obtain the seismic record after absorption compensation. The method is simple to operate, stable to run, and obvious in effect, effectively suppresses the influence of noise interference on absorption compensation, and improves the compensation precision of the seismic signal.

The application discloses a kind of wind power blade clearance nonlinear dynamic prediction model construction methods, it is related to wind power blade monitoring technical field, including the following steps: deployment multi-frequency phased array laser radar matrix, the spatial wind speed data of wind farm front row unit wake region is collected, constructs three-dimensional instantaneous speed voxel library;Based on three-dimensional instantaneous speed voxel library, the profile phase convolution processing is executed to speed voxel under continuous time, extract wake rotation phase information, and map as standard vortex phase spectrum.The application introduces multi-frequency phased array laser radar, phase convolution and high-dimensional tensor analysis, accurately identifies the spatial coherence characteristics of wake vortex, combined with layered integration and multi-scale resonance detection, extracts wake phase superposition path, and constructs wake phase superposition exponential dynamic correction prediction model wind pressure boundary, significantly improves the prediction sensitivity and adaptability of wind power blade yawing risk, effectively solves the problem of underestimating and inaccurate prediction of wake coupling in the prior art.

The invention provides a fan three-dimensional flow field reconstruction method based on multi-point observation. The fan three-dimensional flow field reconstruction method comprises the following steps: S1, multi-point observation data acquisition: acquiring a wind speed time sequence of N actual measurement points; s2, trend and turbulence separation and frequency domain transformation: obtaining a spatially distributed average wind speed field and a disturbance time sequence, and obtaining actually measured frequency domain disturbance distribution; s3, generating a standard background turbulent flow field: based on a Kaimal turbulent flow spectrum model, combining a coherence function, and generating a standard turbulent flow field conforming to statistical characteristics and frequency domain distribution of the standard turbulent flow field through random phase superposition and inverse Fourier transform; s4, observation constraint superposition and spatial coherence reconstruction: establishing a joint vector matrix; constructing a complex covariance matrix based on the coherence function; carrying out Cholesky decomposition, and extracting an actual measurement-simulation structure coupling relationship; standardization processing of actual measurement constraints; constructing simulation disturbance of a belt actual measurement structure; performing inverse Fourier transform to generate a time domain turbulence field; and S5, superimposing a trend field and outputting a final flow field.

The invention belongs to the technical field of image processing, and particularly relates to a hyperspectral image dimension reduction and classification method and system based on discriminant spatial-spectral tensor locality preserving projection and a medium, and the method comprises the steps: inputting hyperspectral image data, and constructing three-dimensional tensor representation; performing blocking processing on the tensor, and extracting local tensor blocks; calculating a weighted covariance matrix based on a nuclear spectrum angle weighting strategy; adopting affine invariant Riemannian measurement to construct a weight matrix of the adjacency graph; solving the multi-modal projection matrix through an alternative optimization algorithm to realize dimension reduction; and finally, inputting the dimensionality-reduced features into a classification model and outputting a classification result. According to the method, by fusing the spatial-spectral tensor representation and the Riemannian geometry, the problems that a traditional method destroys spatial coherence and is sensitive to noise are solved, and the dimension reduction effect and the classification precision are remarkably improved.

The invention discloses a geosynchronous orbit SAR (Synthetic Aperture Radar) sea surface wave imaging quality evaluation method, and belongs to the technical field of geosynchronous orbit space correlation SAR sea surface wave imaging quality evaluation. The method comprises the following steps: acquiring a sea surface wave echo signal, generating an imaging result by using a backward projection algorithm after range-direction matched filtering and spatial correlation processing, calculating image intensity and a sea wave spectrum main energy direction slope, and finally quantifying imaging quality through a spatial correlation evaluation factor. The evaluation factor can effectively represent the spatial coherence of sea surface wave imaging, and the larger the value is, the better the imaging quality is. According to the method, the problem of insufficient applicability caused by low contrast and weak texture features in sea surface wave imaging of an existing evaluation method is solved, the accuracy of geosynchronous orbit SAR sea surface wave imaging quality evaluation is remarkably improved, and reliable technical support is provided for marine environment monitoring.

The application discloses a fire smoke environment life detection method based on 80GHz millimeter wave radar multi-receiving antenna joint voting. In the application, the false alarm caused by random noise and environmental clutter of a single antenna channel is effectively inhibited through voting confirmation after independent detection of multiple antennas. Since real vital signs can be consistently detected on multiple antennas, and interference sources are difficult to simultaneously meet the consistency condition of multiple channels, the reliability of the detection result is significantly improved. Meanwhile, the mechanism is not limited by the number of antennas and array configuration, has good universality and scalability, is convenient to deploy and implement on different hardware platforms, and the spatial coherence accumulation of beam forming enhances the signal-to-noise ratio of weak breathing echoes behind the shelter, so that the trapped personnel sheltered by sofas, tables and chairs and the like can still be effectively detected. Meanwhile, the fine resolution capability in the angle dimension improves the accuracy of target direction estimation, and provides more accurate positioning information for fire rescue.

The invention discloses a farmland protection forest three-dimensional structure parameter extraction method and system, and belongs to the technical field of remote sensing mapping and forestry information. The method comprises the following steps: acquiring oblique photography point cloud data of a target area; automatically identifying and extracting point cloud data corresponding to the farmland protection forest through a deep learning model; constructing a closed three-dimensional model representing the overall outer contour of the forest belt based on the point cloud; and automatically calculating the overall three-dimensional structure parameters of the forest belt based on the model, wherein the parameters comprise the length, the width, the height, the trend and the permeability representing the ventilation and light transmission performance of the forest belt. According to the method, the penetration coefficient model fusing the density difference, the spatial coherence and the feature communication length is innovatively provided, and scientific quantification of the hydrophobicity is achieved.

The application discloses a parameter-controllable random structure light generating device and method, a laser emits a light beam which sequentially passes through a linear polarizer and a beam expander and then is incident to a beam splitter, the light beam reflected by the beam splitter is a first light beam, the first light beam enters a digital micromirror device loaded with a holographic film to be modulated, a second light beam is obtained after the modulation, the second light beam is incident to the beam splitter again, and is incident to a common-path interference system after being transmitted; the second light beam passes through a first lens in the common-path interference system, positive and negative first-order light spots are filtered out through a pinhole diaphragm at a focal plane of the first lens, the positive and negative first-order light spots pass through two half-wave plates respectively, and then pass through a second lens and a Ronchi grating to be combined into a random vector light beam. The application can accurately generate a random structure light beam, and compared with a traditional structure light beam, can simultaneously control the polarization degree and the spatial coherence degree of the generated random structure light beam.

The invention relates to the technical field of medical image processing, provides a fundus focus image segmentation method and system, and solves the problems of insufficient spatial coherence and dissection rationality of a three-dimensional segmentation result in the prior art. The method comprises the following steps: acquiring optical coherence body data and structured light data of a retina, and registering and fusing into an enhanced point cloud; deep features of each point of the enhanced point cloud are extracted, and a global dependency relationship is determined; inputting a segmentation model to carry out three-dimensional instance segmentation to obtain an initial segmentation result; introducing a geometric constraint function corresponding to the prior model of the normal anatomical structure of the retina, adjusting the initial result to enable the lesion boundary to accord with the anatomical structure, and obtaining a target segmentation result; identifying focus key points in the enhanced point cloud according to a target segmentation result; and outputting the three-dimensional masks of the key points, and calculating the volume and the surface area of the focus and the spatial position of the focus relative to the retina layer according to the three-dimensional masks. According to the invention, high-precision three-dimensional segmentation and quantitative analysis of the fundus focus can be realized.

In an optical data transmission system, a beam modulator is configured to embed a data set into an input beam. The multimode optical waveguide network has an in-coupling region for receiving an input beam. The multimode optical waveguide network is configured to direct an input beam to an out-coupling region of the multimode optical waveguide network. The spatial coherence detector is configured to measure the phase and amplitude of the output light field at a plurality of positions. The output light field is at least partially defined by the input light beam, thus exhibiting a distortion effect caused by the light beam passing through the multimode waveguide network. Signal processing is applied to the output of the spatial coherence detector to compensate for distortion effects to restore the data set embedded in the input beam from the output of the spatial coherence detector.

In a method for hip measurement in an ultrasound image, an ultrasound image of the hip is obtained. Also obtained is a spatial coherence map associated with one or more lags of ultrasound waves associated with the ultrasound image; a quality metric is then determined based on the hip ultrasound image and the spatial coherence map, where the quality metric is indicative of the suitability of the ultrasound image for use in measuring a hip. If the quality metric indicates that the ultrasound image is above a threshold quality, then the method includes then indicating that the ultrasound image is suitable for hip measurement.

The invention discloses a method and a device for generating partially coherent light based on intra-cavity frequency doubling cooperative regulation and control, and relates to the technical field of laser. The device comprises a pumping source, a convex lens and a resonant cavity which are connected in sequence, wherein the resonant cavity comprises an input mirror, a laser gain medium, a frequency doubling crystal and an output mirror. The method comprises the following steps: determining the length of a resonant cavity and the position of each component according to the generation requirement of partially coherent light; a pumping source is used for emitting pumping laser, and collimation focusing is carried out through a convex lens; pumping laser enters the resonant cavity, the relation between the pumping light spot radius and the oscillation fundamental mode radius is adjusted in the resonant cavity, active control over matching of a pumping mode and an oscillation mode is carried out, and partially coherent light is output. According to the invention, high conversion efficiency and high output power generated by partially coherent light are ensured, and the problem of cooperative work of multiple parameters such as a frequency doubling process and an anisotropic space coherent structure in a resonant cavity is successfully solved.

The invention discloses a double-sided directional free-form surface light beam regulation and control device suitable for double light sources and a design method thereof, and belongs to the field of non-imaging optical and free-form surface optical design. The device comprises a light source module, a first free-form surface, a second free-form surface, a first target surface and a second target surface. The light source module comprises two different types of light sources, namely a laser light source and an LED light source, located on the two sides of the lens. The two light sources can work independently. And the first free-form surface and the second free-form surface form an integral double-sided free-form surface lens structure which is used for realizing directional light beam regulation and control on the two light sources. The shapes of the two free-form surfaces are independently designed through a bidirectional iteration geometrical optical algorithm, so that the system meets the energy conservation and refraction laws in two light source modes, and compatibility and direction selective regulation and control of different space coherence light sources are achieved. The dual-function integration of laser beam shaping and LED illumination can be realized in a single lens, and the laser beam shaping and LED illumination lens has the advantages of high luminous efficiency, compact structure and high adaptability.

The invention relates to the technical field of earthquake monitoring, in particular to a DAS-based earthquake and seismic facies combined pickup method and system, and the method comprises the steps: collecting an original seismic wave signal through a DAS system, and generating time-space continuous strain data; and carrying out spatial-dimension energy focusing enhancement processing on the data to generate a spatial focusing energy profile. A multi-window adaptive threshold analysis technology is utilized to identify a plurality of potential seismic phase arrival sections in the profile and extract first arrival time and space coordinate initial values of the potential seismic phase arrival sections; and constructing a multi-seismic associated matching network, and calculating to obtain a propagation trajectory and a time sequence of each seismic phase type on the optical fiber path by taking the initial values as input. And performing multi-seismic-phase separation reconstruction on the original signal based on a calculation result to generate separation wave field data of different seismic phase types. According to the method, the spatial coherence enhancement signal of the DAS data is fully utilized, and accurate joint reconstruction and separation of the multi-seismic-phase propagation trajectory are realized through the association network.