841 results about "Phase image" patented technology

The invention relates to the technical field of image data measurement and analysis, in particular to a method and a system for measuring slope deformation in a dangerous mountainous area based on image data. The method comprises the following steps: acquiring high-resolution image acquisition data and GNSS auxiliary data of the slope of the hard mountain area; correcting the high-resolution image acquisition data to obtain corrected slope image acquisition data; performing local feature extraction and matching of each time phase image on the corrected slope image acquisition data to obtain slope preliminary matching point set data; and performing mismatching elimination on the slope preliminary matching point set data to obtain transformation matrix data between the slope images. According to the method, high-resolution image acquisition and GNSS data are combined, through correction, registration, three-dimensional reconstruction and optical flow analysis, slope deformation of the hard mountainous area is accurately obtained and analyzed, and efficient deformation monitoring and visualization results are achieved.

The invention belongs to the technical field of optical precision measurement, and particularly relates to a wafer defect detection system based on dual-channel optical imaging. According to the invention, dark field scattering and diffraction phase microscopy are combined as two optical imaging channels; optical paths of the two imaging modes are compatible and synchronized; through optical system calibration and optimization, high-precision acquisition and alignment of multi-modal images are realized; calibrating the sensitivity and resolution of the system by using a standard sample to ensure that the detection performance meets a preset requirement; carrying out dual-channel synchronous imaging on the wafer sample to be detected, and respectively obtaining a dark field scattering image and a diffraction phase image; carrying out conjoint analysis on the multi-modal data through an image algorithm, extracting size, depth and structure information of defects, and generating a high-precision defect detection report; the system specifically comprises a dark field scattering imaging module, a diffraction phase microscopic imaging module and an image processing unit. According to the invention, the defect of a single imaging mode can be effectively overcome, and the detection precision and robustness of the system are remarkably improved.

The invention discloses a remote sensing image change detection method based on spatio-temporal feature interaction and feature difference enhancement, and belongs to the technical field of remote sensing image processing, and the detection method comprises the following steps: constructing a change detection data set containing a dual-temporal remote sensing image, respectively extracting multi-scale features through an encoder, and obtaining a change detection data set; inputting the data to a spatio-temporal feature interaction module to obtain interacted dual-time-phase features; fusing the interacted double-time-phase features and inputting the fused double-time-phase features into a decoder to generate four groups of same-resolution features with different semantic hierarchies; performing difference enhancement on the four groups of features through a feature differentiator, then performing channel splicing and fusion, and outputting a change detection result; and training the model by using the training set, adjusting and optimizing, and evaluating the precision. The beneficial effects of the invention are that the method can effectively capture the space-time dependency relationship between the double-time-phase images, enhances the discrimination capability of a change region, and meets the requirements of high-precision change detection in a complex scene.

The invention provides a mining area deformation monitoring and risk assessment method, which comprises the following steps of: carrying out registration, de-leveling, filtering and phase unwrapping on acquired image data through InSAR (Interferometric Synthetic Aperture Radar), generating a coherence graph, a filtering interference graph and an unwrapping phase graph, carrying out phase decomposition by using a singular value algorithm, separating a deformation phase, an atmospheric delay phase and noise, and carrying out risk assessment. The method comprises the steps of expanding discrete deformation points into a continuous deformation field through Kriging interpolation, carrying out deformation grading by taking a deformation rate, an accumulated settlement amount and a deformation gradient as core indexes, identifying a deformation high-risk area, arranging measurement robot monitoring points in the high-risk area, and constructing a risk assessment model in combination with real-time deformation data of a measurement robot and geological parameters. And evaluating the risk grade of each part of the mining area through the dynamic risk index model, and triggering graded early warning. According to the method, the InSAR and measurement robot technology is combined to solve the data fusion problem of large-range deformation monitoring and local high-precision monitoring, and the dynamic risk assessment precision under the complex terrain of the mining area is improved.

The invention discloses an interferometric phase unwrapping method based on a double-branch decoding residual network, and the method comprises the following steps: S1, constructing an interferometric phase data set which comprises an interferometric phase diagram of pixel phase wrapping and a corresponding real interferometric phase diagram; s2, constructing a double-branch decoding residual network, wherein the double-branch decoding residual network comprises an input layer, an encoder network, a double-branch decoder network comprising a first branch decoder module and a second branch decoder module, and an output layer which are connected in sequence; training the double-branch decoding residual network by using the interferometric phase data set to obtain a double-branch decoding residual network interferometric phase unwrapping model and the weight of the model; and S3, inputting the interferometric phase diagram subjected to pixel phase winding into a double-branch decoding residual network interferometric phase unwrapping model to obtain a predicted interferometric unwrapping phase diagram. According to the method, the interferometric phase unwrapping under the condition of low signal-to-noise ratio is realized, and the interferometric measurement precision of the synthetic aperture radar is improved.

The invention discloses a remote sensing change detection method based on edge enhanced cross attention and multi-dimensional loss. The method comprises the following steps: inputting a dual-time-phase image into a trained twin backbone network, and extracting multi-scale backbone features of the dual-time-phase image; inputting the multi-scale backbone features into a trained non-sharpening mask cross attention fusion model to extract spatial information and semantic information of the dual-temporal image through cross attention operation, and extracting edge information of the dual-temporal image through a sharpening convolution kernel to obtain multi-level differential features with different scales; inputting the differential features into a trained multi-scale hierarchical connection decoder to obtain a multi-scale preliminary prediction result of the change area; and carrying out weighted fusion on the multi-scale preliminary prediction result to obtain a change prediction result. According to the invention, the multi-scale information of the twin backbone network can be fully utilized, and the capability of extracting the edge information of the dual-time-phase image can be improved, so that the change detection accuracy can be improved.

The invention provides a 2.5 D phase deflection technology-based complex vehicle body surface coating defect detection method. The method comprises the following steps of: 1, projecting four cosine phase diagrams along an X direction and four cosine phase diagrams along a Y direction to the surface of a vehicle body coating through a programmable light source, and capturing eight cosine phase diagrams reflected by the surface of the vehicle body coating by using a high-precision industrial camera; 2, carrying out fusion processing on eight cosine phase diagrams captured by a camera by adopting a phase deflection technology; and step 3, gradient information of the wrapped phase diagram is extracted and fused, and a shape diagram is generated. The image generated by the technology can detect automobile body coating defects which are difficult to identify by a traditional method, can more clearly reflect fine defects such as sagging, scratches, pits and protrusions on the surface of the automobile coating, and also shows excellent detection capability on the complicated surface of the automobile body coating. According to the technology, the defect detection capability is remarkably improved, the subsequent defect extraction process is simplified, and the detection efficiency and accuracy are greatly improved.

The invention relates to the technical field of laser welding of micro-fluidic chips, in particular to a method and device for laser welding of a micro-fluidic chip. The method for laser welding of the micro-fluidic chip comprises the following steps: positioning the micro-fluidic chip, acquiring a local picture of the micro-fluidic chip through a visual detection module, controlling two end parts to automatically adjust positions on a space coordinate axis, and ensuring that a cover plate is tightly attached to the micro-fluidic chip; and the target heating light spot is aligned with the molten pool part, so that the matching degree of a real-time crystal phase image formed on the molten pool part and a target crystal phase image is larger than a set value, the welding procedure is completed in an automatic control and detection mode, and finally annealing is conducted to complete welding. The welding method is high in automation degree, laser welding is conducted through the laser welding head module, heating is fast, efficiency is high, and accurate and automatic welding is achieved by judging a crystal phase image.

The invention discloses an intelligent change detection method for non-aligned two-time-phase images of an inspection unmanned aerial vehicle, and relates to the technical field of unmanned aerial vehicle image processing and change detection, and the method comprises the steps: obtaining two-phase images collected by a low-altitude unmanned aerial vehicle under a fixed route and same sensor parameters; a lightweight registration model is constructed and trained, feature point matching is utilized to predict matching point pairs, a homography matrix is calculated, and accurate registration of non-aligned images is achieved; and constructing and training a change detection model based on image pair interaction feature fusion, analyzing the aligned image after registration, and outputting a change information binary image. The method can effectively solve the problem of non-alignment caused by position and angle differences during two-time-phase image acquisition of an unmanned aerial vehicle, and the technical problems of low precision, poor robustness and insufficient calculation efficiency of a traditional method in change detection, effectively improves the automation level of low-altitude safety monitoring of ground highways and railways, reduces the maintenance cost, and improves the safety of the unmanned aerial vehicle. The important application value is realized.

The invention relates to a remote sensing image change detection method and equipment based on multi-scale difference feature fusion, and belongs to the technical field of remote sensing image processing. According to the invention, a multi-attention feature enhancement module is designed to carry out adaptive weight adjustment and enhancement on a dual-time-phase image and differential information thereof, meanwhile, a cross-scale residual fusion module is adopted to carry out unified fusion on features from different scales, and fused global change features are input to a classification head to carry out change detection. According to the method, the details and the overall structure of the ground object change area are accurately captured, so that the accuracy and the robustness of change detection are effectively improved, and better detection performance can be obtained under the condition that the number of training samples is small.

The invention relates to an optical phased array phase calibration method and device, an electronic device and a storage medium, and is applied to the field of optical detection.The optical phased array phase calibration method comprises the steps that a to-be-calibrated phase image is input into an optical phased array system, and a random disturbance phase image corresponding to the to-be-calibrated phase image is obtained; the random disturbance phase image is obtained by performing phase disturbance within a preset range on the to-be-calibrated phase image. Inputting the to-be-calibrated phase image and the random disturbance phase image into a target convolutional neural network model, and determining a target phase error of the to-be-calibrated phase; the target convolutional neural network model is obtained by training a preset convolutional neural network model based on the far-field training image pair; and calibrating the initial emission electric field of the to-be-calibrated phase image based on the target phase error. According to the invention, the problem that the accuracy of phase error calibration of an existing optical phased array needs to be improved is solved.

The invention discloses a display screen appearance detection method, device and system and a storage medium, which are used for improving the detection precision of display screen appearance defects. The method comprises the following steps: controlling a light source to emit stripe light of different phases to a display screen to be detected through a time-sharing stroboscopic method, and scanning the display screen to be detected by using a line scanning camera and adopting a multi-frame acquisition mode to generate slice image data; performing dephasing processing on the slice image data to generate wrapped phase image data; performing phase unwrapping processing on the wrapped phase image data to generate absolute phase image data; performing image fusion processing on the absolute phase image data and the slice image data according to a to-be-detected defect type to generate a to-be-detected image set; performing defect detection on the to-be-detected image set to generate a plurality of slice data; and integrating the plurality of slice data to generate target defect data.

The invention discloses a remote sensing image change detection method based on a lightweight staggered structure and coarse and fine granularity fusion. The method comprises the following steps: 1, preprocessing an image in a remote sensing image change detection data set; 2, constructing a network model M2M-LINet, and initializing a hyper-parameter; the M2M-LINet comprises a coarse-grained positioning stage, a detail focusing stage and a decoding prediction stage, wherein in the coarse-grained positioning stage, coarse-grained feature mapping of an input double-time-phase image is obtained through a CALM module, and change information is preliminarily positioned; the detail focusing stage comprises an LCB module and an LTB module which are combined in a staggered manner; in the decoding prediction stage, edge features of the image are enhanced through an edge perception enhancement module EAEM, and decoding prediction is carried out through an LTB module and a convolution prediction head; 3, the M2M-LINet is trained and verified, parameters of the M2M-LINet are adjusted and optimized, and a network model M2M-LINet with the optimal parameters is obtained; and 4, predicting a detection result by using the M2M-LINet with the optimal parameter. According to the method, the change detection precision of the remote sensing image is improved with low calculation cost, and small target and boundary change can be accurately detected.

The invention discloses a method and a device for correcting laser focusing aberration in a transparent material, and belongs to the technical field of laser processing. The method comprises the steps that a machining light path containing an object plane, an aspheric reflector and a 4F device is built in Zemax, the surface type of the aspheric reflector is optimized with the minimum focus aberration as the target, and a rise table is derived; an optical path difference and a laser phase are calculated through Matlab, and a phase diagram is generated; and an actual machining light path is built, and a phase diagram is loaded to achieve low-aberration machining. Simulation optimization and phase modulation are combined, aberration caused by refractive index difference is effectively counteracted, wavefront errors are reduced by 58%, the thickness of a machining damage layer is reduced by 70%, the method is suitable for various transparent materials, the machining precision and the material utilization rate are improved, and the method is suitable for high-precision laser machining scenes.

The invention discloses a cultured seaweed identification method and system based on time sequence remote sensing and morphological constraint, and the method comprises the steps: obtaining and preprocessing an original multispectral remote sensing image, forming a multispectral remote sensing image set, and generating a water body range mask; under the constraint of a water body range mask, synthesizing each time phase image, enhancing the texture through local maximum filtering, calculating a seaweed light absorption index, and generating a feature enhanced image; performing threshold segmentation on the feature enhanced image, extracting and vectorizing a potential seaweed pattern spot raster image, and calculating morphological indexes such as area and firmness of each vector pattern spot; and screening the vector pattern spots based on a preset area and a firmness threshold, and outputting a cultured seaweed distribution result. According to the method, the self-defined index and the morphological constraint are combined, the recognition precision of regular form targets such as culture rafts is effectively improved, area time sequence change analysis can be generated, and an automatic and high-precision technical scheme is provided for dynamic monitoring of culture seaweed resources.

The invention relates to the technical field of remote sensing image processing, and particularly discloses a photovoltaic power station remote sensing identification method and system based on time constraint, and the method comprises the steps: reading image data, and obtaining projection parameters and geographical transformation parameters; processing the image data by a minimum-maximum normalization method; selecting a reference year, acquiring reference year photovoltaic power station vector data processed by a normalization method, converting image data and vector data processed by the reference year normalization method into the same coordinate system according to projection parameters and / or geographical transformation parameters, rasterizing the vector data, and generating a binary mask consistent with the resolution of the image data; and by taking the binary mask as a label, designing a deep learning network structure to train image data, and outputting a predicted binary mask. According to the method, the uncertainty of single-phase image classification is remarkably reduced, and the stability and reliability of a multi-phase classification result are ensured.

The invention relates to a method (300) for detecting damage (201) to a part (200) made of conductive material, the damage (201) being a burn or a crack, the part (200) being heated (310) by induction by injecting an energy pulse. The method comprises receiving (330) a temporal sequence of thermal images of the part after heating the part (200), and the temporal signal of the intensity of each pixel of the thermal image is used to construct (370) a phase image of the part by means of a discrete Fourier transformation of the temporal signal.

The invention provides a multi-scale cultivated land change detection system based on frequency domain conversion, and belongs to the technical field of remote sensing image processing and computer vision. Firstly, feature extraction is carried out on a dual-time-phase image to obtain a multi-scale first feature map; performing multi-kernel parallel convolution processing on the first feature map through small convolution kernels with different sizes, and then splicing and fusing the first feature map into a second feature map; performing feature enhancement on the second feature map to generate an extraction feature map; and combining and classifying two paths of extracted feature maps corresponding to the double-time-phase image to obtain a detection map of cultivated land change. Meanwhile, the double-time-phase image for training is subjected to low-frequency feature replacement in the frequency domain, visual feature differences caused by different imaging conditions are effectively eliminated, and false change false detection is reduced. According to the method, false change interference is reduced through low-frequency component replacement, meanwhile, background noise caused by a large convolution kernel and information loss caused by cavity convolution can be avoided through multi-kernel parallel convolution processing of the small convolution kernel, and meanwhile the detection capacity of scale sensitive targets is improved.

The invention discloses a chromosome metaphase division phase image sorting method, which belongs to the field of image processing, and comprises the following steps of: segmenting crossed and adhered chromosomes in chromosome metaphase division phase images; extracting a plurality of features from the segmented chromosome image; setting a reasonable interval of each feature, when each feature of each image is in the corresponding interval, putting the image into a set S1, when a certain feature of each image is not in the interval, putting the image into a set S2, arranging the images in the set S1 according to the number of connected domains from large to small, arranging the images in the set S2 according to the number of chromosomes from large to small, and setting the images in the set S2 according to the number of chromosomes from large to small; the length of the set S1 is added to the serial numbers of all the images in the set S2, then the two sets are combined, overall sequencing is completed, through the method, the images meeting the requirement for karyotype analysis can be automatically screened out, and manual operation is not needed; according to the sorting method, the processing speed is high, the time required for karyotype analysis is effectively shortened, and the efficiency is improved.

The invention discloses a method for dynamically measuring deformation and strain through digital speckle interference, and relates to the field of optical detection. The method comprises the following steps: firstly, building a speckle interference light path, secondly, selecting a to-be-detected area of a detected object, recording four speckle interference patterns with different phase shifts before and after deformation through a polarization camera, performing four-step phase shift on the recorded speckle interference patterns, and then performing subtraction to obtain a deformation wrapped phase diagram; and carrying out sine and cosine filtering and unwrapping processing on the deformation wrapped phase diagram to obtain a real phase diagram, thereby obtaining the deformation and strain of the measured object. According to the invention, the deformation and strain of the measured object can be dynamically detected in real time, the measurement light path is simplified, and the method is a high-precision full-field nondestructive measurement method.

The invention provides a water chlorophyll a concentration inversion method and system based on multi-modal data and a lightweight model, and relates to the technical field of water environment remote sensing evaluation. The method comprises the following steps: firstly, acquiring a Gaofeng No.5 satellite remote sensing image, a sentinel No.3 satellite image and ground actual measurement data, and completing image preprocessing and water body pixel extraction; constructing a hyperspectral index and an aquatic vegetation index, and fusing the hyperspectral index and the aquatic vegetation index with the water body temperature, the pH environmental factors and the spectral reflectivity to form a multi-dimensional feature sample set; a core feature subset is obtained through random forest and XGBoost coupling feature selection, and a lightweight student model is trained based on knowledge distillation; and constructing a to-be-predicted feature sample for the to-be-predicted time phase image and the environment factor, inputting the to-be-predicted feature sample into the lightweight student model to obtain a chlorophyll a concentration predicted value, and generating a spatial distribution map and a quality control map layer. According to the invention, high-precision, low-redundancy and efficient deployment chlorophyll a concentration inversion is realized.

The invention relates to the technical field of minimally invasive ablation, in particular to a liver tumor microwave ablation postoperative contraction deformation prediction method, which comprises the following steps: performing unified standardization processing on different periods of liver MRI images; accurately marking a preoperative liver tumor area, a postoperative ablation area and a postoperative liver tumor ghost area of the liver MRI image to obtain masks of the corresponding areas; designing a distance perception function as an attention parameter of different areas of the liver MRI image; constructing a multi-sequence distance guide complementary network model; selecting a loss function; selecting an elastic registration method to perform pre-operation and post-operation registration; and calculating an ablation safety boundary through the ablation area mask and the adjusted tumor mask. The liver tumor microwave ablation postoperative curative effect evaluation accuracy can be effectively improved, and a foundation is laid for formulating a subsequent treatment plan of a patient.

Snapshot phase-shifting diffraction modules and associated systems and methods are described that enable high spatial and temporal resolution phase imaging with high immunity to environmental factors such as vibrations and temperature changes. One example optical diffraction phase module includes a polarization grating to produce two circularly polarized light beams with opposite polarizations, a first lens to receive the two circularly polarized beams, and a spatial filter positioned at a focal plane of the first lens. The spatial filter includes two openings, one to spatially filter one of the two circularly polarized light beams, and another opening to allow another circularly polarized light beam to pass. The module also includes a second lens to focus the received light onto an image plane and to enable a phase measurement based a plurality of interferograms. The phase module can be incorporated into a microscope system that operates a reflection or a transmission mode.

The invention discloses a phase ambiguity reduction method based on speckle auxiliary phase tracking, which can accurately correct the motion error in a phase shift image by adding a speckle pattern in a phase shift projection sequence and calculating the sub-pixel projection point displacement between a front frame and a rear frame by using a DIC algorithm, obviously improves the precision of three-dimensional measurement, and improves the accuracy of phase ambiguity reduction. The projection points of the phase shift image are aligned to the same camera coordinate through a remapping technology, so that the motion blur error is effectively reduced, the accuracy of a measurement result is ensured, the projection points are tracked instead of tracking the motion of an object, the motion error with z-direction displacement can be corrected, a wrapped phase diagram is obtained through a phase shift method, and the accuracy of the measurement result is improved. And phase unwrapping is carried out by combining a double-frequency order method and a constant geometric constraint method, so that the accuracy and continuity of phase unwrapping are ensured, and the three-dimensional shape of the object is reconstructed by utilizing a triangulation method.

The invention relates to the technical field of optical precision interference measurement, in particular to a correction method for projection distortion detection of a primary mirror of a dynamic interferometer and a related product, and the method comprises the steps: obtaining a first coordinate and a second coordinate; determining a zero distortion point; obtaining an ideal imaging point set and an actual mapping point set after system projection through ray tracing; fitting to obtain a distortion function representing the distortion of the optical system; performing point-by-point correction on a phase diagram obtained by interference measurement, mapping pixel coordinates of the phase diagram into workpiece coordinates, obtaining interference measurement data corresponding to the actual surface of the primary mirror, and completing distortion correction; according to the method for correcting the surface shape projection distortion during interference measurement of the high-precision primary mirror of the large-aperture dynamic interferometer, the problem that in the process of assembling, repairing and developing the large-aperture Twyman type dynamic interferometer, the projection distortion effect of a detection system causes the loss of an accurate matching relation between a workpiece coordinate system and a detection coordinate system is solved.

The invention provides a brain low-frequency conductivity tensor modeling method, a brain low-frequency conductivity tensor modeling system, a medium and electronic equipment. The method comprises the following steps: acquiring a phase image of a fast spin echo sequence of a brain; acquiring a high-frequency conductivity image of the brain based on the phase image; collecting a diffusion weighted amplitude image of the brain based on a multi-excitation diffusion weighted sequence; based on the diffusion weighted amplitude image, using a multi-interlayer model to obtain diffusion microstructure parameters of the brain; generating a low-frequency conductivity tensor image of the brain based on the high-frequency conductivity image and the diffusion microstructure parameters; generating a digital head model based on the brain magnetization preparation fast gradient echo sequence; and generating a craniocerebral low-frequency conductivity tensor model based on the low-frequency conductivity tensor image and the digital head model. According to the brain low-frequency conductivity tensor modeling method and system, the medium and the electronic equipment, more accurate target point planning is provided for the tTIS.

The invention belongs to the technical field of automatic welding, and particularly relates to a welding seam detection method based on surface structured light. Comprising the following steps: (1) three-dimensional reconstruction of a workpiece: performing three-dimensional scanning on the workpiece by using a surface structured light three-dimensional scanner to obtain a coded image II, a phase diagram and three-dimensional data of the workpiece, and reading the coded image II, the phase diagram and the three-dimensional data through a computer; and (2) identifying welding seam feature points: performing welding seam feature identification or manual welding seam marking on the coded image II to obtain a two-dimensional coordinate set of a welding seam on the coded image II. Two-dimensional information and three-dimensional information of a surface structured light scanner are utilized at the same time, an efficient welding seam detection method integrating two-dimensional feature recognition and three-dimensional coordinate mapping is achieved through triangulation mapping, the problem of welding seam detection without obvious structural features is solved, meanwhile, compared with a split type multi-sensor fusion method, the welding seam detection method has the advantages that the welding seam detection efficiency is improved, and the welding seam detection cost is reduced. Calibration errors between devices are eliminated, the welding seam detection precision is improved, and the detection process is simplified.

The invention discloses a method for performing micro-hemorrhage focus recognition on brain images by using a deep learning algorithm, which comprises the following steps of: performing brain skull removal operation on magnetic sensitive weighted imaging images and phase images of a plurality of patients by using an HD-BET tool to obtain an intracerebral region map in a 2D image format; the method comprises the following steps of: firstly, firstly, carrying out brain detection on a brain region map, then, carrying out mesh division on each brain region map, dividing the brain region maps into mesh region maps with the same size, and then, training a YOLOv8 target detection model and a 3D-CNN network model by utilizing the mesh region maps, so that micro-hemorrhagic spots can be quickly predicted through the YOLOv8 target detection model, and then false positive micro-hemorrhagic spots are eliminated through the 3D-CNN network model. And a final micro-bleeding point detection result is obtained.

The invention provides a random forest shallow sea sediment classification method based on multi-temporal remote sensing image fusion, and relates to the technical field of sediment information extraction. Comprising the following steps: 1, collecting and preprocessing multi-temporal image data to obtain remote sensing reflectivity; 2, the water depth of each single-time-phase image is inverted, and the optimal water depth is obtained; 3, calculating bottom reflectivity characteristics of blue and green wave bands based on the optimal remote sensing image; 4, respectively calculating topographic features and spectral features based on the optimal water depth and the optimal remote sensing image; and 5, in combination with the bottom reflectivity features, the topographic features and the spectral features, carrying out random forest feature optimization and classification model training, and generating a substrate classification result. On the basis, the method solves the problems that an existing remote sensing image substrate classification method is insufficient in feature consideration, noise in a single-time-phase image can cause low classification precision, and therefore negative effects can be generated on accurate acquisition of substrate information.

The invention provides a self-adaptive reverse multiple exposure phase fusion technology, and relates to the field of high-dynamic-range three-dimensional reconstruction. The method includes the following steps that firstly, a saturated pixel mask matrix is obtained by judging pixels of a high-reflection area, then uniform grey-scale maps are selected step by step through a gradient adjustment method to obtain the optimal local projection intensity, and local coding and decoding are conducted on the high-reflection area through the optimal projection intensity pattern. And replacing an invalid phase value in the high-SNR image with a local phase value under the optimal projection intensity through an edge matching and phase fusion technology, and obtaining high-quality complete three-dimensional imaging information through coordinate mapping after a complete phase diagram is obtained. According to the method, the optimal projection intensity is adaptively determined through a small number of patterns, meanwhile, the high SNR of imaging patterns is guaranteed, and high-precision complete three-dimensional imaging is carried out on an HDR object. Experimental results show that in a traditional method, a large number of point clouds are lost in a saturated pixel area, the imaging integrity and the measurement precision cannot achieve the normal imaging effect, and the algorithm of the invention shows high integrity and high point cloud density in a dark area and a high-reflection area.