980 results about "Space resolution" patented technology

The invention relates to an underwater image enhancement method based on wavelet Mama, which aims at the problems of color shift, low contrast ratio and fuzzy details of an underwater image, extracts shallow layer features of an underwater low-quality image, inputs the shallow layer features into a multi-scale coding network, and performs joint enhancement on low-frequency color information and high-frequency detail features by using a wavelet Mama unit. Global semantic features are obtained by combining down-sampling layer-by-layer compression, and global modeling of color offset correction and contrast enhancement is completed; the spatial resolution is recovered through up-sampling, and reconstruction and enhancement of texture details and color information are realized in combination with jump connection and a wavelet Mama unit; and finally, the features are mapped to an image domain and fused with input image residual errors, and an enhanced underwater image with natural color, clear texture and balanced contrast is generated. According to the method, frequency domain-space domain joint feature modeling is realized through the wavelet Mama unit, and the color authenticity, the structural definition and the detail perceptibility of the underwater image are remarkably improved.

The invention relates to the technical field of medical image detection, and discloses a medical image quality detection method based on image processing. The method comprises the following steps: acquiring medical image data to be detected, wherein the medical image data comprises a multi-modal scanning image sequence and corresponding acquisition parameters; the medical image data are preprocessed, standardized image data are generated, and the standardized image data comprise unified parameters of spatial resolution, gray scale range and noise level; extracting structural features of the standardized image data, wherein the structural features comprise tissue boundary gradient distribution, texture consistency and local contrast information; constructing a quality evaluation model according to the structural features, wherein the quality evaluation model analyzes a mapping relationship between the structural features and preset quality indexes through a dynamic convolutional network; and outputting a quality defect detection result based on the quality evaluation model, wherein the quality defect detection result marks an image region with artifacts, fuzziness or distortion.

The invention relates to the technical field of petroleum and natural gas engineering, and discloses an oil and gas reservoir optimized mining method based on multi-modal data, which comprises the following steps: constructing an oil and gas reservoir multi-modal data acquisition system, seismic wave field data, logging interpretation data, production dynamic data, micro-seismic monitoring data, underground temperature and pressure time sequence data and shaft structure parameter data are obtained through the acquisition system; and performing space-time alignment processing on the acquired multi-modal data, establishing a unified geological coordinate system and a time reference, and eliminating data isomerism caused by different acquisition frequencies and spatial resolutions. A six-dimensional heterogeneous data acquisition system covering a seismic wave field, well logging interpretation, production dynamics, micro-seismic monitoring, an underground temperature and pressure time sequence and shaft structure parameters is constructed, so that the depiction precision of a reservoir porosity field, a permeability field, a saturation field and a pressure field is essentially improved.

The invention provides a medical image segmentation method and system based on spatial detail enhanced vision, and relates to the technical field of image processing. The method comprises the following steps: performing initial feature mapping on an input and preprocessed medical image to obtain an embedded feature map; the embedded feature map is input into an encoder for feature extraction, and multi-scale features are obtained; the encoder comprises a plurality of encoding stages, and the number of channels is doubled and the spatial resolution is halved through down-sampling operation between the encoding stages; the multi-scale features are subjected to up-sampling and spatial enhancement reconstruction step by step through a decoder, and a high-precision segmentation result is generated; wherein the decoder comprises a plurality of decoding stages, the decoding stages correspond to the encoding stages, and feature fusion is carried out between the corresponding stages of the encoder and the decoder through jump connection. According to the method, the boundary description precision and the segmentation robustness of the low-contrast image can be improved without increasing the linear complexity, and the method is suitable for medical image segmentation scenes of skin lesions, gastrointestinal polyps and the like.

The invention relates to an optical cable monitoring and data analysis method and device, equipment and a medium. The method comprises the following steps: acquiring current state data of an optical fiber link, and dynamically generating adaptive pulse coding parameters and matched filter parameters based on the data by using a deep reinforcement learning agent, so as to generate and inject coded pulses into the optical fiber link; and then receiving an echo signal, performing matched filtering processing on the signal by using the generated matched filter parameter to obtain a filtering output signal, further calculating a performance index vector and a reward signal, and finally updating a strategy network of the intelligent agent by using the reward signal, thereby realizing continuous optimization of pulse coding and matched filtering parameters. Therefore, the system can automatically adapt to the dynamic change of an optical fiber link, the spatial resolution, the dynamic range and the signal-to-noise ratio are effectively improved, and the adaptive capability and the robustness of the system are enhanced at the same time.

The invention relates to an ultrasonic phased array full-focusing imaging method for acoustic wave multi-path propagation compensation of a layered composite material, which is characterized by comprising the following steps of: arranging an ultrasonic phased array probe with N array elements on a composite material piece to be detected, obtaining complete echo data by taking each array element as a transmitting array element and taking all array elements as receiving array elements in sequence in a full-matrix acquisition mode; a two-dimensional pixel grid is established in an imaging area, a Monte-Carlo perturbation mechanism used for representing the acoustic wave propagation uncertainty in the layered composite material is introduced into a propagation model for each emission array element-receiving array element-pixel point combination, and the acoustic beam emission direction, the propagation path and the equivalent acoustic velocity are subjected to random perturbation, so that the acoustic wave propagation uncertainty in the layered composite material can be represented. Generating a plurality of equivalent sound wave propagation paths, calculating two-way propagation time of each path, and performing weighted statistics on energy contributions of different paths according to array element sound beam directivity to obtain equivalent propagation time delays of pixel points; and performing delay correction on a full-matrix echo signal by using the equivalent propagation time delay, and performing coherent superposition on signals of all transmitting-receiving array element combinations to realize dynamic focusing of a full-pixel grid, and finally reconstructing an ultrasonic full-focusing imaging image with high resolution and high signal-to-noise ratio. The time delay error caused by propagation path deviation and sound velocity non-uniformity in the layered composite material can be effectively compensated, the focusing precision and spatial resolution of full-focusing imaging are improved, meanwhile, an existing TFM imaging frame does not need to be changed, and the method has good engineering implementability and popularization and application value.

The invention relates to the technical field of geological early warning, in particular to a dangerous rock mass instability analysis method, system and equipment based on a space-time diagram neural network, by fusing unmanned aerial vehicle LiDAR, multispectral data, meteorological radar data and the space-time diagram neural network (ST-GNN), the system realizes sub-meter spatial resolution and minute-level time response, and the stability of dangerous rock mass instability analysis is improved. The four-dimensional (time and space) analysis result of the instability probability of the dangerous rock mass is obtained through high-precision space-time modeling, the problems that a traditional geological disaster early warning system is low in resolution ratio, slow in response and high in misinformation are solved, the comprehensiveness, accuracy and reliability of instability prediction of the dangerous rock mass are improved, and the early warning effect is good. And full-chain intelligent closed-loop management of real-time data acquisition-dynamic prediction-early warning push-feedback optimization is supported, the emergency decision time is shortened by real-time rainfall superposition risk thermodynamic diagrams, and the attenuation rate of long-term prediction precision is reduced by dynamically fusing newly added geological data and instability events through incremental learning.

The invention relates to the technical field of geological engineering, and discloses a geological safety risk dynamic assessment method based on multi-source data fusion, which comprises the following specific steps: step 1, collecting and standardizing multi-source geological data; 2, performing semantic fusion and conflict resolution on the geologic features; 3, constructing a dynamic risk assessment model; 4, risk situation real-time updating and early warning are carried out; the satellite remote sensing system in the first step adopts the synthetic aperture radar interference measurement technology, the spatial resolution is better than 3 meters, the revisit period is shorter than 7 days, and the earth surface deformation monitoring precision reaches the millimeter level. Through standardized processing and semantic fusion of the multi-source geological data, the problem of heterogeneous data integration is effectively solved, the data basic quality of risk assessment is improved, a space-time coupling neural network model is adopted, nonlinear features and space-time correlation characteristics of geological risk evolution are accurately captured, and prediction precision and timeliness are improved.

The invention discloses a mountainous area wind field downscaling optimization method based on high-precision simulation, and relates to the technical field of meteorological numerical simulation, and the method comprises the steps: obtaining the terrain elevation data and meteorological monitoring data of a target mountainous area, recognizing slope abrupt change points, constructing a spiral sampling path, calculating a slope change value, determining the distance between sampling points, and generating target sampling data; performing orthogonal decomposition on the sampled data to obtain a frequency component, calculating a surface fluctuation coefficient, and constructing a boundary disturbance equation to obtain surface stress distribution; calculating an airflow motion state based on surface stress distribution, solving a vorticity equation to obtain vorticity characteristics of a leeward area, and calculating airflow motion correction parameters; and carrying out downscaling iterative operation on the corrected parameters and the meteorological monitoring data to generate target area wind field data with hectometer-magnitude spatial resolution. The simulation precision of the local wind field under the complex terrain condition is improved.

The present invention relates to a high temperature and drought composite disaster monitoring and early-warning method and system, belonging to the technical fields of disaster risk assessment and early warning. Internal correlation features and abnormality information of high temperature and drought events are input into a model, multi-time-space scale features of the high temperature and drought events can be identified accurately, high event identification accuracy and space resolution are achieved, the progress can be predicted progressively, the drought and high temperature threshold change can be monitored closely, and fine forecasting and early warning can be performed in different periods, regions and intensities, thereby ensuring that indicators are in the same time scale, avoiding the complication of the high temperature and drought process caused by frequent time and space discontinuities of the indicators in a single point or small region, and ensuring the suitability for any periods of the process.

The invention belongs to the technical field of image processing, and particularly relates to a mixed U-Net brain tumor segmentation method and system based on quantum attention, and the method comprises a data preprocessing module which is used for obtaining and preprocessing a medical image; the image segmentation module is used for segmenting an image through a trained QHAU-Net segmentation model, and the QHAU-Net segmentation model comprises an encoder used for extracting multi-level features of the input image; the bottleneck layer is used for receiving the semantic features of the highest dimension output by the encoder and generating enhanced features modulated by quantum through a quantum multi-head attention module; the decoder is used for recovering the spatial resolution of the feature map and fusing the features through jump connection; the quantum boundary enhancer is used for performing quantum enhancement optimization on the output features of the decoder and outputting the output features; and the result output module is used for outputting and visualizing the segmentation result graph. The problem of low boundary medical image segmentation precision in medical image processing is solved.

The invention discloses a corn disease recognition method based on an improved generative adversarial network. The corn disease recognition method comprises the steps that S1, an initial low-resolution image of a corn field is collected and obtained through an unmanned aerial vehicle; s2, performing super-resolution reconstruction on the initial low-resolution image by using an improved generative adversarial network model; the model construction comprises the following steps: S2.1, constructing a shallow feature extraction layer; s2.2, constructing a deep feature extraction network based on a plurality of RRDB nested residual dense blocks; s2.3, constructing an attention module based on a space and channel dual attention mechanism; s2.4, a multi-scale texture enhancement module is constructed through multi-scale convolution and smooth branches; s2.5, constructing a global residual connection layer; s2.6, constructing an adaptive hybrid up-sampling module based on transposed convolution and stable up-sampling; s2.7, performing mapping output on the features after up-sampling; and S3, carrying out disease prediction on the high-resolution reconstructed image. According to the method, details such as spatial resolution and texture of the unmanned aerial vehicle high-altitude flight remote sensing image are improved, and then the corn disease monitoring precision is improved.

The invention discloses a total primary productivity estimation method and system based on multi-model coupling deep learning, and the method comprises the steps: obtaining the multi-source data of meteorological data, remote sensing images, latent heat flux, sensible heat flux and solar radiation, carrying out the quality control, missing value processing and nearest neighbor interpolation of different data sources, and carrying out the prediction of the total primary productivity. Unifying to a target spatial resolution and a time resolution; in a light energy utilization rate (LUE) model family, a solar radiation phase factor is introduced into a photosynthetically active radiation absorption ratio (FPAR) to obtain a phase modulation type FPAR, drought duration is introduced into a water stress function f (W) to obtain an exponential decay type f (W), and a GPP time sequence of a plurality of improved mechanism models is calculated according to the exponential decay type f (W); extracting spatial texture features from a remote sensing image stack by using a convolutional neural network (CNN), and performing cross-modal fusion on the spatial features and the GPP estimated by the plurality of improved mechanism models in a gating mode to form fusion representation; carrying out learning and collaborative optimization on the fusion representation of the GPP and CNN spatial features estimated by the plurality of improved mechanism models by adopting a gradient lifting tree model; and high-precision estimation of the GPP is realized through multi-model collaborative optimization. The method aims at solving the problem that a traditional light energy utilization rate model is insufficient in response under the extreme environment conditions of drought and intense radiation, the adaptability limitation of a traditional single model under the complex environment is broken through, and therefore high-precision GPP estimation under the complex environment is achieved.

The invention discloses a cell occlusion relation estimation method and system based on a pixel-level focus evaluation curve. The method comprises the following steps: acquiring a multi-focal plane cell image sequence of the same view at different focusing positions; performing cell instance segmentation on the image sequence to obtain a two-dimensional cell instance mask; calculating the definition value of each pixel in the mask on each focal plane, and constructing a pixel-level focus response curve; determining a focus depth interval of each pixel according to the peak position of the focus response curve and a preset relative threshold value; and comparing focus depth intervals of different cell pixels in the cell overlapping region, and judging the up-and-down shielding relationship between the cells. According to the method, automatic analysis from the original microscopic data to the cell upper and lower layer relation is realized through pixel-level depth modeling and interval comparison, the spatial resolution and credibility of occlusion relation judgment are improved, the compatibility with a conventional microscopic imaging process is high, and deployment and popularization are easy.

The invention relates to the technical field of remote sensing image processing and computer vision, in particular to a hyperspectral remote sensing image reconstruction method based on iterative optimization and depth prior. According to the method, complementary information of a low-resolution multispectral image and a high-resolution panchromatic image is utilized, the spatial resolution and the spectral fidelity of the high-spectral image can be improved at the same time, and the complementary information of the low-resolution multispectral image and the high-resolution panchromatic image is effectively utilized; through a spectrum low-rank decomposition and alternate optimization strategy, the complexity of a high-dimensional optimization problem is reduced, and the stability of iterative solution is improved; and after the deep denoising network is introduced, the denoising capability and the feature expression capability of the subspace coefficient image are further enhanced, so that the generated hyperspectral image is superior to the existing method in the aspects of spatial details and spectral authenticity, and the hyperspectral image has relatively strong robustness and wide application prospects.

A deep learning-based super-resolution fluorescence lifetime imaging microscopy (SR-FLIM) method includes the steps of: S1, performing fluorescence microscopic imaging on a sample to obtain confocal intensity images and stimulated emission depletion (STED) intensity images at a same location; S2, co-registering the acquired confocal and STED intensity images; S3, pairing the co-registered confocal and STED intensity images as input (Input) and ground truth (GT) to assemble a dataset; S4, partitioning the dataset into training and validation sets following a predefined ratio; and S5, constructing a network, and selecting hyperparameters and an optimizer. This method may achieve SR-FLIM within a conventional confocal FLIM system, surpassing spatial resolution limitations of FLIM, breaking through resolution barriers of conventional optical microscopy, while preserving normal fluorescence lifetime characteristics of fluorescent probes.

The invention discloses a point cloud-driven thoracic cavity whole organ dynamic reconstruction and respiration monitoring method and system, and the method comprises the steps: constructing a three-dimensional geometric model corresponding to each target structure in a thoracic cavity based on the thoracic medical image data of a target object, and converting the three-dimensional geometric model into static point cloud data; driving the static point cloud data to perform dynamic deformation simulation in a respiratory cycle based on a respiratory movement rule of the target object to obtain dynamic point cloud data synchronized with a respiratory time phase; performing time-space synchronization association on the dynamic point cloud data and the electrical characteristic parameters, and constructing a digital twin thoracic cavity model; deploying a virtual sensing assembly which dynamically deforms along with the thoracic cavity in the digital twin thoracic cavity model, simulating and monitoring the dynamic breathing process of the target object, and generating a virtual physiological signal; and performing signal processing on the virtual physiological signal to obtain a dynamic reconstruction image reflecting respiratory movement. Continuous simulation of the full-breathing movement process is achieved, and the temporal-spatial resolution and diversity of a virtual database are improved.

The invention discloses a signal reconstruction and noise suppression method and system for a distributed Raman temperature measurement sensing system, and the method comprises the steps: obtaining original anti-Stokes light and Stokes light time domain signals of the distributed Raman temperature measurement sensing system, and constructing a space-time signal matrix; performing preprocessing and data enhancement on the space-time signal matrix; constructing a dual-path feature fusion network for Raman signal reconstruction and noise suppression, and initializing the network; defining a loss function fusing temperature physical constraints, and training the dual-path feature fusion network; and inputting an original Raman signal to be processed into the trained network, outputting the reconstructed anti-Stokes and Stokes signals with the high signal-to-noise ratio, and carrying out temperature demodulation according to the anti-Stokes and Stokes signals with the high signal-to-noise ratio. While the spatial resolution is maintained or improved, the noise of the Raman temperature measurement system is effectively suppressed, especially the real signal of a temperature abrupt change point can be recovered, the temperature demodulation precision, stability and reliability are improved, and reliable data are provided for infrastructure safety monitoring.

The invention provides a large-range landslide susceptibility evaluation method and system based on a physical constraint U-Net model, and relates to the technical field of landslide disasters, and the method comprises the steps: obtaining an evaluation factor of a target region, and constructing a historical data set related to the evaluation factor and a landslide region; combining a U-Net model and physical constraints to construct a landslide susceptibility evaluation model; training the landslide susceptibility evaluation model through the historical data set; and inputting the real-time evaluation factor of the target area into the trained landslide susceptibility evaluation model for prediction to obtain a large-range landslide susceptibility evaluation result of the target area. The method solves the problems that the evaluation result obtained by the existing landslide susceptibility evaluation method has lower spatial resolution compared with the original input data, and the evaluation result is poor in effect when the evaluation area is larger.

The invention discloses a TAVR postoperative blood flow parameter complementing method based on CT-MRI fusion, and belongs to the technical field of crossing of medical image processing and computational fluid mechanics. A CT valve frame registration technology is adopted, an invisible artificial valve leaflet three-dimensional structure is reconstructed through valve frame information in a CT image and is coupled with a dynamic heart model, a complete valve mechanical and functional simulation platform is constructed, the problem of an imaging blind area is solved, hemodynamic parameters of an inlet and an outlet of a postoperative area are further collected in combination with 4D Flow MRI, and the artificial valve leaflet three-dimensional structure is obtained. The blood flow conditions of the upstream and the downstream of the artificial valve area at each stage of contraction and relaxation of the TAVR postoperative part are comprehensively restored, the temporal-spatial resolution and the physical authenticity of the boundary condition in fluid-solid coupling simulation are greatly improved, and the more real passive response of the valve leaflet under the action of the blood flow is obtained; and a reliable basis is provided for quantitative and mechanical risk analysis of postoperative complications such as thrombosis, perivalvular leakage and valve leaflet fatigue.

The invention discloses a distributed optical fiber temperature measurement verification device and method.The device comprises a constant temperature device, a lifting device, a measurement and control host and automatic verification software, the constant temperature device comprises a constant temperature bath and a constant temperature control device, the constant temperature bath comprises a mixing area and a working area, the lifting device drives a test tool to ascend and descend, and the test tool fixes a temperature measurement optical cable; the temperature measurement optical cable is connected with the temperature measurement host, the temperature measurement host feeds back a temperature measurement value to the measurement and control host, the measurement and control host compares the measurement value with a temperature value of a medium in the constant temperature device, and distributed optical fiber temperature measurement verification is achieved. The method passes a temperature measurement range and accuracy test; performing a temperature resolution test; performing a spatial resolution test; testing positioning accuracy; a single-channel test time test comprises six tests, data are automatically collected and analyzed, and a verification report is generated. The device is reasonable in structural design and high in universality, automatic control over the verification process is achieved, data collection, analysis, calculation and report generation are all automatically completed, and verification efficiency is improved.

The invention belongs to the technical field of temperature characterization, and discloses a semiconductor device temperature characterization method and system. The semiconductor device temperature characterization scheme provided by the invention comprises the following steps: bearing a semiconductor device and adjusting the position of the semiconductor device by using a device bearing and adjusting unit; respectively emitting detection laser and heating laser by using a detection laser emitting unit and a pumping laser emitting unit, and irradiating the detection laser and the heating laser to a test area of the semiconductor device; collecting and recording a temperature measurement signal reflected by the semiconductor device by using a temperature measurement unit; the device operation control unit is used for providing an electric signal for the semiconductor device and synchronously triggering the temperature measurement unit to record a temperature measurement signal so as to ensure the synchronization of online work and temperature measurement of the semiconductor device; and obtaining transient temperature characterization information of the semiconductor device based on the temperature measurement signal by using the main control unit. According to the invention, temperature characterization of nanosecond-level time and submicron-level spatial resolution of the semiconductor device can be realized.

The invention relates to the technical field of water and soil loss dynamic monitoring and early warning methods in water conservancy projects, and discloses a water and soil loss dynamic monitoring and early warning method, which comprises the following steps: synchronously acquiring multi-dimensional data of terrain, rainfall, soil humidity and vegetation coverage through unmanned aerial vehicle remote sensing and a ground sensor network; registering and correcting the multi-source heterogeneous data based on a space-time fusion algorithm; the improved SWAT model is combined with a machine learning algorithm to dynamically simulate a water and soil loss process; and triggering graded early warning according to a preset risk threshold and generating prevention and control suggestions. The system comprises a data acquisition module, a data fusion processing module, a water and soil loss simulation prediction module and an early warning response module. According to the scheme, the real-time performance, the spatial resolution and the early warning accuracy of water and soil loss monitoring are remarkably improved, and efficient decision support is provided for ecological protection of water conservancy projects.

The invention belongs to the technical field of remote sensing image processing, particularly relates to an urban remote sensing image segmentation method and system based on bidirectional coordinate attention and multi-scale adaptive feature fusion, and provides a remote sensing image semantic segmentation neural network architecture taking an attention re-calibration module as a decoder core. Wherein the encoder path gradually extracts multi-scale feature representation through cascaded residual convolution blocks and down-sampling operation to form a feature pyramid of which the spatial resolution is reduced step by step and semantic information is enhanced step by step; the decoder path gradually recovers the spatial resolution through cascaded up-sampling and feature refining operations to generate a precise segmentation mask; a space-channel dual attention re-calibration module oriented to a decoding stage is provided, through explicit coding of space coordinate direction information and combination of global channel dependence modeling, features beneficial to semantic discrimination are adaptively enhanced in the feature fusion and resolution recovery process, and therefore segmentation precision and consistency are improved.

The invention discloses a plateau mountain area avalanche susceptibility intelligent assessment method, and relates to the technical field of high mountain ice and snow disaster space early warning and risk mapping. According to the method, the influence of the microtopography is captured by adopting high resolution, the evaluation accuracy is improved, the control effect of the slope microtopography (the scale of 30-50 m) on the accumulated snow stress field can be accurately captured by adopting a factor map layer (based on data such as 12.5 mALOSPALSARDEM and Landsat-8OLI images) with the spatial resolution of 12.5 m, and compared with traditional 30mDEM and hectometer-level meteorological reanalysis data, the method has the advantages that the evaluation accuracy is greatly improved, and the evaluation efficiency is improved. Spatial differences of key factors such as terrains, earth surfaces and meteorology can be depicted more carefully, more accurate basic data support is provided for avalanche susceptibility assessment, and missing of key microtopography influence factors due to insufficient resolution is avoided.

The invention discloses a boiler furnace three-dimensional combustion temperature field reconstruction method and system, and the method comprises the steps: S1, multi-mode sensing network deployment, S2, time-space synchronization data collection, S3, sound wave flying time extraction and acoustic modeling, S4, flame image processing and radiation temperature conversion, S5, three-dimensional temperature initial field generation, and S6, multi-mode joint objective function construction. S7, carrying out regularization constraint iteration solution; and S8, carrying out three-dimensional temperature field visualization output. According to the method, high penetrability and sensitivity of sound waves to medium temperature and rich radiation information contained in flame images are fully utilized, and the precision and spatial resolution of temperature field reconstruction are remarkably improved by establishing a joint objective function and performing collaborative optimization. Meanwhile, a regularization method and an adaptive filtering technology are introduced, interference caused by measurement noise and uncertain problems is effectively suppressed, and the stability and robustness of the reconstruction process are enhanced.

The invention provides a power transmission line fire risk grade zoning prediction method and system, and relates to the technical field of intelligent power grid environment perception and risk early warning, and the method comprises the steps: data collection and multi-source space-time fusion: collecting the regional characteristics of the historical fire occurrence space-time, real-time meteorological parameters, vegetation types and combustible rates of a power transmission line to form a multi-source data set, the multi-source data set realizes alignment coding through a unified geographic grid and a timestamp, and a fire risk sample set is generated. Through a multi-source data deep fusion and grid space-time alignment mechanism, unified coding and coupling modeling of regional features such as historical fire occurrence space-time, real-time meteorological parameters, vegetation types and combustible rates are realized; risk zoning precision and spatial resolution are remarkably improved, zoning granularity can be refined in practical application, the resolution is far better than that of a traditional method, and the overlapping rate of a risk zoning result and actual fire distribution is improved.

The invention relates to the technical field of bill text recognition, and discloses a bill text recognition system and method based on deep learning. The method comprises the following steps: acquiring target bill original image data containing a multi-channel pixel matrix and spatial resolution information; based on a matching result of the bill edge features and a preset template, geometric distortion correction is carried out on the original image, and a corrected bill image is generated; inputting the corrected image into a pre-training text region detection network to obtain positioning information containing text line boundary coordinates and region confidence; text line image blocks are extracted according to the positioning information, character segmentation preprocessing is executed, and a character-level image sequence is generated; calling a deep character recognition model to classify the sequence character by character, and generating an initial text recognition result; semantic verification and error correction are performed on the initial result based on the bill type knowledge base, final structured text data are generated, and the processing requirements of bills of different types and qualities can be met.

The invention provides a defect detection method and device for improving a YOLO model based on an attention mechanism. The method provided by the invention comprises the following steps: acquiring image data of a to-be-detected industrial product; the improved YOLOv10 model performs feature extraction, multi-scale feature fusion and defect positioning identification on the industrial product image data to obtain a defect detection result of the industrial product; wherein the SE module is used for strengthening a channel dependency relationship and detail representation of local detail features, and the CBAM module is used for modeling channel attention and space attention on global semantic features; the bridging layer controls middle and low layer features and high layer features to keep uniform channel dimension and spatial resolution by adjusting the size and stride of a convolution kernel; introducing a 1 * 1 convolutional layer after Concat operation of multi-scale feature fusion, and performing channel compression, linear fusion and semantic alignment on multi-scale fusion features; and constructing a loss function fusing the GHM loss and the dynamic IoU loss.

The invention discloses a three-dimensional electromagnetic inverse scattering imaging method of a diffusion model based on physical constraint embedding. The method comprises the following steps: 1, in a specific three-dimensional electromagnetic inverse scattering imaging environment, defining a dielectric constant of an imaging area, arranging a transmitting and receiving antenna array at the periphery, and obtaining scattering field data of a target area for constructing an input and output sample pair of a diffusion model; 2, constructing a diffusion model on the three-dimensional grid space; 3, introducing a physical consistency constraint term based on a Maxwell equation set in the training stage of the diffusion model; and 4, synthesizing the generation loss and the physical constraint loss of the diffusion model, constructing a joint optimization objective function, and carrying out iterative updating on network parameters by adopting an end-to-end back propagation algorithm until the model converges. According to the method, the spatial resolution and continuity of three-dimensional imaging are remarkably improved, the stability and robustness of inverse scattering reconstruction are improved, and the physical consistency is enhanced. And high-precision and interpretable three-dimensional electromagnetic inverse scattering imaging can be realized.