21 results about "Three dimensional morphology" patented technology

The invention discloses a cell fate distinguishing and differentiation prediction method based on a three-dimensional form, and belongs to the technical field of cell fate distinguishing and differentiation prediction. In order to solve the problem that cell fate differentiation is difficult to distinguish and predict under a small sample size, the method comprises the following steps: acquiring a living biological sample, and imaging by using a three-dimensional delay microscope to obtain cell three-dimensional image data; segmenting the cell membrane boundary and extracting geometric and high-order three-dimensional morphological characteristic parameters; labeling cell fate and differentiation states based on means such as cell lineage tracking or gene fluorescence labeling; and finally, determining an optimal morphological index and a minimum sample size through single-side double-sample t test, and applying the optimal morphological index and the minimum sample size to two new cell populations and outputting a cell fate distinguishing and differentiation prediction result. The method can realize cross-species, low-loss, low-cost and high-precision cell fate differentiation and differentiation prediction, and is suitable for development research, stem cell culture and disease diagnosis.

The invention provides a tailing particle classification method based on multi-dimensional morphological feature compression and clustering, and belongs to the technical field of geotechnical engineering, mineral processing and particle classification, and the method comprises the following steps: S1-S6, obtaining an STL file database of tailing particles; s7, calculating and extracting morphological parameters to form an original parameter data set; s8-S10: obtaining a three-dimensional morphological parameter set which retains the original morphological feature information; s11-S12, determining an optimal clustering number by adopting an elbow rule, a contour coefficient and Gap statistics, and performing particle morphology clustering grouping on the three-dimensional morphology parameters by adopting a K-means algorithm; and S13, carrying out stacking test simulation on the clustered particle categories by using a discrete element method, analyzing the stacking density and the average coordination number of different categories of particles, and verifying the physical significance of a classification result. The tailing particle classification method can systematically and efficiently integrate particle multi-dimensional form information, can directly associate the information with macroscopic physical and mechanical properties, and is data-driven, clear in mechanism and explainable in result.

The application discloses an XCT image-based CFRP defect automatic identification and classification method and system, and belongs to the field of carbon fiber reinforced composite material manufacturing and processing quality detection.The application obtains the three-dimensional features of defects through binarization three-dimensional label statistics classification, obtains the two-dimensional features of defects through manual labeling based on the three-dimensional label of defects, and after fusing the three-dimensional features and the two-dimensional features of defects, combines the features with a deep learning method to obtain an efficient real defect identification and classification model, which is simple and efficient and effectively improves the CFRP defect identification precision.The application uses the three-dimensional morphological parameters of defects when labeling the data set, and avoids the unreliability introduced by directly labeling on two-dimensional slices.The application trains a semantic segmentation model for identifying and classifying CFRP defects based on a small amount of two-dimensional CT images, and the semantic segmentation model is used for predicting internal defects of similar materials, so that a large amount of tedious and time-consuming manual image processing for obtaining internal defect data of materials is avoided, and a large amount of labor cost is saved.

The invention relates to the technical field of biological information processing and automatic sorting, and discloses a chick gender automatic identification method and system based on multi-modal fusion, and the method comprises the steps: collecting original three-dimensional point cloud data through a three-dimensional form sensing module, and calculating the key region coordinates of chicks; based on the coordinates, controlling the addressable micro sampling array to execute active sampling adaptation, and obtaining measurement response time sequence data of the target volatile matter; acquiring background response time sequence data by utilizing a background channel sensor array; and performing differential processing on the two to generate a differential biochemical response time sequence vector. And then, respectively extracting morphological feature vectors and biochemical fingerprint feature vectors, inputting the morphological feature vectors and the biochemical fingerprint feature vectors into respective independent classifiers to obtain prediction categories and confidence coefficients, finally, judging a bimodal prediction result through preset cross validation gating fusion logic, outputting a final decision, and controlling a sorting execution module to complete physical sorting. And the accuracy of biochemical sampling is improved through active sampling adaptation.

This invention provides a tailings particle classification method based on multi-dimensional morphological feature compression and clustering, belonging to the fields of geotechnical engineering, mineral processing, and particle classification technology. The method includes: S1-S6: acquiring an STL file database of tailings particles; S7: calculating and extracting morphological parameters to form an original parameter dataset; S8-S10: obtaining a three-dimensional morphological parameter set retaining the original morphological feature information; S11-S12: determining the optimal number of clusters using the elbow rule, silhouette coefficient, and Gap statistic, and using the K-means algorithm to cluster the three-dimensional morphological parameters into particle morphology groups; S13: using the discrete element method to conduct a stacking test simulation on the clustered particle categories, analyzing the stacking density and average coordination number of different particle categories, and verifying the physical meaning of the classification results. This tailings particle classification method can systematically and efficiently integrate multi-dimensional particle morphological information and directly correlate this information with macroscopic physical and mechanical properties. It is data-driven, has a clear mechanism, and the results are interpretable.

The invention relates to the technical field of smart animal husbandry and computer vision, and discloses a live pig weight estimation method based on semantic key point guidance and three-dimensional morphological feature fusion, which comprises the following steps: step 1, joint perception and semantic decoupling; step 2, space mapping and point cloud cleaning; step 3, carrying out self-adaptive attitude normalization; 4, orthogonally intercepting the rigid trunk; step 5, deep extraction of morphological characteristics; and step 6, intelligent regression prediction. According to the invention, a common fixed-proportion excision method or a neck positioning method depending on geometric depression in the prior art is abandoned; by introducing an orthogonal truncation mechanism guided by a semantic anchor point (ear root), a virtual'adaptive rigid collar 'is constructed on a front-end truncation plane, and the truncation boundary can be dynamically adjusted along with swinging of the head of a pig and is always perpendicular to the spine axis. According to the mechanism, interference of non-rigid deformation such as large-amplitude head torsion and head lowering of the pig on trunk volume calculation is eliminated in principle, and the stability of effective trunk point cloud is ensured.

The invention provides a detection and rating method for non-metallic inclusions, which comprises the following steps: scanning a sample by adopting an X-ray tomography system to obtain three-dimensional gray data, and marking the non-metallic inclusions; obtaining three-dimensional morphological parameters and spatial position information of the non-metallic inclusions; counting the obtained three-dimensional morphological parameters and spatial positions to obtain the volume fraction, size distribution concentration ratio, spatial distribution uniformity and shape distribution data of the inclusions; calculating the data by adopting an evaluation model to obtain an inclusion evaluation index, and rating the nonmetallic inclusions of the steel grade on the basis of the inclusion evaluation index. According to the method, the problems of contour distortion and form misjudgment of a traditional method are solved, the objectivity of inclusion rating is effectively improved, an IEI rating result is directly output, a visual basis can be provided for steel quality control and performance pre-judgment, and industrial application is supported.

The invention relates to the technical field of erosion gully parameter conversion, in particular to an erosion gully multi-dimensional morphological parameter conversion method and related equipment. And constructing a multi-dimensional factor database comprising ditch shape parameters, terrain, rainfall, vegetation, soil, land utilization, lithology and management factors. And adopting a recursive feature elimination algorithm to screen main control factors influencing the three-dimensional form, inputting a screening result into a pre-trained prediction model, and outputting an elevation value of the cross section of the erosion gully. And finally, generating a three-dimensional form spatial distribution result of the erosion gully in the target area by combining the elevation value and a factor database. According to the method, accurate prediction from two-dimensional data to three-dimensional form is realized, and a scientific basis is provided for erosion gully treatment and water and soil conservation.

The application discloses a kind of stem filament three-dimensional morphology characterization and filamentous degree evaluation method and system, belong to tobacco processing technical field.The method includes: obtaining stem filament three-dimensional point cloud data, after pre-processing, using cascade segmentation strategy, combined with normal vector distribution entropy criterion and improved watershed algorithm realizes the accurate separation of adherent stem filament;To single stem filament point cloud cluster extraction three-dimensional space skeleton line;Along the skeleton line method plane is analyzed to cross section projection point set, calculates real three-dimensional length, average width and thickness, bending degree, cross section uniformity index and structural integrity coefficient and other basic morphological parameters;On this basis, construct the comprehensive filamentous degree index of fusion multi-attribute penalty function, and calculate thickness unevenness index, structural defect suspicion index and other derivative parameters to assist process diagnosis.The application realizes the objective quantitative evaluation of stem filament "thin, long, straight, even, integral" quality, solves the problem of traditional two-dimensional measurement distortion and subjective evaluation misalignment.

The invention provides a lingual surface three-dimensional shape reconstruction method and system based on deep learning, and relates to the technical field of three-dimensional reconstruction, and the method comprises the steps: activating an image collection array based on a light source illumination parameter to carry out the multi-angle two-dimensional sensing of a target tongue part, and obtaining an original lingual surface image sequence; performing visual cutting, extracting a plurality of lingual surface target images, performing deep learning, and constructing a lingual surface depth map; performing three-dimensional shape analysis, constructing lingual surface three-dimensional point cloud coordinates, mapping the coordinates to a plurality of lingual surface target images for identification, and performing comparative analysis on an identification result and a standard image library to generate a comparative parameter set; and performing correction, performing three-dimensional reconstruction on the target tongue according to a correction result, and constructing a three-dimensional morphological model. The technical problems that a lingual surface analysis method in the prior art mostly depends on two-dimensional images or manual observation, subjectivity is often high, information loss is serious, and the real three-dimensional form of the lingual surface is difficult to obtain are solved.

The application discloses a kind of based on morphological extraction point cloud local surface feature method, steps are as follows: S1 acquires point cloud data and carries out data pretreatment to point cloud data;S2 searches the neighborhood point of each point of point cloud in three-dimensional space, calculates normal information, then based on normal information, the curvature of each point is calculated, and curvature value is obtained;Then refer to the curvature value range corresponding to local surface feature and filter out target surface;S3 uses progressive morphological filter to adjust filter parameter constantly, according to the curvature change of selected local area, the size of filter window is adjusted accordingly, non-target point in point cloud data after curvature screening is filtered out, and the point of target local area is extracted;S4 uses dilatation, corrosion three-dimensional morphological operator and dilatation corrosion combination open operation and close operation to optimize target local area, and extract the local surface feature area of complete smooth point cloud data.The method is extracted to local area with obvious curvature change, high accuracy, and wide adaptability.

This invention relates to a three-dimensional defect segmentation method, system, and storage medium, belonging to the interdisciplinary fields of computer vision, deep learning, and nondestructive testing. The three-dimensional defect segmentation result is obtained by inputting the industrial CT volume data of the object to be inspected into a 2.5D segmentation network composed of an encoder, a spatial feature displacement module, and a decoder. This invention achieves efficient three-dimensional context awareness, significantly improving defect segmentation accuracy and three-dimensional morphological integrity. The segmentation result has topological connectivity and smooth boundaries, and its computational complexity and memory usage are far lower than those of a full 3D network. It can meet the real-time requirements of industrial online inspection and has broad application prospects in high-end manufacturing.

This invention relates to a three-dimensional defect segmentation method, system, and storage medium, belonging to the interdisciplinary fields of computer vision, deep learning, and nondestructive testing. The three-dimensional defect segmentation result is obtained by inputting the industrial CT volume data of the object to be inspected into a 2.5D segmentation network composed of an encoder, a spatial feature displacement module, and a decoder. This invention achieves efficient three-dimensional context awareness, significantly improving defect segmentation accuracy and three-dimensional morphological integrity. The segmentation result has topological connectivity and smooth boundaries, and its computational complexity and memory usage are far lower than those of a full 3D network. It can meet the real-time requirements of industrial online inspection and has broad application prospects in high-end manufacturing.

The invention relates to the technical field of image recognition, in particular to an aggregate three-dimensional morphological feature analysis method and system based on laser scanning, and the method comprises the steps of data intelligent collection, three-dimensional point cloud data intelligent preprocessing, morphological feature parameter calculation, intelligent analysis and association and application. In the prior art, morphological analysis, virtual experiment and performance prediction of aggregate are generally regarded as relatively independent links, or only forward prediction from process, material to performance can be carried out, and closed-loop analysis cannot be formed; according to the method, on the basis of analyzing the aggregate accumulation characteristics through a virtual compaction experiment, a double-prediction model architecture is introduced, the raw material ratio is reversely deduced from morphological characteristics and the process through the first prediction model, and the process parameters are inversed through the second prediction model in combination with morphological and raw material information; and finally, inputting all the features into a performance prediction model for comprehensive performance prediction. And unprecedented deep analysis capability is provided for aggregate quality diagnosis, production process optimization and performance customization.

This invention relates to the field of cloud-edge collaborative computing technology, specifically to a method and system for processing integrated air-space-ground metal mineral exploration data. The method includes the following steps: extracting node resistance anisotropy features to generate corrected topological weights and constructing a Laplace tensor matrix to output a reduced-rank profile matrix; establishing a high-order tensor for cross-domain routing record flow by combining the profile spatial gradient and the network node signal-to-noise ratio; constructing an anisotropic correction matrix based on local covariance matrix analysis to extract three-dimensional normal vectors; solving the implicit scalar field and extracting isosurfaces to finally complete three-dimensional reconstruction. In this invention, by analyzing the reduced-rank structural features of the multi-dimensional exploration network and integrating node dynamic attributes to determine link steady-state scores to guide the priority cross-domain flow of compressed tensors, and by relying on the proportion of spatial local distortion to correct the grid gradient and reshape the implicit surface constraint field, this effectively avoids long-distance congestion transmission of massive heterogeneous data and improves the accuracy of three-dimensional morphological reproduction of complex underground ore bodies.

This invention provides a digital intelligent MSE automatic design method based on biomechanical simulation, belonging to the field of biomechanical simulation technology. The method includes: acquiring three-dimensional morphological data and physiological load data of the target anatomical structure and mapping them to obtain a load distribution matrix; performing tensor operations on the load distribution matrix and tissue mechanical response characteristics to obtain a stress prediction field; identifying high-stress concentration areas and low-stiffness weak areas based on the stress prediction field; extracting contour curves and calculating the normal gradient to determine the expansion constraint direction; minimizing the stress distribution variance and the total tissue deformation energy and determining the spatial configuration parameters of the MSE device; using the spatial configuration parameters as boundary conditions to iteratively calculate the displacement field and strain field; correcting the spatial configuration parameters based on the difference between the cumulative deformation and a preset physiological threshold and the iterative deformation energy change rate; and outputting the corrected configuration parameters as design instructions when the convergence condition is met.

The invention discloses an urban texture fracture area identification method based on a three-dimensional form percolation model, and the method comprises the following steps: recognizing building data, sunshine shadow data, mountain water, river and lake data and block range data in aerial photography data through the collected aerial photography data; processing the building data, the sunshine shadow data, the mountain water, river and lake data and the block range data into vector data under the same coordinate system; extracting three-dimensional morphological attributes of the building; a seepage association network based on building buffer area overlapping is constructed, and texture units are dynamically extracted by setting different seepage threshold values; and finally, carrying out clustering analysis on the texture units, and accurately identifying an urban texture fracture region according to an adjacent proportion and a base area condition. According to the method, the automatic and quantitative identification of the urban space structure is realized, the limitation that the traditional method depends on manual interpretation and is difficult to capture the continuity of the three-dimensional form is effectively overcome, and a scientific basis can be provided for the repair and connection work of the urban texture in urban updating and planning.

The invention discloses a microalloyed high-strength steel production system, and relates to the technical field of steel production supervision. A real-time three-dimensional morphological field is established in a data sensing network according to data acquired by a thermal sensor array and a laser imaging device, a real-time equipment parameter set is generated according to the operation process of production equipment, a historical three-dimensional morphological field is established, the historical three-dimensional morphological field is divided into short-term process three-dimensional morphological fields, and the short-term process three-dimensional morphological fields are obtained. The short-term process three-dimensional morphological field is divided into detection cores, the short-term process three-dimensional morphological fields associated with the same production raw material factors are mutually overlapped and mapped, dynamic reference clouds and dynamic reference cloud ridge lines are set for the detection cores, a standard three-dimensional morphological field is obtained, and the real-time three-dimensional morphological field is divided into verification cores. And deploying the short-term process three-dimensional form field from the standard three-dimensional form field through the parameter matching factor to match with the verification kernel, and further adjusting the production equipment corresponding to the parameter matching factor by setting a ripple benefit regulation and control mechanism.

The invention provides a rotor three-dimensional morphological filtering method based on unequal interval sampling, and belongs to the technical field of rotor measurement. In order to solve the problems that traditional three-dimensional morphological filtering can only process data in one direction once, and data fusion easily causes model error expansion, the method comprises the following steps: carrying out identification analysis on a geometric structure of a rotor; the number of target points and adjacent sampling points is set for area division, and the weight of the sampling points of each area is distributed; carrying out multi-scale unequal interval sampling on the rotor; carrying out superposition processing on the sampled data, and fitting a model; performing shape detection on the surface of the rotor; constructing a multi-scale morphological probe according to the structural characteristics of each region of the shaft diameter; constructing a morphological pyramid of the whole rotor, fusing and processing the collected data, and completing axial-radial bidirectional filtering at one time; and sequentially performing morphological operation on a subsequent layer of the morphological pyramid until all data are processed by the multi-scale morphological probe to obtain a rotor three-dimensional model.

This invention discloses a method and system for quantitative identification of pipeline blockage based on three-dimensional morphology and dynamic response, belonging to the field of intelligent pipeline detection technology. This invention simultaneously acquires three-dimensional point cloud data of pipeline blockages and micro-vibration dynamic force response signal sequences. It calculates the three-dimensional occupancy gradient (3DOG), characterizing the radial spatial distribution density change of the blockage, and the dynamic response entropy (DRE), quantifying the complexity of the material's response energy distribution. A pre-set nonlinear fusion model yields the Pipeline Blockage Comprehensive Index (PCCI). Based on the PCCI threshold range, the blockage severity is graded, and a clearing strategy is automatically matched. This invention is the first to achieve simultaneous quantification of the spatial morphology and material properties of blockages in two dimensions, establishing a complete intelligent evaluation chain. It effectively solves the problems of traditional detection relying on manual experience, having a single evaluation dimension, and being unable to quantify, significantly improving the intelligence level of blockage detection in high-speed rail tunnels and various pipelines, and enhancing the scientific nature of clearing decisions.

The invention provides an artificial skin flap customization design method fusing mirror image optimization and biomechanical analysis. The method comprises the following steps: S1, carrying out dual-region multi-modal data acquisition; s2, fusion generation of a defect area three-dimensional model; s3, performing healthy area mirroring and aesthetic optimization design; s4, designing a bionic blood vessel channel network; and S5, outputting and manufacturing. According to the method, the epidermal form and the subcutaneous fat thickness are integrally and accurately reconstructed, so that the designed skin flap is close to a natural tissue in the aspects of surface profile and internal structure, and the tissue for transplantation is perfectly matched with a recipient defect in the aspects of three-dimensional form and volume.