569 results about "Spatial consistency" patented technology

The invention discloses a time sequence marine ecological environment early warning monitoring data abnormal point detection method, and relates to the technical field of marine ecological environment analysis, and the method comprises the steps: collecting a plurality of types of marine monitoring indexes, constructing a multi-source and multi-dimensional time sequence data set, and carrying out the preprocessing and unification of the data into a standard input format. And based on the causal relationship and co-evolution characteristics between the factors, constructing a graph structure model, and depicting a coupling propagation path between the factors. Generating a local time sequence model by adopting a sliding window, constructing an anti-fact path, comparing the propagation difference of an original path in a map structure, and identifying a key disturbance moment; and in combination with historical event similarity and space consistency information, the early warning credibility of abnormal points is evaluated, and efficient and interpretable graded early warning response is realized. According to the method, accurate identification and credible early warning of abnormal points are realized, causal modeling, anti-factual reasoning and hierarchical response capabilities are realized, and the intelligence and interpretability of marine ecological monitoring are improved.

A defect prediction method based on a multi-feature parallel multi-stage neural network (MF-PMSNN), includes: obtaining a trajectory dataset, and preprocessing data of a defect of a workpiece in additive manufacturing (AM); building an MF-PMSNN, and evaluating an output classification result based on evaluation indicators; and performing real-time defect prediction, and deploying a trained MF-PMSNN model to a production environment. The present disclosure combines and effectively matches thermal imaging-based in-situ monitoring data and X-ray computed tomography (XCT)-based in-situ monitoring data to ensure temporal and spatial consistency between the thermal imaging-based in-situ monitoring data and the XCT-based in-situ monitoring data. In this way, a molten pool status and a pore of the workpiece can be captured more comprehensively. The MF-PMSNN is proposed to obtain a molten pool status and the porosity distribution in the data and perform defect prediction.

The invention relates to the technical field of image processing, in particular to a remote sensing monitoring method and system for cultivated land protection, and the method comprises the following steps: obtaining a remote sensing image sequence of a cultivated land region at a differential time node, extracting pixel gray values of the same land parcel, arranging the pixel gray values according to time, calculating the ratio of a difference value between adjacent frames to an intra-frame gray average value, and obtaining a remote sensing image sequence; and marking positions continuously exceeding the fluctuation reference ratio, and generating a pixel region set. According to the method, a sampling frequency threshold optimization strategy is adopted, abnormal areas are screened in combination with albedo change rate, short-time fluctuation interference is reduced, time sequence consistency is enhanced, a solar incident angle and a terrain slope are introduced to calculate projection influence, a slope surface shielding area is eliminated, and space grid marking and boundary integrity judgment are adopted; the method improves the continuity of cultivated land distribution information, combines time sequence analysis, sampling optimization, physical characteristic compensation and spatial consistency correction, enhances the fine recognition of the dynamic change of cultivated land, and improves the monitoring precision and data reliability.

The invention relates to the technical field of computer vision and industrial automation, in particular to a shoe sole three-dimensional gluing contour extraction system and method based on vision single-line laser, and aims to solve the problems that virtual calibration target spots cannot be accurately generated based on shoe sole geometry, the positions and sizes of the target spots are difficult to determine by combining curvature extreme values and principal component analysis in the prior art, and the production cost is low. The problem that a double-branch deep learning model cannot be adopted to fuse feature prediction transformation, and the re-projection error is increased is solved; a virtual calibration target spot is automatically generated based on sole geometry through a feature fusion calibration module, a grid is generated through point cloud processing and Poisson reconstruction, the position and size of the target spot are determined by combining a curvature extreme value and principal component analysis, a corresponding relation is established by utilizing two-dimensional and three-dimensional feature matching, initial alignment is realized through ICP and re-projection error optimization, and the target spot position and size are determined. A double-branch deep learning model is adopted to be fused with feature prediction transformation, iterative optimization is carried out through space consistency errors, and re-projection errors are reduced.

The invention discloses a cross-modal remote sensing target detection method and system based on spatial consistency constraint and deep feature alignment, belongs to the field of computer vision and remote sensing science and technology and the technical field of machine learning and deep learning, and solves the problem that a conventional cross-modal method does not fully consider feature hierarchy difference. According to the invention, an improved teacher-student network model is constructed and comprises a student branch network, a teacher branch network and an optimization module for performing pseudo-label optimization, non-monitoring learning and supervised learning on the student branch network and the teacher branch network; training the improved teacher-student network model by adopting the target domain data set and the source domain data set to obtain a trained improved teacher-student network model; and carrying out cross-modal remote sensing target detection on a to-be-detected target domain image by adopting the trained improved teacher-student network model. The method is used for cross-modal remote sensing target detection.

The invention provides a liver focus three-dimensional modeling method, and belongs to the technical field of image processing based on computer vision. Firstly, a multi-view spatial registration method based on optical flow optimization is designed, pixel-level displacement information of different view images is estimated by calculating an optical flow field, accurate image alignment is achieved, and spatial consistency of three-dimensional reconstruction is improved. And secondly, a three-dimensional reconstruction strategy based on two-dimensional focus segmentation is proposed, the two-dimensional focus segmentation is completed by adopting a lightweight U-Net variant, and a segmentation result is mapped to a three-dimensional space through a voxel probability projection method, so that 3D focus reconstruction is realized, and the calculation cost is reduced. And finally, extracting high-frequency features of the three-dimensional model by adopting a local edge enhancement method based on a Laplacian operator, and strengthening a focus boundary and a key anatomical structure through interpolation optimization, so that the three-dimensional model is more accurate and clearer. Compared with a traditional method, the method has the advantages that the mode of purely depending on image superposition is avoided, and the accuracy of three-dimensional modeling is improved.

The invention relates to an underwater topographic measurement method based on laser radar and visual fusion, which comprises the following steps: synchronously acquiring underwater point cloud, images and physical environment sensing data, and endowing a unified space-time label to realize the space-time consistency of multi-source data; according to the method, the data quality of different modes is improved by means of preprocessing, denoising, scale normalization, feature enhancement and the like, an environment interference weight matrix is constructed in combination with an environment sensing model, feature extraction and matching algorithm parameters are adaptively adjusted according to different environment states, and multi-scale feature description, spatial consistency and physical constraint criteria are established, so that the multi-modal data quality is improved. According to the scheme, high-reliability feature matching between the point cloud and the image is achieved, finally, through environment-driven iterative optimization and fusion, the robustness and precision of space registration are improved, high-precision multi-modal data automatic registration can be stably achieved in the underwater dynamic environment, adaptability is high, and environment perception and space measurement quality is effectively improved.

The invention discloses a magnetic field measurement method and system based on a multi-sensor fusion technology, and relates to the technical field of sensor fusion and magnetic field measurement, and the method comprises the steps: deploying a multi-sensor data array, carrying out the adaptive initialization of a bistable SR parameter range, defining an SR system differential equation, and carrying out the iterative optimization through employing an MPA population. Carrying out Hilbert transform edge detection on enhanced signal component data, calculating an array inclination angle, carrying out abbe error and bidirectional projection error compensation, and carrying out metasurface grid coordinate quantization mapping; the collected and cross-scale magnetic field data set is preprocessed and packaged into data cells, quality evaluation and weight distribution are carried out on the data cells, and extended Kalman filtering data fusion is carried out; by introducing a bistable stochastic resonance system and an MPA population optimization algorithm, a weak magnetic field signal is obviously enhanced, and by calculating an array inclination angle and compensating an Abbe error and a bidirectional projection error, the space consistency of a measurement result is improved.

The invention discloses an image feature matching optimization method based on intra-class space consistency in the technical field of computer vision and image processing. The method comprises the following steps: feature point extraction and preliminary matching: extracting feature points from a query image and a reference image and performing preliminary matching; initialization and transformation model estimation: initializing a matching point set and a residual error, and calculating an initial transformation model; error calculation and matching point set updating: calculating the error of the matching point pair, and updating the matching point set by adopting a dynamic screening method; performing residual optimization: judging whether the optimal condition is reached or not based on the residual, and deciding whether to continue iteration or not; performing intra-class space consistency clustering and isolated cluster elimination: performing clustering analysis after the optimal residual error is obtained, and eliminating isolated clusters based on an intra-class space consistency separation ion structure; and outputting a result: outputting a matching point set after the isolated clusters are removed. The method solves the problem that a traditional feature matching method is difficult to completely remove mismatching in a complex scene and is sensitive to noise.

The invention relates to the field of brain tumor analysis, and discloses a brain tumor image analysis system based on artificial intelligence, comprising: a spatial alignment unit for acquiring original image data of the brain of a subject; performing spatial alignment on the original image data according to a cross-modal registration algorithm to obtain standardized image data; the feature extraction unit is used for performing tumor region initial segmentation on the standardized image data according to a three-dimensional convolutional neural network so as to obtain a coarse segmentation probability graph; and extracting three-dimensional geometric feature parameters of the tumor candidate region according to the coarse segmentation probability graph. According to the method, the original image data is spatially aligned through the cross-modal registration algorithm, and the spatial consistency between different image sources is ensured, so that the image data under different modals can be accurately compared and analyzed, and an accurate spatial reference is provided for subsequent tumor region identification and processing.

The invention relates to a remote sensing water depth inversion system and method based on image space constraint and residual compensation, and belongs to the technical field of earth remote sensing detection and space remote sensing. Comprising the following steps: preprocessing satellite data, including active satellite data preprocessing and passive satellite data preprocessing; a BP neural network water depth inversion model is constructed, the preprocessed satellite data is used for training, and a preliminary water depth estimation value is output; establishing a residual compensation model and introducing spatial consistency constraint, and in the residual compensation process, introducing a spatial smooth penalty term by calculating the difference of water depth values of adjacent pixels to suppress local mutation; and carrying out optimization training on the model by using a joint loss function which is formed by weighted combination of mean square error and space consistency constraint loss, and outputting a final water depth inversion result. According to the method, an image space constraint loss term and a space smoothing penalty term in residual compensation are introduced, so that the consistency and the smoothness of an inversion result in space are ensured.

The invention relates to the technical field of image analysis, in particular to an automatic marking robot calibration method based on visual guidance, which comprises the following steps of: establishing an image set comprising different resolution levels based on operation site image data acquired by a marking robot, and performing corner detection and straight line segment detection on each level image in parallel. According to the method, through a parallel detection mechanism of a multi-resolution hierarchical image set, angular point and straight line segment features under different scales are synchronously extracted, and a multi-scale feature point set with high robustness is constructed in combination with cross-hierarchical coordinate stability measurement and response intensity quantification. And performing dynamic screening and grouping association on the feature points based on a preset geometric constraint condition, eliminating noise interference and false detection features, and generating a candidate calibration structure set with spatial consistency. Weighted contribution value fitting is adopted, cross-scale stability and detection confidence of feature points are integrated, and geometric accuracy and anti-interference capability of calibration reference point coordinates are improved.

The invention provides a semantic aerial view visual relocation method and device in a non-exposed scene, electronic equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring a multi-view image sequence under a non-exposed scene (such as a tunnel, an underground pipe gallery or an underground parking lot); semantic recognition is carried out based on a pre-trained semantic target detection model, and spatial consistency semantic features are extracted through a semantic-geometric dual-channel fusion mechanism combining a semantic mask and geometric constraints; the method comprises the following steps of: realizing three-dimensional reconstruction by using a voxel micro-renderable modeling method (VGGT), and generating a dense three-dimensional semantic point cloud fusing semantics and a geometric structure; two-dimensional semantics are mapped to a three-dimensional space through a projection and back projection relation, and point cloud semantics are endowed; main structure planes such as the ground, the left wall surface and the right wall surface are extracted, and a two-dimensional semantic aerial view with semantic annotation is generated; and pose estimation is carried out based on a reciprocal matching strategy guided by a semantic mask, so that visual repositioning with high precision, high robustness and semantic interpretability is realized. The method breaks through the problems of low precision, sparse features and poor semantic consistency of traditional visual repositioning in a non-exposed environment, and can be widely applied to the fields of intelligent transportation, underground inspection and unmanned system positioning.

The invention relates to a road disease high-precision positioning method, system and device based on multi-scale dynamic weighted curvature correction and a medium, and the method comprises the steps: constructing a road center line geometric object, obtaining a geographic coordinate, and constructing a multi-scale curvature field characterization parameter according to a spatial mapping relation between the geometric object and the geographic coordinate; establishing a correction model, calculating and generating an initial correction factor, and performing optimization to form a dynamic correction factor; the optimal projection position of the vehicle is determined according to the geographic coordinates and the road center line geometric object, the cumulative distance of the projection position is corrected based on the dynamic correction factor, and road pile number information is generated in combination with pile number reference parameters and constraints; the relative orientation is determined by comparing the distance between the vehicle position and the road end point or verifying the consistency of the driving track vector and the vector space in the tangential direction of the road, and the positioning result is output in combination with the road pile number information. According to the invention, a full-link precision guarantee system from data analysis to result output is formed, and a reliable positioning reference is provided for road maintenance.

The invention discloses an efficient image super-resolution reconstruction method and system based on degradation region guidance, and the method comprises the following steps: S1, carrying out the region-level degradation type recognition and severity quantification of an input low-resolution image, and generating a global degradation distribution map with spatial consistency; s2, according to the global degradation distribution map and in combination with semantic-texture collaborative features, repairing a region which is judged to be seriously degraded by adopting a high-capacity branch, and repairing a region which is judged to be slightly degraded by adopting a light-weight branch; s3, fusing the output of the high-capacity branch, the output of the lightweight branch and the global detail enhanced image to generate a final high-resolution image; wherein the global detail enhanced image is obtained by enhancing the semantic-texture collaborative features.

The invention relates to the technical field of mechanical arms, in particular to a man-machine interaction system of an intelligent mechanical arm with a body. The system decomposes a long time sequence task into sub-tasks through a large language model, and introduces a task acquisition module, a task queue management module, a task re-planning module and a control execution module, uses visual detection to identify gestures and environment events to trigger temporary tasks, and maintains interruptible marks and safety anchor points based on priorities and interruption risks. And during interruption, the mechanical arm with the timestamp and the environment state are collected, during recovery, the states are compared, local re-planning is carried out, transition sub-tasks are automatically generated for failure sub-tasks, and the mechanical arm is controlled to execute after constraint verification. According to the system, on the premise that the structure consistency, the space consistency and the time consistency are guaranteed, safe interruption and efficient recovery in a long-time-sequence task can be achieved, and the autonomy, the real-time performance and the operation safety of the mechanical arm in a man-machine cooperation scene are remarkably improved.

The invention discloses a tumor space-occupying brain network neural image alignment method based on multi-modal fusion, and belongs to the technical field of medical image processing and artificial intelligence crossing. The method comprises the following core steps of multi-modal image heterogeneous feature decoupling, tumor occupation deformation field modeling, functional network topological structure maintenance, cross-modal feature adversarial alignment, dynamic deformation constraint optimization and clinical interpretability verification, and construction of a three-dimensional non-rigid registration network based on a double attention mechanism. And differential homeomorphic mapping of a tumor focus area and normal brain tissue is realized through the cascaded spatial transformation module. Aiming at the problems of insufficient multi-modal feature alignment and brain network topology distortion in the prior art, the invention provides a function connection constrained cross-modal fusion strategy, a graph convolution network is adopted to encode resting state function connection features, and network node displacement caused by tumor occupation is dynamically corrected in combination with deformable convolution and a bidirectional feature competition mechanism; a space consistency loss function based on white matter fiber bundle tracing is designed, and through diffusion tensor imaging feature guide structure-function bimodal joint optimization, the problems of insufficient registration precision in a focus area and whole brain network connection distortion of a traditional method are solved. Experiments show that the registration precision of the method in glioma cases reaches 0.82 mm and is improved by 37% compared with that of a traditional method, and dissection-function consistency of functional network reconstruction around tumors is remarkably improved.

A reference point is arranged on the ground to calibrate a ground world coordinate system, a robot base coordinate system is established based on robot base features, and a rigid transformation matrix between the two is calculated to achieve spatial alignment. A multi-sensor network is constructed, a visual sensor and a laser radar are associated to a double-reference coordinate system, and data are synchronously collected and unified to the same coordinate system. And according to the spatial consistency error of the data, dynamically evaluating the credibility weight of the sensor, constructing a joint error equation, and based on least square method optimization, weighting fusion data and outputting calibration parameters. According to the invention, the problems of dependence on single sensor precision, lack of a system-level calibration scheme and complex calibration process are solved, and the calibration precision and efficiency of the robot system are improved.

The invention discloses a hybrid model driving-based space-time consistency regulation and control method for a central heating system, which comprises the following steps of: performing hydraulic equilibrium analysis on the central heating system, and performing hydraulic analysis on a space structure of a heating pipe network by using a graph theory method based on an energy conservation law, simulating the hydraulic working condition of the official website by using a basic loop analysis method; based on an energy conservation and mass conservation heat balance equation, a heat source and a heat user are indirectly connected through heat exchanger transition, a dynamic physical model of each subsystem is established based on a lumped heat capacity method, a refined model of the heat supply system is established, and the thermal dynamic characteristics of the system are researched and analyzed; based on a Pearson's correlation function, considering thermal inertia of equipment such as a pipe network of the central heating system and a building fence structure, and determining a regulation and control period and time of the regulation and control model; according to the method, by controlling the opening degree of the regulating valve of the heat exchange station, the suitability degree of the room temperature of a heat user is guaranteed.

The invention discloses a 2.5 D promptable medical image segmentation method and device based on SAM, and relates to the technical field of computer vision and medical image processing, and the method comprises the steps: carrying out the preprocessing of to-be-processed three-dimensional medical image data, and obtaining a plurality of 2.5 D data blocks; each 2.5 D data block serves as input, a trained SAM model is used for outputting a corresponding segmentation result, and a residual learning mechanism and a cross-slice attention mechanism are introduced into the SAM model. According to the method, the 2.5 D segmentation thought is introduced into the SAM, and by combining a residual learning mechanism and a cross-slice attention mechanism, the SAM model can better maintain the spatial consistency between the slices and fully capture the relevance between the slices, so that the expression ability of image features is improved, the expression of the model on a complex structure image is enhanced, and the image quality is improved. Therefore, the accuracy of the medical image segmentation result can be improved.

The invention discloses a BIM (Building Information Modeling)-based building component attribute matching method and a BIM-based building component attribute matching system, contour line segments of a target and a reference component are segmented, curvature change values are recorded, and the local feature expression capability is enhanced. Then, dynamic adjustment parameters are introduced to optimize eigenvector weighting processing, so that the matching process adapts to different shape complexity, and mismatching is avoided; and then, identifying a local similar region by comparing the corrected matching parameter set, and generating a registration mapping relation, thereby improving the accuracy of space consistency judgment. A matching priority matrix is constructed based on the mapping relation and the shape distribution difference, high-matching-degree candidate components are quickly screened, and the matching efficiency is improved. Finally, a feedback adjustment mechanism is adopted in the attribute verification stage, it is ensured that the matching result meets the preset precision requirement, and self-adaptive optimization is achieved. The method effectively solves the problems of low precision and poor stability in non-standard geometric component matching in the prior art, and has important practical value.

The invention discloses a power transmission tower deformation monitoring method and device and a computer program product, and the method comprises the steps: obtaining a multi-angle collection image of a random speckle pattern on the surface of a power transmission tower; constructing an image sequence based on the multi-angle acquired images, reconstructing a three-dimensional displacement field on the surface of the power transmission tower by using a stereoscopic vision reconstruction algorithm, and exporting a strain tensor field; performing time sequence processing on the strain tensor field under a plurality of time nodes through a strain evolution analysis model of a continuous time sequence, and extracting a strain change trend of each preset structure part; carrying out space consistency mapping and weighted interpolation processing according to the strain change trends of the plurality of preset structure parts, and calculating to obtain the overall structural dependent variable of the power transmission tower; and S5, in combination with historical strain data and the current overall structural dependent variable, identifying an abnormal deformation area of the power transmission tower by comparing the spatio-temporal evolution mode of the strain tensor field. The method has the effect of improving the accuracy of deformation monitoring of the power transmission tower.

The invention discloses an intelligent building fault prediction method based on an adaptive algorithm, and the method comprises the steps: obtaining operation state data collected by a plurality of sensor nodes, carrying out the normalization preprocessing, inputting an improved adaptive prediction model, and obtaining the implicit state representation; an initial fuzzy rule set is generated through fuzzy rule induction, a real-time evolution strategy is introduced, fuzzy rules and neural connection weights are dynamically adjusted according to error feedback, and a self-adaptive prediction model updated through evolution is formed; and then performing multi-step prediction on the key operation parameter time sequence by using the updated model, comparing a prediction result with a dynamic fault threshold value to generate a candidate fault indication, and determining the position and category of a fault to be pre-warned in the building system through time sequence stability verification and space consistency analysis. The method can significantly improve the accuracy and real-time performance of intelligent building fault prediction, and has a good application prospect.

The invention relates to the technical field of channel environment monitoring, and discloses a channel environment safety monitoring and early warning method and system based on multi-sensor fusion, which align visible light video, infrared thermal imaging, gas and vibration sensor data through space-time reference to ensure isomerous data synchronism and space consistency. Modal exclusive time sequence feature extraction is carried out, and unique space-time mode information is mined. A context sensing mechanism is introduced, key information such as an operation and maintenance plan is intelligently extracted in combination with a current environment state, dynamic weighting and deep fusion are performed on multi-sensor time sequence features through an attention fusion mechanism, and the importance of different sensor features is adaptively adjusted. And finally, inputting the fused channel environment state into a feedforward neural network model, and outputting an event probability and a risk score. Therefore, safety monitoring accuracy and reliability are remarkably improved, false alarms are reduced, potential safety hazards are recognized in advance, quantitative decision basis is provided, and continuous and safe operation of a channel environment is guaranteed.

The invention relates to an unstructured environment-oriented heterogeneous data fusion sensing system for an agent with a body, which can be applied to the technical field of intelligent control. The system comprises a sensing module and a control module, the sensing module is used for acquiring multi-dimensional physical attribute data, three-dimensional space form data and dynamic mechanical response data of the intelligent body in real time; the control module is used for performing timestamp synchronous marking on the multi-dimensional physical attribute data, the three-dimensional space form data and the dynamic mechanical response data, and realizing spatial consistency mapping of the multi-dimensional physical attribute data, the three-dimensional space form data and the dynamic mechanical response data through coordinate system conversion to obtain heterogeneous data; performing feature matching on the heterogeneous data to generate a fusion perception vector; based on the fusion perception vector, generating an autonomous decision instruction corresponding to the body agent; and controlling the body agent to execute a target action corresponding to the autonomous decision instruction. By adopting the system, the control accuracy of the intelligent body can be improved.

The invention discloses a natural language and machine vision-based alignment method, which comprises the following steps of: providing a three-level alignment architecture, and respectively extracting local-global features of a visual scene and grammar-semantic features of a natural language by adopting a double-flow feature extraction network; through a space-time attention enhanced cross-modal alignment module, a dynamic gating mechanism is adopted to complete feature space adaptive projection; a joint optimization strategy is constructed based on comparative learning, and vision and language embedding space consistency is optimized by using a multi-granularity comparative loss function. And meanwhile, semantic topology constraint and visual causal reasoning are introduced, so that the calculation complexity is reduced, and the task robustness is improved. Through experimental test data, the Top-1 accuracy rate of the method in image-text retrieval reaches 92.3%, the visual question and answer F1 value is 83.4%, the model parameter quantity is reduced by 40%, the method can be widely applied to the fields of intelligent interaction systems, automatic driving scene understanding, industrial quality inspection knowledge base construction and the like, and the semantic perception and reasoning ability of a multi-modal system is remarkably improved.

The invention relates to the technical field of nondestructive testing, in particular to an automatic defect detection and segmentation technology based on eddy current pulse thermal imaging, which comprises a thermogram sequence preprocessing step, a defect signal feature extraction step and an image segmentation post-processing step, in the thermogram sequence preprocessing step, an excitation peak frame image is dynamically selected through an image entropy difference, a static background is inhibited by combining image difference operation, and a geometric coil mask is generated by utilizing edge detection; and a defect signal feature extraction step: carrying out defect matrix reconstruction on the robust principal component analysis model based on the difference image by adopting an augmented Lagrange multiplier method. And an image segmentation post-processing step: designing a dual-threshold segmentation method based on reconstruction matrix local space consistency, and screening thresholds by combining an adaptive threshold of image gray level distribution and an area screening threshold of a defect candidate region. Automatic detection and segmentation of defects in eddy current pulse thermal imaging are achieved, and the purposes of defect identification and quantitative detection are achieved.

The invention relates to the technical field of space-time prediction, in particular to a brucellosis space-time prediction method and system based on artificial intelligence, and the method comprises the following steps: obtaining case information, calculating an incidence relation, extracting a path sequence, screening a trend direction, and matching a new case to generate prediction data. According to the method, continuous identification of a propagation path is realized by constructing a propagation incidence relation between cases and introducing a spatial directivity index, a non-trend propagation process is screened out through an angle average value and a variance, spatial consistency of path screening is enhanced, and through joint matching of spatio-temporal characteristics of newly-added cases and an existing propagation trend, the propagation path screening efficiency is improved. The sensitivity of prediction to the trend attribution of a new case is improved, the spatial directivity of a potential disease area is enhanced through reverse projection of a trend path and area positioning drop point frequency analysis, and through linkage processing of multi-level path extraction, trend judgment and area coding, the probability of occurrence of the new case is lowered. And the capturing capability of the prediction data on the propagation and evolution characteristics of the brucellosis is improved.

The invention discloses a defect segmentation loss evaluation method based on space consistency optimization, and relates to the field of computer vision and defect segmentation. The problem that an existing loss calculation method is difficult to give consideration to defect space consistency and fair modeling of defects of different scales, and consequently loss evaluation accuracy is poor is solved. According to the method, a prediction result and a defect area in a real label are subjected to instantiation modeling, a prediction and real defect set is constructed, and one-to-one optimal matching of the prediction result and the real defect set is realized through a Hungary matching algorithm; in order to enhance the modeling capability for defects of different sizes, a Wasserstein distance is introduced to measure the spatial similarity of matched defect pairs, defect segmentation loss is calculated for successfully matched defect pairs based on the distance, a fixed loss value is directly applied to predicted defects which are not successfully matched, and the sum of the defect segmentation loss and the fixed loss value is used as a total loss value of a sample. The method is mainly applied to an industrial product surface defect segmentation task.

The invention provides a 3D small sample segmentation method and system based on registration alignment and interlayer consistency. The method comprises the following steps: S1, supporting memory extraction; and S2, carrying out sequential segmentation. According to the invention, a 3D medical image is regarded as a series of 2D slice sequences, and sequential processing is carried out by fully using the time sequence modeling capability of SAM2. In this way, the FSMIS-SAM2 can maintain good space consistency between the slices, and the segmentation precision is remarkably improved. The core technology of the FSMI-SAM2 framework is to process a 3D medical image sequence by using a memory enhancement converter of the SAM2. In addition, the invention further provides an innovative inference initial position estimation method, the most appropriate initial slice can be determined for inference through accurate registration according to changes of the anatomical structure of a patient, it is ensured that a high-confidence-coefficient segmentation result can be obtained in the initial stage of sequence inference, and high-quality initialization is provided for subsequent slice segmentation.