48 results about "Phase unwrapping" patented technology

A system and method for compressing synthetic aperture radar (SAR) images with enhanced phase recovery and unwrapping capabilities is disclosed. The system performs preprocessing on input SAR images, applies discrete cosine transform (DCT) to create subbands, and utilizes a multi-pass amplitude compression technique. A specialized neural network performs phase unwrapping using compressed amplitude information and interferogram wrapped phase data. The system employs a channel-wise transformer fusion block (CTFB) for feature fusion and a multi-stage context recovery subsystem with optimized loss functions for both amplitude and phase recovery. The method achieves improved compression efficiency and phase recovery accuracy, particularly beneficial for Interferometric SAR (InSAR) applications.

The invention belongs to the technical field of distributed optical fiber sensing (DAS) system signals, and discloses a time axis phase unwrapping method and system based on an FPGA (Field Programmable Gate Array), and the method comprises the steps of phase data acquisition, stable group formation, phase progressive unwrapping, assembly line and cache cooperative processing and quality measurement closed-loop correction. According to the system, real-time phase extraction of interference signals is achieved through the FPGA demodulation unit, phase change is dynamically unwrapped through the reliable period screening unit and the phase progressive unwrapping unit, a self-adaptive unwrapping mechanism is formed in combination with anomaly judgment and threshold updating logic, and through collaborative design of an annular cache and a three-level assembly line comparator, the interference signal is obtained. According to the method, parallel transmission and time sequence synchronization of data are achieved, the precision and stability of phase unwrapping can be effectively improved in a complex noise environment, hardware delay and resource occupation are remarkably reduced, and the method is suitable for real-time signal unwrapping processing of a high-precision distributed optical fiber sensing system.

The invention provides a high-precision phase unwrapping method fused with a Transform self-attention mechanism. The high-precision phase unwrapping method is suitable for digital speckle interferometry, optical imaging and medical image processing. According to the method, on the basis of a U-Net structure, position coding and multi-head self-attention (MHSA) modules are introduced into a bottleneck layer, so that global feature dependence is established, and a long-range context is captured; a multi-head cross attention (MHCA) module is added in a decoding stage, attention weighting and noise suppression are performed on jump connection features, and selective fusion of key features is realized. A composite loss function is adopted in training, error variance loss and total variation loss are included, and phase periodic characteristics and smooth constraints are considered. Through simulation data training containing noise and complex phase jump, the unwrapping precision and robustness are significantly improved.

The application discloses a phase unwrapping method and device, system and storage medium, comprising: constructing a U-NetHD absolute phase generation network for directly mapping the wrapped phase to the unwrapped phase; constructing a wrapped phase dataset; training the U-NetHD network by using an unwrapped dataset, unwrapping the interference phase through the absolute phase generation network to obtain an initial unwrapped phase; re-wrapping the initial unwrapped phase and performing conjugate multiplication with the original interference phase to obtain a residual phase; unwrapping the residual phase through an L1 norm unwrapping method to obtain a residual unwrapped phase; and finally, summing the initial unwrapped phase and the residual unwrapped phase to obtain the final unwrapped phase. The technical scheme of the application can solve the problem of low unwrapping precision in low coherence and large gradient regions, and effectively improve the phase unwrapping precision.

The present application relates to the technical field of optical interferometry, and particularly relates to a high-precision phase demodulation method based on a PhaseUnet++ network.The method comprises the following steps:1. Zernike polynomials are used to generate training data and test data for network training;2. a neural network PhaseUnet++ is established and trained;3. normalized cosine interferograms I cos and auxiliary diagrams I sac are input into the trained network, and normalized sinusoidal interferograms I sin are output;4. the wrapped phase distribution is obtained by using an arctangent function, and the unwrapped phase distribution φ(x,y) is obtained by using a phase unwrapping algorithm.The present application has high precision, can process interferograms of different sizes, has strong noise resistance, can realize phase measurement of dynamic samples, has high reconstruction result precision, simple steps, wide application range, and great application prospect in the field of optical interferometry research.

The invention discloses an InSAR (Interferometric Synthetic Aperture Radar) phase unwrapping method based on a terrain gradient sensing network, and relates to the field of synthetic aperture radar interferometry. The method comprises the following steps: acquiring a phase unwrapping data set and dividing the phase unwrapping data set into a training data set, a verification data set and a test data set; secondly, constructing a terrain gradient sensing network; inputting the training data set into the terrain gradient sensing network for iterative training, and storing an optimal model weight after training is completed; and finally, inputting the test data set into the trained terrain gradient sensing network, and outputting a predicted unwrapping phase diagram. According to the scheme, under the conditions of low signal-to-noise ratio and data missing, noise suppression and detail fidelity can be considered at the same time, residues and fringe artifacts are suppressed, phase continuity and physical consistency constraints are met, the unwrapping precision, stability and cross-scene generalization ability are improved, and a reliable phase unwrapping scheme is provided for synthetic aperture radar interferometry.

The invention provides a double-state interference morphology measurement method and system for a discontinuous precision surface, and belongs to the technical field of optical precision measurement. Based on a Fizeau interferometer, interference images in a focusing state and an out-of-focus state are respectively acquired, two-dimensional structure information and height information of a sample with a discontinuous precise surface are respectively extracted, and two-state data fusion is carried out, so that three-dimensional morphology reconstruction is realized. The method breaks through the limitation of the optical diffraction limit on the resolution, avoids the failure problem of a traditional phase unwrapping algorithm at a discontinuous boundary, remarkably enhances the environmental interference resistance, improves the calculation efficiency, is suitable for high-precision surface measurement of a vacuum chuck and the like with a large number of discontinuous characteristics, and has a wide application prospect. And a reliable solution is provided for high-end manufacturing quality control.

This invention discloses a double-sideband swept-frequency laser ranging and velocimetry device and method based on phase unwrapping. The device includes a single-frequency laser, a double-sideband modulator, an amplifier, a first coupler, a first demultiplexer, a second coupler, a second demultiplexer, a third coupler, a circulator, an optical antenna, a first photodetector, a second photodetector, a synchronous acquisition card, and a phase unwrapping and range / velocimetry demodulation module. This invention can obtain the distance values ​​corresponding to all sampling times within one sweep cycle, achieving an order-of-magnitude increase in system measurement rate; simultaneously, it can adapt to dynamic targets with arbitrary motion forms, without requiring a constant target velocity.

The application discloses a kind of based on FPGA's distributed optical fiber sensing signal two-dimensional phase unwrapping method, device and medium, the method steps include: step 1, phase is calculated using Cordic by FPGA and is scaled to obtain input phase;Step 2, the first phase difference is calculated in spatial dimension, and the first phase compensation amount is calculated by comprehensively judging the result of the first phase difference size of each clock edge;Step 3, the input phase is compensated using the first phase compensation amount, and spatial dimension phase unwrapping is completed;Step 4, spatial dimension phase unwrapping data is written into cache, and data is taken out at intervals and is arranged according to time dimension;Step 5, the second phase difference is calculated at each time in time dimension, and the second phase compensation amount is calculated and the positive and negative state is judged, and the input phase is compensated to obtain time dimension unwrapping result.The application can realize high-speed pipeline calculation on FPGA, and real-time two-dimensional phase unwrapping is carried out and high-precision unwrapping result is obtained.

The invention relates to the technical field of electronic information, in particular to a recursive three-dimensional phase unwrapping method, which comprises the following steps of: initializing, performing radiometric calibration and registration on a multi-temporal SAR (Synthetic Aperture Radar) image, calculating NAD and screening coherent scatterers, executing static three-dimensional unwrapping and filing parameters; recursive unwrapping: adaptively determining an NAD threshold value, distinguishing updated / newly added candidates, performing time smooth recursion and space iteration on an updated set, eliminating anomalies by space-time consistency, and performing static unwrapping and grid connection on a newly added set; initialization updating is carried out, Kalman recursion deformation parameters and noise variance are adopted, and switching and unified correction are carried out when a reference point is unstable. According to the method, the unwrapping error under the non-stationary deformation is reduced, the scatterer density is maintained, the increment processing efficiency is remarkably improved, and near real-time monitoring is supported.

The application discloses a kind of folding phase auxiliary binary coding fast phase unwrapping method, comprising: obtaining the folding phase in fringe pattern by least square method;According to the positive and negative of the folding phase, the binary image is obtained;By the binary image, the reverse binary image is obtained;Based on connected domain coding function, the connected domain coding of the binary image and the connected domain coding of the reverse binary image are constructed;The fringe order is obtained by the connected domain coding;The unwrapping phase is obtained by combining the fringe order with the folding phase.The application realizes fast phase unwrapping by folding phase auxiliary, and has strong anti-noise ability, including the influence of phase abnormal point, robustness of Gaussian noise, defocus etc..

The invention provides a phase unwrapping method based on a channel-space-frequency three-branch U-Net. The method comprises the following steps: acquiring a phase unwrapping data set and dividing the phase unwrapping data set into a training data set, a verification data set and a test data set; secondly, constructing a channel-space-frequency three-branch U-Net network; inputting the training data set into the channel-space-frequency three-branch U-Net network for iterative training, and storing an optimal model weight after training is completed; and finally, inputting a test data set into the trained channel-space-frequency three-branch U-Net network, and outputting a predicted unwrapping phase diagram. According to the method, the unification of noise suppression and detail reservation can be realized, the phase unwrapping precision and robustness in a low signal-to-noise ratio scene are effectively improved, the unwrapping result of a data missing area is enabled to better conform to a physical law, and a reliable phase unwrapping scheme is provided for ground surface deformation monitoring of synthetic aperture radar interference measurement.

The application discloses a least square unwrapping method based on residual point correction for unconnected regions, and the method comprises the following steps: collecting four original interference patterns of a measured sample containing unconnected regions; extracting an original wrapped phase pattern by a four-step phase shifting method; adding a digital carrier frequency to obtain a reference phase pattern; repairing the unconnected regions in the original wrapped phase pattern by using a sample block matching method, comparing with the reference phase, and repeatedly iterating to obtain a final wrapped phase pattern; obtaining an unwrapped phase pattern by using a least square unwrapping algorithm based on residual point correction; and obtaining a correction coefficient and finally obtaining a reconstructed phase pattern. The application combines the advantages of sample block matching and residual point correction, has a good unwrapping effect on the wrapped phase of unconnected, high-steep and high-noise regions, has high unwrapped phase precision, and has important significance for target pellet detection, optical element surface shape detection and other fields.

The application discloses a fast spatial phase unwrapping method, which comprises the following steps: calculating the truncated phase of each pixel point according to a plurality of acquired phase shift images; generating image A and image B according to the truncated phase value; respectively performing connected domain analysis on the image A and the image B to obtain a plurality of stripe connected domains; numbering the stripe connected domains in the image A and the image B respectively, and obtaining the stripe order code according to the numbering sequence; or, generating an auxiliary image by phase shift, and then obtaining the adjacency relationship between the stripe connected domains in the image A and the image B according to the phase overlapping relationship between the auxiliary image and the image A and the image B, obtaining the stripe order according to the adjacency relationship, and finally calculating the continuous phase by using the stripe order code and the truncated phase to complete the spatial phase unwrapping; the method can shorten the processing time of the existing spatial phase unwrapping method to one-tenth of the original time, and is beneficial to realizing the fast structured light three-dimensional measurement.

The application discloses a detail loss optimization method and system for phase unwrapping in deep learning, and a storage medium. The method comprises the following steps: constructing a generative adversarial network for phase unwrapping, including a generator and a discriminator; the generator is used for unwrapping wrapped phase data and generating an image; the discriminator is used for discriminating the true and false results of the image generated by the generator; a hybrid loss function of the generative adversarial network is determined, and the generative adversarial network is trained to process phase unwrapping data. By setting the generative adversarial network with the hybrid loss function, the trained GAN retains the texture details of the phase unwrapping results as much as possible, effectively suppresses the speckle noise, and significantly reduces the speckle noise in the valley.

The application discloses a three-dimensional measurement method based on nonlinear cosine coding, and comprises the following steps: step one, generating N-step phase shift coding, constructing a wavelength-De Bruijn element mapping table, generating nonlinear cosine coding and nonlinear cosine component coding; step two, projecting phase shift stripes and nonlinear cosine coding stripes to a measured object frame by frame, acquiring a modulation pattern and decoding to obtain wrapped phase and nonlinear cosine components; step three, constructing an approximate phase, performing phase unwrapping on the wrapped phase to obtain a local continuous phase, and dividing the local continuous phase into local phases; step four, minimizing cosine component errors of the local phase regions to retrieve phase orders, thereby realizing phase unwrapping and obtaining error absolute phase; and step five, performing phase repair on the error absolute phase to obtain error-free absolute phase, and combining with structural light system parameters to reconstruct a three-dimensional model of an object. The application can realize high-precision phase unwrapping by using only a single frame pattern, and greatly improves the three-dimensional measurement efficiency.

The application discloses a chemical shift encoding imaging method, device and equipment based on phase unwrapping, which comprises the following steps: acquiring an initial image, determining a field map candidate solution of the initial image; performing phase conversion on the candidate field map solution, taking the correct solution and the interval between the inverse solutions of the candidate field map solution as the target within a set range, determining an intermediate field map solution based on a phase unwrapping method; determining the real phase of the intermediate field map solution, and converting the real phase to the field map candidate solution space to determine a target field map solution; determining a first chemical composition signal and a second chemical composition signal based on the target field map solution, and performing chemical shift encoding imaging based on the first chemical composition signal and / or the second chemical composition signal. By converting the problem of selecting one of the phase vectors in different chemical compositions into a phase unwrapping problem, the stability and accuracy of chemical shift component separation are ensured under the condition that the acquisition parameters do not meet specific conditions, and the chemical shift encoding imaging effect is improved.

The invention discloses a phase unwrapping algorithm for optical coherence elastography, and relates to the technical field of phase unwrapping. The algorithm is based on an improved U-Net architecture, a network comprising an encoder, a decoder and a jump connection mechanism is constructed, the core is cooperated with an EGA module through a multi-branch Inception residual module in the encoder, the multi-branch Inception residual module extracts different receptive field phase features through five parallel convolution branches, the EGA module calculates gradients through a Sobel operator, attention weights are generated, and the attention weights are calculated through the EGA module. Strengthening phase edge and gradient mutation region response; the decoder removes residual noise through transposition convolution up-sampling, fusion of high and low layer features for fine reconstruction, and optional smoothing filtering. The method solves the problems that a traditional algorithm is low in precision, poor in robustness and low in speed under the scenes of low signal-to-noise ratio, strong speckle disturbance and severe phase gradient change, achieves high precision, high robustness and rapid unwrapping of OCE phase signals, adapts to complex tissue deformation and large-view-field rapid reconstruction requirements, and can be widely applied to the fields of clinical diagnosis, biomedical research and the like.

The invention aims to provide a phase unwrapping method based on extended Kalman filtering. The method comprises the following steps: A, constructing a two-dimensional phase unwrapping program MCC-EKFPU; and B, performing phase unwrapping processing on the interferogram by using a two-dimensional phase unwrapping program MCC-EKFPU to obtain an unwrapped phase of the interferogram. According to the method, the maximum correlation entropy criterion extended Kalman filtering theory is introduced into interferogram wrapping phase unwrapping, and an interferogram phase gradient estimation technology and a robust path tracking strategy are combined, so that the method has higher phase unwrapping precision and better robustness; the method has good application potential in interferogram phase unwrapping application with high accuracy requirements.

A phase order correction method for multi-frequency cascaded phase unwrapping is proposed. First, the low-frequency unfolded phase is directly calculated using the multi-frequency fringe method, and then the mid-frequency unfolded phase is obtained to acquire the initial fringe order obtained through a small frequency ratio. Subsequently, based on the shifting wrapped phase method, a fringe order misaligned by half a cycle from the initial fringe order is obtained. Accurate phase reconstruction is achieved through fringe order correction. These two methods complement each other to simultaneously address the accuracy degradation caused by pseudo-2-step transitions and impulse noise errors, while preserving phase map details. This invention obtains the correct order corresponding to the high-frequency wrapped phase through a phase recovery method at two fringe frequencies, and uses this to calculate the high-frequency unfolded phase, achieving fast and accurate phase unwrapping.

The embodiment of the invention relates to the technical field of radio, and discloses an array element calibration method, a phase unwrapping method, a storage medium and an integrated circuit. The method comprises the following steps: acquiring N target echo signals of an array element by rotating a rotary table on which an antenna array is fixed; n target echo phases corresponding to the target array element under the N target angles are obtained, the target echo phases are phases aligned with a preset reference array element, and the target angle is obtained by using the linear relation between the target echo phases and the target angles. And predicting the target echo phase of the target array element at the nth target angle according to the setting position of the target array element under the condition that the difference between the actually detected target echo phase x 'k, n and the actually detected target echo phase x' k, n is greater than a preset threshold, performing 2pi compensation on x 'k, n, and obtaining the unwound target echo phase. Therefore, the calibration of the array element does not depend on manually writing a logic judgment expression to realize the unwrapping of the phase, thereby greatly reducing the manpower and time cost, and facilitating the batch correction operation.

The application provides a point cloud reconstruction method based on partition phase unwrapping overcoming misalignment restriction, and relates to the technical field of structured light three-dimensional measurement, and comprises the following steps: synchronously collecting a left image set and a right image set by using a binocular camera, respectively solving a stripe order, a reference wrapped phase and a mask image; determining a first type of crossing point and a second type of crossing point according to the reference wrapped phase; constructing three misaligned wrapped phases with the reference wrapped phase as an intermediate phase, and classifying and calculating the absolute phase in the stripe order according to the number of the first type of crossing point and the second type of crossing point corresponding to each stripe order value; correcting misaligned points for each row of absolute phase by using RANSAC linear fitting; sliding a preset matching window on the left phase image and the right phase image pixel by pixel, combining a full-field illumination image, obtaining a disparity map of the left phase image and the right phase image, and further reconstructing a three-dimensional point cloud. The application can significantly reduce the problems of flying points and phase jumps, and effectively improve the point cloud reconstruction quality.

The application provides a phase unwrapping method and device applied to ship-bridge collision bridge body vibration monitoring, and relates to the technical field of distributed optical fiber acoustic wave sensing. The method comprises the following steps: based on a self-attention mechanism, self-adaptive noise suppression is performed on a window phase signal sequence to obtain a denoised phase signal sequence; according to the denoised phase signal sequence, multi-scale time-frequency feature fusion is performed through a convolutional neural network to obtain a space-time matrix and a time-frequency feature; according to the space-time matrix, spatial correlation feature extraction is performed through a two-dimensional convolutional network to obtain a spatial feature; based on a long short-term memory network, weighted fusion is performed on the time-frequency feature and the spatial feature to obtain a space-time fusion feature; and according to the space-time fusion feature, unwrapping phase mapping is performed through a full connection layer to obtain a real phase signal sequence. The application is a phase unwrapping method for distributed optical fiber acoustic wave vibration in a ship collision scene, which has high accuracy and good robustness.

The application provides a multi-baseline phase unwrapping method based on accurate intercept correction, and relates to the technical field of radar signal processing, and comprises the following steps: acquiring parameter information corresponding to main and auxiliary images of an interferometric synthetic aperture radar in an observation scene and original interferometric phases, and constructing a reference intercept vector based on the parameter information and an external digital elevation model database; traversing actual intercepts corresponding to all pixels in the observation scene through a sliding window, obtaining a core intercept based on the frequency of the actual intercepts, correcting the core intercept by using the reference intercept vector, and obtaining a corrected intercept set; calculating a fuzzy number by using a limited range search method or a Chinese remainder theorem method based on the corrected intercept set and the number of baselines to obtain a fuzzy value; and calculating an unwrapped phase corresponding to each baseline by using a phase unwrapping equation based on the fuzzy value and the original interferometric phases. In this way, the practicability and reliability of the multi-baseline phase unwrapping method are significantly improved.

The embodiment of the invention relates to the technical field of phase unwrapping, in particular to an aperture radar interferometry multi-baseline phase unwrapping method and device. A specific embodiment of the method comprises the steps of performing data preprocessing on an obtained initial tile data set to generate a sample phase data set; based on the sample phase data set, performing model training processing on the initial phase unwrapping model to generate a phase unwrapping model; performing data preprocessing on the acquired to-be-detected picture to generate winding phase data; and inputting the wrapped phase data into the phase unwrapping model to obtain unwrapped phase data. According to the embodiment, the wrapped phase can be converted into an accurate real phase, and high-precision multi-baseline phase unwrapping is realized.

The invention relates to a fast phase unwrapping method combining DeBruijn and a half-cycle Gray code. The fast phase unwrapping method comprises the following steps: step 1, generating four phase shift patterns, and a sum; step 2, generating a DeBruijn pattern with a period of T and a DeBruijn pattern with a period of T / 2, and generating a Gray code pattern with a period of T / 2 at the same time; step 3, determining a projection pattern as,..., and projecting the projection pattern to a plane; step 4, calculating a normal fourth-step phase shift; step 5, obtaining a wrapped phase; step 6, searching a corresponding position in the original code element list to respectively obtain a phase order sum; and step 7, defining boundary conditions of phase intervals, introducing a tiny correction value, and selecting and calculating absolute phases of all the intervals through a formula according to a numerical range wrapping the phases. The method has the beneficial effects that the problem of edge jump is solved, and the unwrapping time is shortened.

This invention discloses a phase unwrapping method and system for large-gradient deformation regions in mining areas, achieving high-precision unwrapping results in areas where unwrapping is difficult. Specifically, it includes: converting an image (interferometric phase map) to the frequency domain based on a two-dimensional fast Fourier transform (FFT2); obtaining a pre-unwrapped image by solving the Poisson equation and performing an inverse two-dimensional Fourier transform (iFFT2) on the second-order difference result of the image; diluting the interference fringes by performing a complex conjugate multiplication of the original image and the pre-unwrapped result to obtain a noise phase map; and performing an L-type phase unwrapping on the noise phase map. 1 Norm-based phase unwrapping yields the remaining effective phase information within the noisy phase. Finally, the pre-unwrapped result is multiplied conjugately with the remaining phase to obtain the final unwrapped result. This invention solves the problem of low unwrapping accuracy in regions with large gradient changes, effectively improving unwrapping accuracy.

The application discloses an InSAR phase unwrapping method based on a Fourier domain global hybrid network, and relates to the technical field of remote sensing, and the method comprises the following steps: generating synthetic data of diversified continuous phases through data simulation, superimposing compound noise to obtain a data pair of noisy wrapped phases and real phases; a phase unwrapping model is constructed, a UNet encoder-decoder structure is used as a backbone network, a Fourier mixed residual block FMRB is embedded in a bottleneck layer, an all-space pyramid pooling ASPP module is connected, and a convolution block attention module CBAM is integrated; the phase unwrapping model is trained, a compound loss function is used as an optimization objective, phase unwrapping model parameters are iteratively optimized until the phase unwrapping model converges; and an InSAR interferogram to be unwrapped is input into the phase unwrapping model, and unwrapped phase results are output. The application comprehensively constrains the unwrapping process from multiple physical dimensions through the compound loss function, and the robustness in a complex noise environment and the physical reasonableness of a solution are enhanced.

The application discloses a kind of based on reference plane speckle auxiliary fast phase unwrapping method and application, including the following steps: acquisition reference plane speckle diagram, and reference plane absolute phase is obtained by space phase unwrapping method;Three frames of phase shift fringe patterns and object speckle diagram of imaging object are collected, and object wrapped phase is solved according to three frames of phase shift fringe patterns of imaging object;Disparity map is generated based on the speckle texture features matching in object speckle diagram and reference plane speckle diagram;Phase unwrapping is carried out to object wrapped phase using reference plane absolute phase, and composite phase containing layered feature is generated;Using disparity map, the phase order of composite phase is recalibrated layer by layer, and the accurate absolute phase of imaging object is obtained.The application combines the advantages of traditional phase shift code and speckle code, and the calculation of absolute phase can be realized without additional auxiliary image, and the number of projection is significantly less than that of traditional method, which greatly improves the efficiency of three-dimensional imaging measurement.

The invention discloses an InSAR (Interferometric Synthetic Aperture Radar) phase unwrapping method based on a Fourier domain global hybrid network, which relates to the technical field of remote sensing and comprises the following steps of: generating synthetic data of diversified continuous phases through data simulation, and superposing composite noise to obtain a data pair of a noisy wrapped phase and a real phase; the method comprises the following steps: constructing a phase unwrapping model, taking a UNet encoder-decoder structure as a backbone network, embedding a Fourier mixed residual block FMRB in a bottleneck layer, connecting a cavity space pyramid pooling ASPP module, and integrating a convolution block attention module CBAM; training the phase unwrapping model, and iteratively optimizing parameters of the phase unwrapping model by taking the composite loss function as an optimization target until the phase unwrapping model converges; and inputting the InSAR interferogram to be unwrapped into the phase unwrapping model, and outputting an unwrapped phase result. According to the method, the unwrapping process is comprehensively constrained from a plurality of physical dimensions through the composite loss function, and the robustness and the physical reasonability of the solution in a complex noise environment are enhanced.