393 results about "Slicing" patented technology

The invention relates to the technical field of 3D printing models, in particular to a 3D printing model slice feature compensation method and system and a medium. The method comprises the following steps that a 3D printing model is collected, a model feature skeleton is split, transverse slice features are generated, then a 3D printing environment is collected, virtual simulation is carried out, environment stress analysis is carried out on the skeleton slice features to obtain environment stress data, then deformation scale deduction and thermal deformation prediction are carried out based on the environment stress data, and a thermal deformation model is obtained. On the basis, reverse deformation simulation is carried out, deformation repair requirements are evaluated, a deformation compensation path is planned, finally, compensation simulation of a model feature skeleton is achieved, non-target deformation is analyzed, a compensation path is adjusted, and therefore the feature compensation method is optimized. According to the 3D printing model slice feature compensation method, the accuracy and quality of the 3D printing model are integrally improved.

The invention discloses a 3D printing partition filling path planning method based on stress distribution. The 3D printing partition filling path planning method is suitable for FDM (fused deposition modeling) printing. The method comprises the following steps: firstly, performing finite element mechanical simulation on a to-be-printed model to obtain node coordinates and stress distribution data; then, an improved Gaussian mixture model (GMM) algorithm is adopted to carry out stress partitioning on the model, and an area boundary is optimized in combination with image processing and a Snake algorithm. Differentiated path planning of a high-stress area, a conventional area and a non-filling area is realized through an adaptive slicing and partition filling strategy, and a G-code file with multiple adjustable parameters is automatically generated. According to the method, the structural strength and the material utilization rate of a printed piece are effectively improved, and the method is suitable for high-performance 3D printing manufacturing of a complex structure.

The invention discloses a retrieval enhanced warehouse-level code completion method based on semantic topology and disambiguation redundancy pruning. Firstly, a key value code library for accurate mapping of source code fragments and induction sub-graphs is constructed through a specific slicing algorithm; in the retrieval stage, four-level layered optimization is adopted, structural similarity is evaluated by extracting a deep semantic relationship, trimming accurate duplicate items and using a new graph-based measurement (tradeoff is performed on editing according to topological importance), results are reordered to maximize correlation and diversity, candidate items are systematically refined, and the retrieval efficiency is improved. The problems of retrieval redundancy solidification and surface similarity misleading are solved, meanwhile, cross-module dependence is analyzed through an external perception identifier disambiguator, and the problem of cross-file symbol ambiguity is solved; and finally, fusing an optimization result to generate a prompt to drive the LLM to generate higher-quality output. According to the invention, through coordination of semantic and structural signals, strong performance can be obtained even in a large-scale and resource-limited code library. Meanwhile, the design allows it to be orthogonally complementary to other cross-file methods, providing collaborative improvements when used in combination.

The invention relates to the technical field of computer application, in particular to an industry document generation method based on a large language model and retrieval enhancement, which comprises the following steps: acquiring a first user demand of a building engineering project to be input, performing data preprocessing on the first user demand, and acquiring the preprocessed first user demand; performing semantic alignment on the preprocessed first user demand by adopting a cross-modal embedding model to generate a multi-modal fusion vector; according to a dynamic semantic slicing strategy, performing slice segmentation on the multi-modal fusion vector, and obtaining a plurality of slice units containing complete semantics as segmented multi-modal fusion vectors; inputting the segmented multi-modal fusion vector into a pre-trained special retrieval enhancement large model for the building industry, and generating an industry document which conforms to a building engineering specification and a first user demand; according to the method, the problems that the generated building related document content in the building industry is low in professionality and accuracy and cannot meet the industry requirements are solved.

The embodiment of the invention provides a front-end large file uploading method and device, and the method comprises the steps: receiving a large file uploaded by a user through a front end, carrying out the dynamic fragmentation of the large file according to a set dynamic fragmentation model, determining each corresponding fragment, carrying out the priority judgment of each fragment, and determining a corresponding priority order, the method comprises the following steps: determining a corresponding concurrent uploading fragment quantity according to a network bandwidth, a load and a preset concurrent quantity rule, carrying out concurrent uploading operation on each fragment according to a priority sequence and the concurrent uploading fragment quantity according to a preset multiplexing technology, judging whether concurrent uploading of each fragment is successful or not, and if the concurrent uploading fails, executing the concurrent uploading operation on each fragment. According to the method and the device, file fragmentation and concurrent uploading can be carried out on the basis of dynamically adjusting the sizes of the slices and the concurrent quantity, and the efficiency and the flexibility of uploading the large file are improved. The method and the device have the advantages that the file fragmentation and the concurrent uploading can be carried out on the basis of dynamically adjusting the sizes of the slices and the concurrent quantity, and the uploading efficiency and the uploading flexibility of the large file are improved.

The invention provides a small target detection method based on a superpixel mask and a dynamic kernel. The small target detection method comprises the following implementation steps: acquiring a training and testing sample set; constructing a target detection network model based on the superpixel mask and the dynamic kernel, and carrying out iterative training on the target detection network model; and obtaining a small target detection result. The greedy slicing module carries out initial image block segmentation according to the center point and the scale, the defect that space correlation features are segmented due to segmentation is avoided, the dynamic kernel module carries out multi-scale feature fusion on each image block, the receptive field can be dynamically adjusted, wide-area context information of targets of different scales can be flexibly adapted, and the image segmentation efficiency is improved. The detection precision is effectively improved; a superpixel mask generator generates a foreground region mask irrelevant to the category for each extracted feature image, and a dynamic kernel module performs multi-scale feature fusion on each foreground image block containing a target cluster obtained by a greedy slicing module; and the influence of processing useless background information on calculation overhead when the multi-scale features are obtained is avoided.

The invention discloses a binary vulnerability static mining method and system based on a large model, and relates to the data security and computer technology, and the method comprises the steps: extracting an external function from a decompiled binary file, recognizing a taint rendezvous point possibly causing a vulnerability by using the large language model LLM in combination with a rule base, and receiving an externally input taint source; generating an attacked target according to the calling point of the taint convergent point; performing reverse slicing based on the vulnerable target to generate a function call chain; screening the generated generation function call chain to obtain a dangerous flow; selecting an adaptive prompt template according to the type of the taint convergent point of the dangerous flow so as to perform block processing on the long function body and then perform combination to generate a prompt sequence; and utilizing the prompt sequence to guide LLM to perform vulnerability detection on the dangerous flow. According to the method, the accuracy, efficiency and automation level of binary program static vulnerability mining can be improved, and the defects of a traditional method in the aspects of complex dependency analysis and unknown vulnerability detection are overcome.

According to the emergency safety management GIS big data vector slicing service map, the vector data are thinned and simplified under each level by using the optimized global constraint algorithm, and the simplified and thinned vector data are cut and sliced and are published in a service form. The style of the tile is designed, and the client renders and visualizes the requested vector tile according to the specified style. And designing a cutting thought, and cutting point, line and surface elements in the vector data. And organizing and storing the generated tiles by using a quadtree index. A cache strategy of a client is designed, existing WMS and WMTS service standards are expanded, a vector tile map style is designed, and vector tiles are published in a service form. According to the method, a prototype system is created, static and dynamic slicing services of the vector map are realized, loading and rendering are carried out at a client, and particularly, when the emergency safety management quantity is huge and the quality is poor, the map precision and efficiency advantages are obvious.

The invention provides a 3D printing path planning method and system for a three-dimensional curved surface. The method comprises the steps that printing area identification is conducted on a 3D printing model, a curved surface area is obtained, slicing processing is conducted according to the dynamic slice layer thickness, a printing path is obtained through a preset printing path planning method according to the dynamic slice layer thickness, node representation is conducted on the printing path, weight values between path nodes are determined, and a printing path sorting table is generated; 3D printing is executed, detection is conducted through a preset visual sensor, curved surface parameters are obtained, and 3D printing parameters are adjusted according to the curved surface parameters; according to the method, self-adaptive layering processing is carried out by fusing the material performance, the printer parameters and the 3D printing requirements, the path is represented in a nodal mode, the weight values between the nodes are constructed, the path sorting table is generated accordingly, transition path planning and optimization are carried out, 3D printing path planning and optimization of the three-dimensional curved surface are achieved, and the 3D printing efficiency and the printing quality are improved.

The invention belongs to the technical field of geographic information technology and remote sensing image processing and service, and discloses a slice-free image service system and method based on dynamic projection and real-time mosaic, which abandons the traditional pre-slicing mode, directly provides online service based on original image data, and improves the service efficiency. The problem of redundant storage caused by pre-generation and storage of massive tiles is fundamentally eliminated, and storage space is saved by up to 90%. Meanwhile, when the original data is updated, the system does not need to carry out a time-consuming re-slicing process, real-time updating and publishing of services can be realized, and the core pain points of long data updating period and high delay in the traditional technology are thoroughly solved. Through the integrated dynamic projection engine and the real-time mosaic module, the on-demand service request of the client for any coordinate system, any spatial range and any resolution can be responded.

The invention discloses a method and a system for solving compilation errors in intelligent contract version evolution. The method comprises the following steps: collecting and cataloguing major changes of all records based on an official Solidity website, screening to obtain major changes causing a compiling problem, and constructing a knowledge base of version migration; combining codes with compiling problems and compiling error information of the codes; slicing the smart contract code by using a static analysis tool, and retrieving a version migration knowledge base by using a retriever based on compiling error information to obtain related version migration knowledge; and guiding the large language model to generate a high-quality patch in combination with a result obtained by the analysis. According to the method, a static analysis tool, structured retrieval of expert knowledge and the generation capacity of a large language model are organically combined, the problem of compiling errors caused by Solidity language evolution is systematically solved, and developers can be effectively assisted in coping with version migration and code repair tasks.

The invention relates to the field of concrete 3D printing, and discloses a path optimization method for concrete 3D printing, which comprises the following steps: firstly, slicing a three-dimensional model of a to-be-printed concrete member, converting the sliced model into a geometric model, and initializing position information of geometric points in the geometric model; according to a graph theory principle, dividing the geometric model into a plurality of areas which can be printed at one time; the method comprises the following steps: constructing slices of a printing model into geometric figures which can be identified by a computer, extracting and numbering vertexes of the geometric figures, respectively recording coordinates of the vertexes, setting an adjacent matrix to describe a connection condition between the vertexes, then storing the vertexes into a partition matrix according to a rule, the partitions can be traversed at a time; finding out a shortest printing path in the plurality of divided areas by using an ant colony algorithm; and the printing platform is controlled to move according to the planned optimal printing path.

The invention discloses an XR-based immersive three-dimensional modeling and straight-through 3D printing system and method, and the method comprises the steps: S1, collecting user input, and converting the user input into an interaction instruction; s2, generating three-dimensional model data; s3, superposing the three-dimensional model into a real physical environment according to a proportion of 1: 1, and outputting a virtual-real fusion display result; s4, performing printability analysis on the model, marking a problem area, guiding a user to modify and providing an automatic repair suggestion; s5, generating a support structure, and providing visual preview and interactive editing; s6, slicing and path planning are executed, and a printing instruction file meeting the control requirement of the 3D printer is generated; s7, communication connection with the 3D printer is established, and printing state information is fed back; and S8, providing storage and calling support of the three-dimensional model file, parameter configuration and environment map data. According to the invention, seamless integration of immersive XR modeling, MR real-time verification and one-key direct-through 3D printing is realized.

The invention relates to the field of BIM model visualization, in particular to a BIM design terrain and GIS live-action texture fusion method, BIM design terrain and GIS live-action texture data are processed respectively, when GIS live-action texture data changes, data updating can be achieved only by replacing slice data, and the convenience of data updating is improved; according to the method, the relation between the image slice nodes of the GIS live-action texture data and the space coordinates of the BIM design terrain is established through the coordinate system transformation matrix, so that coordinate transformation between the image slice nodes and the space coordinates of the BIM design terrain is facilitated, the space coordinates of the BIM design terrain correspond to the texture coordinates of the GIS live-action texture data, and finally fusion of the BIM design terrain and the GIS live-action texture data is realized.

The invention particularly relates to a product integral forming automatic optimization method applied to 3D printing, which comprises the following steps: model importing and processing: slicing a three-dimensional model, including comprehensively analyzing a geometric coefficient and a visual coefficient to obtain an evaluation coefficient, and judging a non-manifold structure according to the evaluation coefficient; preparing materials; calibrating the equipment; printing a product; and post-processing. According to the method, the geometric coefficient and the visual coefficient are comprehensively analyzed to obtain the evaluation coefficient, and the non-manifold structure is judged according to the evaluation coefficient; in geometric coefficient calculation, the number of vertex edges of the model is meticulously counted and analyzed, so that the local abnormal condition of the topological structure of the model can be quantified; in the aspect of visual coefficient, the color consistency of the model is considered; compared with a single detection means, the multi-dimensional evaluation mode can more comprehensively and accurately detect the potential non-manifold structure in the model, and discover the problem possibly causing printing failure or quality reduction in advance, thereby greatly improving the printing success rate.

This disclosure relates to video coding and more particularly to techniques for signaling picture information for coded video. A method for the techniques comprises: receiving a slice header syntax structure; determining whether the slice header syntax structure includes a picture header syntax structure; and in a case where the slice header syntax structure includes the picture header syntax structure, parsing a first syntax element specifying a picture parameter set identifier from the picture header syntax structure and parsing a second syntax element specifying a picture order count from the picture header syntax structure.

The invention discloses an unmanned aerial vehicle aerial photography water surface small target detection method and system based on dynamic slicing and correlation constraint, and belongs to the technical field of computer vision, artificial intelligence and geographic information processing, and the method comprises the steps: S1, data acquisition: obtaining high-resolution video stream data collected by an unmanned aerial vehicle flight platform; s2, carrying out dynamic slicing, and segmenting the original high-resolution image frame into a plurality of sub-image slices of which the sizes are matched with the input of the detection network; s3, target detection: inputting the subgraph slices into the model for reasoning, and outputting the category and local pixel coordinates of the target; s4, result fusion: mapping detection results of the sub-image slices back to an original image coordinate system to generate a global detection result; and S5, performing space-time positioning, and converting the pixel coordinates of the global detection result into geographic latitude and longitude coordinates by using a coordinate conversion algorithm. According to the method, the problems of leak detection and false detection caused by extremely high image resolution, small target size and background noise interference in a water surface environment are effectively solved.

The invention discloses an RAG content generation method and system based on dynamic slicing and self-adaptive fusion, and relates to the field of natural language process.The method comprises the steps that in response to a user query instruction, semantic fusion processing is conducted on a plurality of related text blocks, and a preliminary text sequence is traversed; detecting a logic breakpoint between adjacent text blocks through a context association identifier, recalling a middle connection text block corresponding to the logic breakpoint, and inserting the middle connection text block into a sequence to generate a fusion text stream; performing adaptive information density adjustment on the fused text stream to obtain optimized context content; generating structured prompt information based on the optimized context content and the user query instruction; and calling the target large language model to generate a user query result based on the structured prompt information. The answer accuracy and logic continuity of the target large language model for user query are effectively improved, and the utilization efficiency of model computing resources is optimized.

The invention discloses a programming wire with gradient toughening FFF printing and a preparation method thereof, and relates to the technical field of programming wire preparation, and the preparation method comprises the following steps: establishing a filament model on three-dimensional software; establishing a spiral line and taking the spiral line as a path; the volume proportion is designed according to requirements; in the filament model, generating a spiral line to determine the distribution of the flexible material, and calculating the total length of the spiral line in real time; calculating the radius of a small circle of the flexible material; a spiral boss is generated through a three-dimensional modeling software lofting tool; calculating the extrusion amount and the total extrusion amount of the required filaments printed once according to a formula; importing the filament model into slicing software for dyeing and slicing treatment; and manufacturing of the programming wire with gradient toughening FFF printing is completed through 3D printing. According to the preparation method, material waste is reduced, gradient distribution of the two materials in the printed part is achieved, and therefore the mechanical property stability and the surface quality of the printed part are remarkably improved.

The invention discloses a method and system for converting a CAD drawing into spatial data and making a map, and the method comprises the following steps: S1, data uploading: uploading a CAD file to a background server through a Web end, storing the CAD file, and detecting and verifying a CAD file coordinate system; s2, layer catalog analysis and selection configuration: analyzing a layer catalog and element primitive information of a CAD file stored in a background server, and selecting a layer to be converted; s3, constructing a multi-thread conversion task, and performing spatial data fitting on the graphic elements by using an algorithm; s4, storing the converted spatial data into a database, and constructing an attribute field and style mapping relation; s5, generating a map tile pyramid based on a dynamic slicing technology, and issuing a map service address; and S6, the front end calls the service address for rendering, and supports real-time collaborative editing and data export. According to the method, quick and efficient conversion of each layer element of the CAD file is realized, the conversion data precision is high, the quality is stable and reliable, and the data redundancy is less.

One embodiment provides a graphics processor comprising: a memory interface; a plurality of interfaces to the plurality of compute engines; a cluster of processing resources, the cluster of processing resources comprising a plurality of processing resources configured to execute instructions on behalf of the plurality of compute engines; and virtualization circuitry configured to enable time division virtualization of the plurality of processing resources via the plurality of computing engines, the virtualization circuitry is to concurrently process workloads during a time slice via dynamic assignment of workloads from a plurality of guest machine software environments to a plurality of interfaces to a plurality of compute engines.

The invention discloses a cloud-based 3D printer intelligent control method and system, and the method comprises the steps: after receiving an STL model file and printing parameters uploaded by a user at a cloud, executing a slicing operation to generate G code data, and transmitting the G code data to a printer; then, the cloud end continuously receives a G code line number currently executed by the printer, context analysis is carried out on G code data according to the G code line number, and a real-time printing context vector reflecting the current working state and the expected action of the printer is generated; meanwhile, in combination with a camera video frame collected by a printer, AI reasoning based on context perception is carried out on visual data by utilizing the real-time context vector, so that a more accurate exception decision is generated and an alarm is sent to a client. In this way, the accuracy and robustness of printing anomaly detection are remarkably improved, and therefore more intelligent, efficient and reliable remote printing experience is provided for a user.

The invention discloses an Internet webpage content feature extraction method based on artificial intelligence, which comprises the following steps: S1, acquiring a webpage HTML source file and a rendering image, and preprocessing the webpage HTML source file and the rendering image; s2, performing semantic classification on DOM nodes, encoding the DOM nodes into three types of identifiers, and constructing a node label sequence; s3, performing time sequence synchronization on the node attribute vector, the node tag sequence and the visual area set, and performing block-level slicing; s4, inputting the block-level slices into a gated recursive attention network, extracting multi-modal joint representation, and executing attention aggregation; s5, performing structure alignment on the feature fusion sequence, calculating a cross-node consistency distance, and screening a target slice set with high structure cohesion; and S6, mapping the target slice set to a content feature space, and generating a content feature tag set. According to the method, the structural accuracy, the semantic integrity and the multi-modal fusion precision of webpage content extraction are improved.

A method of forming a volumetric effect on a game map of a video game is provided. In the method, a variable density volume of the game map is determined based on a mesh of the game map and a bounding box. The variable density volume includes the volumetric effect. The variable density volume is sliced into a plurality of texture layers. Each of the plurality of texture layers includes information associated with at least one of the volumetric effect or the mesh. An interaction between a virtual object on the game map and the volumetric effect is determined based on movement of the virtual object. The volumetric effect is rendered based on a cut out of the virtual object in the variable density volume and the determined interaction.

The invention is suitable for the field of three-dimensional printing, and provides a method for generating a slice map and a 3D printing method. A method of generating a slice map includes providing a three-dimensional model including a plurality of facets, each facet associated with a surface feature; outputting a plurality of slice images through a plurality of tangent planes intersected with the three-dimensional model, wherein each slice image comprises a first intersection point of the tangent plane and the plurality of facets; through a plurality of straight lines perpendicular to the tangent plane, determining second intersection points of the plurality of straight lines and the plurality of patches; and determining an association attribute of the slice map at least based on the surface feature associated with the first intersection point and the surface feature associated with the second intersection point. According to the method, the accuracy of the obtained slice image can be improved.

The invention provides a 3D printing slicing method for a large-size complex model. The 3D printing slicing method comprises the steps that a to-be-printed model is imported into a slicing system; optimizing triangular patches of the model, and dynamically grouping the triangular patches of the model according to height intervals; establishing a z-direction scanning line array according to a surrounding frame of the model; solving a basic intersection point of the triangular patch corresponding to each height interval and the scanning line array, calculating a compensation intersection point by adopting a compensation algorithm, marking the compensation intersection point, obtaining a layer cutting intersection point set corresponding to each height interval, and storing the layer cutting intersection point set in a local disk; wherein the layer cutting intersection points in the layer cutting intersection point set comprise basic intersection points and compensation intersection points; and generating slice data based on the layer cutting intersection point set corresponding to each height interval. According to the method, the single-time fixed processing height can be manually adjusted according to the model size and the current memory state, and by means of the mechanism, the optimal memory utilization rate can be achieved while the processing precision of GB-level large industrial models or MB-level small parts is guaranteed.

The invention discloses a ROS2-based surface exposure rapid prototyping method, device and system and a medium, and relates to the technical field of additive manufacturing. The method comprises the steps that all hardware nodes are reset and initialized, a forming platform is flush with a powder laying plane, powder is supplemented, and a scraper is reset. And importing the three-dimensional model of the part to be formed into ROS2-based slicing software, generating a support structure after attitude optimization, slicing according to a preset layer thickness to obtain a slice layer, and performing rasterization processing to generate a black and white mask image. And circulating subsequent steps for each slice layer until all slice layers are completed: descending a servo motor of a forming bin according to a preset layer thickness, providing powder by a powder supply bin, and then spreading and slicking the powder. And sending the black and white mask image corresponding to the current slice layer to the DLP to execute projection. The SLM laser node performs selective laser melting according to the projected area. And a molten pool image is captured in real time and analyzed, an analysis result is input to an ROS2 decision node to dynamically adjust process parameters, and closed-loop control is achieved.

The invention relates to the technical field of computer graphics and three-dimensional modeling optimization, in particular to a 3D model lightweight method and system based on point cloud hierarchical slicing and dynamic mapping mapping.When the 3D model lightweight method and system are used, the storage efficiency, the rendering performance and the visual fidelity are improved by establishing a point cloud and mapping bidirectional coupling mechanism, and the 3D model lightweight effect is improved. In the aspect of storage efficiency, an efficient geometric and texture association storage system is constructed by relying on a map curvature adaptive segmentation and dynamic slice binding technology, an excellent geometric and texture compression effect is realized, the storage cost of a high-precision three-dimensional model is greatly reduced, and in the aspect of real-time rendering performance, through a dynamic loading strategy during operation, the real-time rendering efficiency is improved. By combining efficient visibility screening acceleration and an intelligent resource preloading technology, the real-time scheduling efficiency of the model is greatly improved, and the working efficiency of scenes such as design iteration and model editing is remarkably improved.

A method of printing a plurality of parts with varying layer heights in a single print job in a 3D printer includes providing a plurality of sliced 3D digital representations of a plurality of parts with varying slice heights in the Z direction, positioning the digital representation of parts on a build platform of the 3D printer and determining a printing sequence of the layers of the parts based upon a lowest Z height between the plurality of sliced 3D digital representations. The method further includes extruding the layers of the plurality of disjoined parts with one or more print heads based upon the determined printing sequence and intermittently indexing z height between the sequenced layers until each of the plurality of parts is printed.

The invention relates to an efficient slicing-free method, device and system for mass framing image data. The method mainly comprises the steps of performing first grouping based on a coordinate system of framing images, performing second grouping based on the number of closed graphs after the first grouping, and performing third grouping based on the number of files after the second grouping; establishing a virtual grid file after virtual combination for the framing image data after third grouping, and performing thinning sampling processing to generate a thinning grid file; taking the rarefied grid file as a simplified data set, and taking the framing image file before grouping as an original data set; when rendering is carried out, if the rendering level is lower than a preset level, retrieval and rendering output are carried out through the simplified data set; and if the rendering level is higher than the preset level, performing retrieval and rendering output through the original data set. According to the invention, the rendering efficiency at a low level can be greatly improved, and the rendering performance during low-level rendering is ensured.