25 results about "Skeletonization" patented technology

The invention mainly relates to the field of digital image photogrammetry. In order to solve the problems that road crack detection lacks end-to-end quantitative analysis capability and crack extraction results are fragmented, the invention provides a road crack extraction and grading method based on topological structure constraints, and the method comprises the following steps: crack end point extraction and width calculation, skeleton of a crack center line, identification of a connected region and end points, and classification of the end-to-end quantitative analysis capability of the end-to-end quantitative analysis capability of the end-to-end quantitative analysis capability of the end-to-end quantitative analysis capability. The crack width corresponding to each center point is calculated through Euclidean distance transformation; potential breakpoint pairs are screened according to attribution, distance and direction constraints, then a topological connection rule is constructed, self-adaptive repair of the fracture is achieved, and the geometric integrity of the fracture is recovered; and performing quantitative analysis on the repaired cracks, accurately calculating the length, the average width, the maximum width, the area and the geographic coordinates of each crack, and performing danger grade division on each crack according to a preset threshold value to form structured disease data and a grading result which can be directly used for road risk assessment.

This application provides a method and system for modeling multi-class lung airway tree models based on shape-aware sampling. The method includes: acquiring three-dimensional medical image data containing multiple types of tracheal anatomical annotations and determining a training set; determining sampling patches for the images in the training set based on a fractal dimension-driven sampling patch adjustment strategy; determining a skeleton-weighted loss term based on the training set using a shortest path cost skeletonization method; determining a total loss function based on the skeleton-weighted loss term and volume overlap loss; training a pre-defined neural network for multi-class lung airway tree modeling based on the sampling patches and total loss function of the images in the training set; and modeling the three-dimensional medical image of the lung to be modeled using the trained multi-class lung airway tree modeling model. This application effectively improves the volume overlap accuracy and topological consistency of multi-class tracheal model reconstruction, alleviating branch breakage and topological errors in discrete voxel space.

This invention discloses a method, apparatus, and medium for spatial coordinate unfolding and reconstruction, comprising: extracting spatial expression points labeled as target tissue types from spatial group data to generate a standardized binary image; performing a closing operation and smoothing edges, extracting multiple contour regions and merging them into a main contour, and performing an image skeletonization algorithm on the main contour to generate a skeleton path map; selecting the longest path from the shortest paths as the central main path; backprojecting the central main path in the image coordinates to the original spatial coordinate system to determine the central main path and its starting and ending points; constructing a tangent-normal coordinate system and calculating the distance from the spatial expression points to a first reference point; generating edge reference lines; projecting all spatial expression points to the unfolded coordinate system to determine the projection target; and projecting according to the projection target to obtain the unfolded coordinates. This invention automatically generates complete skeleton paths and highly smooth edge curves, effectively restoring the true spatial positions of the host and / or microorganisms and spatial expression points.

The method for three-dimensional branch reconstruction of tree point cloud based on mixed skeletonization and branch level simulation belongs to the field of forestry science research, computer vision technology and tree three-dimensional modeling technology. The method supports taking leafless single tree point cloud or leafy single tree point cloud as input, selecting the optimal weight file of the predicted point movement of the sparse convolutional neural network model trained on the leafless or leafy synthetic tree dataset, and applying it to the point cloud contraction process based on sparse convolution. The mixed skeletonization of the preprocessed input point cloud is obtained through the point cloud contraction and fast hierarchical clustering, and the initial skeleton of the single tree and the initial radius corresponding to each skeleton point are obtained. The two-stage skeleton refinement generates a single tree quantitative structure model, and the updated single tree quantitative structure model is obtained by adjusting the branch transition and reconstructing the branch arterial snake model. The model participates in parameter extraction together with the input point cloud and is attached to the single tree quantitative structure model file and output.

This application provides a medical image processing method and system. The method includes: acquiring vascular image volume data from photon-counting computed tomography scans of a target region; based on the vascular image volume data, performing skeletonization processing on the vascular network, extracting multiple skeleton points in the vascular network, and forming a vascular centerline network according to the connection relationships between the skeleton points; identifying endpoints and bifurcation points from the vascular centerline network, thereby dividing the vascular centerline network into several continuous and unbifuzzy vascular segments, and constructing a vascular centerline diagram structure based on each vascular segment; establishing and solving a blood flow analysis model based on the vascular centerline diagram structure to obtain the blood flow analysis results of the vascular network; wherein, the blood flow analysis results include the blood flow rate and / or pressure difference between the two ends of each vascular segment. This application improves the accuracy of the blood flow analysis results.

The application provides a mountainous area night animal tracking method and system based on multispectral and thermal imaging data fusion, and relates to the technical field of night animal tracking, and comprises the following steps: synchronously acquiring a multispectral image sequence and a thermal imaging image sequence in a mountainous area at night; performing multiscale Retinex enhancement processing on a registered near-infrared image and a thermal image respectively to obtain a strengthened candidate target image; extracting a multi-source texture signature of a fusion local binary pattern feature vector and a thermal gradient direction histogram for each candidate target, and combining geometric features obtained through morphological skeletonization to construct a multi-dimensional feature set; realizing target correlation between adjacent frames by using a nearest neighbor greedy matching algorithm to form a preliminary trajectory segment; and finally eliminating abnormal trajectories, solving trajectory intersection and occlusion conflicts through a sparse dynamic time warping algorithm, and generating an accurate and complete animal moving trajectory. The application can solve the conflicts caused by trajectory intersection and occlusion, and output continuous and accurate single animal moving trajectories.

The application discloses a skeleton quality driven tubular structure segmentation closed loop optimization method and system, which is applied to the technical field of biomedical image processing and computer aided diagnosis, and the method comprises the following steps: using a trained segmentation model to infer medical volume data to obtain a segmentation mask; skeletonizing the segmentation mask to extract a skeleton graph; performing topological defect detection on the skeleton graph to obtain a structured defect report; reversely mapping three-dimensional coordinates in the report back to a voxel space, taking each defect point as a center, a preset radius as a range, and generating a defect density weight graph according to a severity score; and using the weight graph as a spatial weighting parameter of a loss function to optimize the segmentation model; and the application reversely maps the topological defects detected by skeletonization into a weight graph and integrates the weight graph into a loss function for closed loop iteration, realizes directional repair on high-occurrence areas such as fractures and false branches, and improves the conduction efficiency of segmentation improvement to skeleton quality.

The application discloses a tunnel face joint information calculation method based on YOLOv8-seg, which comprises the following steps: a face joint data set is established, and data in the data set is labeled; the labeled joint data set is divided into a training set, a test set and a verification set, a YOLOv8-seg model is optimized, the joint data set is input into the optimized model for training, and a joint photo with a mask marked area is output after the training is completed; the joint information of the mask area is further processed according to the photo output by the model, the pixel height of the joint is calculated, and skeletonization is performed to extract to a single-pixel width; and finally, the actual height and length information of the joint are calculated according to the processing result of the previous step.

The invention discloses a structure surface crack detection and quantitative analysis system based on deep learning, and relates to the crossing field of computer vision, deep learning and structure health monitoring. The system adopts a multi-stage cascade architecture. The method comprises the following steps: firstly, realizing crack intelligent identification through an improved OfficientNet-B0 model (integrating an ECA attention mechanism and an optimized MBConv structure); then, precise segmentation is completed based on a Mask-RCNN architecture; and finally, combining morphological processing and a camera calibration technology to realize size quantification. The core innovation points comprise: (1) a lightweight identification model of a double attention mechanism; (2) segmentation optimization of CLAHE enhancement and morphological refinement; and (3) a sub-millimeter quantization method based on skeletonization. Experimental results show that the recognition accuracy rate reaches 99.0%, the segmentation IoU is about 0.623, and the width quantization error lt is achieved; the thickness is 0.1 mm.

The application discloses a forearm shallow blood vessel segmentation method and device based on an NSVA-Net deep learning network, and relates to the technical field of medical image processing. The method comprises the following steps: collecting near-infrared images of the forearm shallow blood vessels of a subject by using a near-infrared light source and a near-infrared camera; performing denoising, contrast enhancement, cutting of the forearm region and normalization on the near-infrared images to obtain preprocessed images; performing two-stage image enhancement of RCAE and CLAHE on the preprocessed images to obtain enhanced images; inputting the enhanced images into a pre-trained NSVA-Net deep learning network to obtain a segmentation probability map of the forearm shallow blood vessels; performing threshold segmentation, connected domain analysis and skeletonization processing on the segmentation probability map to remove artifacts and isolated noise, and obtaining a blood vessel segmentation result and a center line; and superimposing and displaying the blood vessel segmentation result and the center line on the near-infrared images to output a final segmentation image. By using the application, the segmentation recall rate and the structural continuity of the forearm shallow small blood vessels of dialysis patients can be significantly improved.

The invention provides a self-adaptive palmprint multi-dimensional quantitative analysis method and system fused with anatomical features, two technical paths of anatomical guidance and deep learning segmentation are provided, accurate derivation can be performed by using medical priori knowledge, powerful identification can be realized by using a deep learning model, the two paths can be used independently or in combination, and the method and the system have good application prospects. The applicability of the technology is widened; key parameters of Canny and Hough algorithms are dynamically linked with image resolution and ROI (Region of Interest) size, so that the detection stability under images with different qualities and sizes is ensured; the palmprint ROI dynamically generated by the anatomy datum point is used for checking and restraining the output of the palmprint segmentation model, errors which possibly occur in the AI model and do not conform to the physiological common sense are effectively solved, and the recognition result is more reliable and accurate; through the techniques of skeletonization, topology analysis and the like, accurate quantification of features such as thickness, curvature, branch form and the like of palmprint which cannot be objectively measured in the past is realized, and unprecedented data dimensions are provided for scientific research of hand diagnosis.

The application relates to the technical field of brain science image processing, and discloses a brain tissue neuron morphology three-dimensional reconstruction and connection group analysis method and system. The method comprises the following steps: feature point extraction and matching and optical flow registration are performed on continuous ultrathin section electron microscope images to obtain three-dimensional image body data; a three-dimensional deep convolutional neural network is used for semantic segmentation of cell bodies, dendrites, axons and dendritic spines, and secondary registration is driven through confidence feedback; synapses are detected and classified into excitatory or inhibitory types; topological perception skeletonization reconstruction is performed, and segmentation correction is constrained through topological anomaly feedback; and a synapse connection matrix is constructed, and graph theory analysis is performed.

The application discloses a coal body hydraulic fracturing fracture three-dimensional space network reconstruction and characterization method, which is based on real CT three-dimensional body data, can realize the transformation of fracture structure from geometric visualization to parameterization and evaluation, can directly reflect the real connection relationship of the fracture in the three-dimensional space by identifying and three-dimensionally reconstructing the internal fracture of the coal body before and after fracturing, avoids the connection misjudgment and scale deviation caused by two-dimensional slice inference, improves the authenticity of the evaluation result from the source, and simultaneously adopts skeletonization and node-edge modeling on the fracture network, converts the three-dimensional voxel fracture structure into a node-edge form fracture network model, significantly reduces the calculation complexity under the premise of ensuring the fracture space connection relationship, enables the connectivity and flow conductivity characteristics of the fracture network to be expressed in the form of unified parameters, and is especially suitable for the fracture network characterization of the laboratory true triaxial test and the digital coal rock sample data based on CT scanning.

A method for providing a gripping point for mechanical interaction on an object includes (i) providing at least one image of the object, the image being a result of an acquisition by at least one sensor, (ii) providing object prompts based on the provided image by utilising an object detector and / or an unsupervised or semi-supervised segmentation method, (iii) performing a segmentation on the provided images using the provided object prompts to generate an object mask representing the object, (iv) performing a morphological skeletonization in the provided image based on the generated object mask to obtain a polyline, (v) reducing the polyline to a defined number of points, and (vi) selecting at least one of the points to provide the gripping point for mechanical interaction on the object. A computer program, an apparatus, and a storage medium for this purpose are also disclosed.

A method for generating a skeleton in an image of a cavity of an organ of a body includes receiving a map of the cavity, the map including surface voxels and interior voxels. A subset of the interior voxels is generated, of candidate locations to be on the skeleton. The subset is pruned by removing outlier candidate locations. Using a geometrical model including a statistical model, the candidate locations remaining in the pruned subset are spatially compressed. The compressed candidate locations are connected to produce one or more centerlines of the skeleton. At least the skeleton is displayed to user.

The invention discloses a crack feature quantification method and device based on camera calibration and skeletonization processing, and relates to the technical field of crack image recognition. The method comprises the following steps: performing image calibration according to a crack image based on a Zhang calibration method to obtain an internal and external parameter matrix and a distortion coefficient of a camera, a corner pixel coordinate of a two-dimensional grid calibration plate and a calibrated crack image; performing inverse perspective correction on the calibrated crack image based on the angular point world coordinates of the two-dimensional grid calibration plate according to the internal and external parameter matrixes, the distortion coefficient and the angular point pixel coordinates to obtain a corrected crack image; according to the corrected crack image, skeleton extraction of unit pixel width is carried out by using a skeletonization processing algorithm to obtain a crack skeleton image; and according to the crack skeleton image, performing calculation by using a feature quantization algorithm to obtain crack quantization features. The method is a non-contact, planar-coverage, low-cost and high-robustness crack feature quantification method.

This invention provides a method and apparatus for identifying cracks in ancient buildings based on image processing, belonging to the field of image processing. The method includes: acquiring a close-up image of cracks in an ancient building; segmenting the cracks using a region growing algorithm combining local Otsu's method and local mean; performing skeletonization processing on the segmented crack regions to obtain a crack skeleton with a single pixel width; and calculating the length and average width of the crack based on the area of ​​the crack region and the size of the crack skeleton, thereby achieving quantitative identification of the crack. This effectively reduces noise and background texture interference, improves the accuracy of seed point selection, and provides a more reliable and accurate technical means for detecting cracks in ancient buildings, helping to promptly identify potential safety issues in ancient buildings.

The invention provides an intelligent mirror multi-dimensional health assessment system, and relates to the technical field of health assessment, and the system comprises an acquisition module which is used for carrying out human body skeletonization processing on an acquired continuous three-dimensional image sequence of user motion and position information of user limbs in a space, extracting three-dimensional space coordinate data of key joints of a user body, and transmitting the three-dimensional space coordinate data to a processing module; obtaining a real-time skeleton model of user motion; and the pointing module is used for establishing a first reference vector pointing to the central points of the left and right shoulder joints and a second reference vector pointing to the central points of the left and right hip joints by taking the central point of the trunk of the user as a space base point based on the real-time skeleton model, so as to determine a main reference plane of the movement of the user. According to the method, through extraction of motion skeleton data, construction of a reference and dynamic feature plane, calculation and evaluation of key indexes, generation of motion attitude correction parameters and visual calibration guidance, evaluation and real-time optimization of motion quality are realized, the motion effect is improved, and the damage risk is reduced.

The invention provides a tunnel face joint fissure recognition method based on semi-supervised learning, and relates to the field of tunnel engineering.The tunnel face joint fissure recognition method comprises the steps that training images of tunnel face joint fissures are collected and marked, a first training set is constructed, and a second training set is constructed after the images are subjected to enhancement preprocessing; segmenting the second training set image by using a semantic segmentation model, obtaining a joint fissure space region, and performing multi-scale feature extraction; building a deep learning model introducing a semi-supervised learning framework, and training by using the feature map and the regional mask to obtain a prediction model; performing segmentation and feature extraction on a to-be-identified image, and inputting the image into the prediction model to obtain a region mask; skeletonization, connected domain analysis and multi-scale feature extraction are carried out based on the joint fissure mask prediction result, and the length and the inclination angle of the joint fissure are calculated to complete recognition.

The invention discloses a space coordinate expansion and reconstruction method, equipment and medium, and the method comprises the steps: extracting a space expression point which is marked as a target tissue type from space group data, and generating a standardized binary image; executing closed operation and smoothing edges, extracting a plurality of contour areas and combining the contour areas into a main contour, and executing an image skeletonization algorithm on the main contour to generate a skeleton path diagram; selecting the longest path from the shortest paths as a central main path; back-projecting the central main path under the image coordinates to an original space coordinate system, and determining the central main path and a starting point and an ending point thereof; constructing a tangent-normal coordinate system, and calculating the distance from the spatial expression point to the first reference point; generating an edge reference line; projecting all spatial expression points to the expanded coordinate system, and determining a projection target; and performing projection according to the projection target to obtain expansion coordinates. According to the method, a complete skeleton path and a high-smoothness edge curve are automatically generated, and the real positions of hosts and / or microorganisms and spatial expression points in the space are effectively restored.

The invention relates to the technical field of industrial automation and machine vision, and provides a welding position recognition method and device based on point cloud projection skeletonization and a medium. According to the method, after three-dimensional point cloud of a workpiece is obtained, preprocessing and coordinate system alignment are carried out, vertical part point cloud is cut out and projected to a two-dimensional plane to generate an image; carrying out binarization and refinement to obtain a skeleton diagram; straight lines are detected from the skeleton drawing, a vertical welding seam is recognized according to the straight line included angle relation, and a horizontal welding seam is recognized by extracting an image contour; and finally, the identification result coordinates are inversely transformed to an original space. According to the method, three-dimensional recognition is converted into two-dimensional image processing, the robustness to point cloud missing is high, vertical and horizontal welding seams can be accurately recognized at the same time, the universality is good, and the calculation efficiency is high.

The application discloses a production line assembly real-time detection method and system applying computer vision, and relates to the technical field of computer vision, and comprises the following steps: generating a lifting saliency map of an FPC assembly part; obtaining an outer contour line based on a bright field image of the FPC assembly part, and performing bilateral extension on the outer contour line to obtain a welt band area; generating a welt enhancement map according to a welt band saliency map corresponding to the welt band area; establishing a local coordinate system based on the outer contour line corresponding to the bright field image; performing binarization and skeletonization processing on the welt enhancement map to obtain a bright ridge center line map, and generating a welt unwound ridge line map according to the local coordinate system corresponding to the bright ridge center line map; obtaining an ordered center line point set according to an unwound connected area corresponding to the welt unwound ridge line map, and calculating a fitting deviation of the ordered center line point set based on a fitting arc corresponding to the ordered center line point set; and the application improves the recognition accuracy of a local edge abnormal area.

The invention relates to a method (100) for providing a capture point for a mechanical interaction on an object (1), comprising the following steps:-providing (101) at least one image of the object (1), the image being a result acquired by at least one sensor (2),-providing (102) an object cue on the basis of the provided image by using an object detector and / or an unsupervised or semi-supervised segmentation method, -performing (103) a segmentation on the provided image using the provided object cues to generate an object mask representing the object (1),-performing (104) a morphological skeletonization in the provided image based on the generated object mask to obtain a polyline,-reducing (105) the polyline to a defined number of points, and-generating (105) the object (1). -selecting (106) at least one point to provide a point of gripping for the mechanical interaction on the object (1). The invention relates to a computer program, a device and a storage medium for this purpose.

The invention discloses a tunnel lining defect non-destructive testing method, and relates to the technical field of non-destructive testing. During operation of a system, an image acquisition device is used for acquiring continuous image sequences of a tunnel lining, spatial registration is realized based on feature point matching between adjacent image sequences, an initial multi-source image set is obtained, the initial image set is preprocessed, and the initial multi-source image set is obtained; obtaining an enhanced multi-source image sequence, carrying out feature extraction and defect detection, extracting hidden crack edge features by adopting a multi-scale convolutional network, inhibiting dynamic interference in combination with an image-to-image time sequence difference detection model, generating a segmentation image of a hidden crack defect region, and carrying out segmentation on the hidden crack defect region; the method comprises the following steps: identifying hidden cracks, acquiring width and depth indication characteristics and position parameters of the hidden cracks through crack skeletonization mapping, inputting the width and depth indication characteristics and position parameters into a multi-task learning model, constructing a defect severity index based on the identified hidden crack parameters, performing health assessment on the evolution state of the hidden cracks in combination with time sequence trend prediction, and automatically generating maintenance priorities, risk levels and maintenance suggestions.

This application relates to the fields of industrial automation and machine vision technology, and provides a welding position recognition method, device, and medium based on point cloud projection skeletonization. The method acquires a 3D point cloud of a workpiece, performs preprocessing and coordinate system alignment, then crops out the vertical portion of the point cloud and projects it onto a 2D plane to generate an image. This image is then binarized and thinned to obtain a skeleton map. Straight lines are detected from the skeleton map, and vertical welds are identified based on the angle relationships between the lines. Horizontal welds are identified by extracting the image contours. Finally, the coordinates of the recognition results are inversely transformed back to the original space. This invention transforms 3D recognition into 2D image processing, exhibits strong robustness to point cloud defects, can accurately identify both vertical and horizontal welds simultaneously, and has good versatility and high computational efficiency.