49 results about "Labelling algorithm" patented technology

The invention relates to an annotation data processing method and device, computer equipment, a storage medium and a computer program product. The method comprises the steps that a data annotation request sent by a terminal is received, the data annotation request carries sample identification data and annotation algorithm identification data, according to preset annotation algorithm attribute configuration data, algorithm input parameter attributes corresponding to the annotation algorithm identification data are determined, a sample set corresponding to the sample identification data is obtained, and a data annotation result is obtained; and matching attribute feature data of the target attribute from the sample set to obtain a to-be-labeled data set, and pushing the to-be-labeled data set to a terminal. By the adoption of the method, the unified annotation data obtaining capacity can be provided for different annotation algorithms and sample sets, the to-be-annotated data can be rapidly and accurately obtained only by configuring the data and determining the attributes needed by the algorithms according to the preset annotation algorithm attributes for the different annotation algorithms, and then the data annotation efficiency is improved.

The invention belongs to the technical field of graph neural networks, and discloses a multi-semantic view and adaptive contrast learning combined graph classification system and method, and the system comprises a multi-semantic view extraction module which captures high-order structure features through a WL remarking algorithm and a word embedding model, and extracts graph features from different semantic views; the self-adaptive comparative learning module is used for reserving important node features from a plurality of semantic views by self-adaptively generating an enhancement graph and enhancing the relationship among the views through comparative learning of cross-semantic views, and the self-adaptive comparative learning is used for optimizing the expression of the node features through consistency loss; the multi-semantic view fusion module uses an attention mechanism to merge node representations from a plurality of views to distinguish importance of nodes in different views so as to generate richer and more comprehensive graph embedding.

The invention discloses a dead pixel detection method and system combining spatial domain analysis and time domain analysis, and belongs to the technical field of image processing. The method comprises the following steps: an airspace analysis stage: extracting edge information through a Sobel operator; carrying out image binaryzation, filling potential holes by using morphological closed operation, and strengthening dead pixel areas; identifying all candidate dead pixel regions by adopting a connected domain labeling algorithm; in the time domain analysis stage, the spatial position and area characteristics of candidate dead pixel areas are combined, the candidate dead pixel area corresponding relation between cross frames is established, the continuity of the same candidate dead pixel area on a time axis is judged, and only two assumptions are met, the time domain analysis is completed. And finally judging the candidate dead pixel regions with the areas and the contrast ratios of the dead pixel regions exceeding the set threshold values as stable dead pixels and outputting the stable dead pixels. By analyzing the stability of the dead pixels in the time sequence, abnormal dead pixels continuously existing in multiple frames can be effectively recognized, and particularly, the method is more stable and accurate when detecting small-area and low-contrast dead pixels.

The invention discloses a multi-unmanned aerial vehicle distribution and bus charging combined path optimization method, which aims at minimizing task total time, constructs a mixed integer programming model based on a space-time network, and comprehensively considers unmanned aerial vehicle electric quantity constraint, demand point full coverage, bus time window and charging pile number limitation. An original model is decoupled to limit a main problem and a sub-problem by adopting branch pricing and a Dantzigzag-Wolfe decomposition theory, the main problem deals with demand coverage and charging resource allocation, and modeling is a set coverage problem; for single-machine path generation, the latter is modeled as a resource-constrained and replenishable shortest path problem with a time window dependent feature. A multi-unmanned aerial vehicle initial solution is constructed through a random generation method, a dual variable is iteratively solved after a main problem is initialized, a sub-problem is solved based on a multi-label algorithm, and global optimal solution search is realized in combination with a column generation mechanism and a branch strategy. According to the invention, through joint optimization of the departure time and path planning of the unmanned aerial vehicle, the collaborative optimization problem of distribution and charging is accurately solved.

The invention provides an RDMA (Remote Direct Memory Access) out-of-order receiving and selective retransmission method, which comprises the following steps of: recording an arrival state of a data packet by adopting a Bitmap algorithm and / or a segmentation marking algorithm of a sliding window, dynamically generating ACK (Acknowledgement Character) or NAK (Negative Attached Keying) feedback information according to the algorithm, and only retransmitting a lost data packet according to the received NAK feedback information. The RDMA out-of-order receiving and selective retransmission method supports Bitmap statistics of a sliding window or a segment marking method of saving a memory, is compatible with an existing protocol and realizes an ACK mechanism of selective retransmission, and avoids the problem of redundant transmission in a traditional mechanism. The invention further discloses computer equipment, a data transmission system and a computer readable storage medium.

The invention discloses a seed point-based three-dimensional building regular contour automatic extraction method and device, and belongs to the technical field of surveying and mapping. For a building with a right angle between walls, triangulation network data is extracted through a live-action three-dimensional model, a seed point coordinate is taken from a roof point of the building, a connected domain marking algorithm is used for obtaining the range of the roof, and the range of the roof is calculated by setting an elevation threshold value and a normal vector threshold value. According to the method, most of non-wall facade data of the ground and the roof are effectively removed, so that building outer wall triangulation network data are obtained, a projection histogram statistical method is used for processing, peak positions are selected as wall boundaries, and finally the regular contour of the building is drawn. According to the method, the regular contour of the building is accurately described through the projection histogram technology, the steps are simple, and the automation degree and the extraction precision are high.

The invention relates to a skeleton line intersection region redundancy restoration method based on distance constraint and center substitution, and belongs to the field of image processing, and the restoration method comprises the steps: extracting a 3 * 3 neighborhood of each foreground pixel point in a skeletonized image, temporarily setting a center pixel as a background value, and detecting an intersection point through an 8-neighborhood connection marking algorithm; realizing cross point clustering through Euclidean distance calculation and a breadth-first search algorithm based on the spatial proximity relationship of the cross points; and carrying out group screening, ignoring a group only containing a single cross point, and carrying out topology reconstruction on a group containing a plurality of cross points. According to the method, the redundant pixel cluster is accurately identified, the quality of the skeleton line is improved, the redundant pixels in the diagonal line area are accurately identified and cleared through the template matching algorithm, the skeleton line is ensured to strictly meet the single-pixel width requirement, and false endpoints and line length measurement errors are avoided.

The invention relates to a spectral separation regional adaptive fringe projection three-dimensional measurement method, which comprises the following steps of: projecting transverse and longitudinal sine phase shift fringe patterns to the surfaces of a plurality of measured objects, and acquiring a fringe sequence by using a camera; respectively obtaining an object contour mask and a saturated mask; the coordinates of the white area of the contour mask are converted into the coordinates of a projector, and a projection mask is obtained; obtaining a single object projection mask image by using a connected domain labeling algorithm; setting two groups of mask regions by using the marked regions and the saturated mask to generate a longitudinal sine phase shift fringe pattern with adaptive intensity, sequentially projecting two groups of region fringes to the surface of a measured object, and simultaneously collecting region fringe images of spectrum separation; performing independent phase demodulation and phase fusion on the fringe patterns of different channel colors corresponding to the object, and performing three-dimensional reconstruction according to a fused phase result; according to the method, the measurement efficiency is remarkably improved on the premise of eliminating stripe aliasing interference.

A computer-implemented method for a digital security system receives unlabeled event data associated with a computing environment, clusters via an unsupervised machine learning model the unlabeled event data into clusters of unlabeled event data where unlabeled event data in one cluster are more similar to each other than to unlabeled event data in other clusters, selects a respective subset of unlabeled event data for each cluster of unlabeled event data, translates via a large language model artificial neural network each unlabeled event datum in each respective subset of unlabeled event data into a description for the unlabeled event datum, and applies a label via a labeling algorithm to at least one unlabeled event datum in a respective cluster responsive to and representative of the respective description for the unlabeled event datum in the respective subset, thereby transforming the at least one unlabeled event datum to a labeled event datum.

The embodiment of the invention provides an industrial monitoring method and device, electronic equipment and a medium. According to the embodiment of the invention, the data model and the algorithm program of the industrial project are obtained, and the algorithm program is analyzed to generate the algorithm flow chart, so that the monitoring parameters in the data model and the program variables in the algorithm program are accurately bound based on the variable binding address. In the execution process of the industrial project, the real-time data of the monitoring parameters are dynamically mapped to the corresponding program variables in the algorithm flow chart, and the corresponding target flow chart nodes in the algorithm flow chart can be determined and marked according to the actual execution state of the industrial project, so that the operation and maintenance threshold is reduced, the automation level of industrial monitoring is improved, and the industrial monitoring efficiency is improved. Problems in the production process can be found and solved in time, and finally the stability and efficiency of industrial production are improved.

To provide a method, a device, and a non-transitory computer-readable medium for generating and managing projects and tasks based on messages exchanged between members and assigned representatives.SOLUTION: A message tagging method in a chat forward task facilitation communication system includes receiving a set of messages between a member and a representative while the set of messages are being exchanged, processing the set of messages to identify task flags associated with the set of messages, tracking a chat flow within a chat interface, processing the chat flow using a tagging algorithm that automatically assigns at least one of the task flags to messages of the set of messages, and generating a message history including the set of messages and corresponding task flag assignments with associated tasks of a set of tasks.SELECTED DRAWING: Figure 18

The invention discloses an unmanned aerial vehicle distribution scheduling model for transferring orders and a solving algorithm thereof. Model establishment comprises the following processes: 1) defining state parameters and variables in unmanned aerial vehicle distribution; 2) determining a target function of the model, wherein the target function of the model represents the total time for minimizing the delivery of all orders by the unmanned aerial vehicle; and 3) determining constraint conditions. According to the method, a scheduling model is solved, a main problem set coverage model is firstly established, then an initial path for unmanned aerial vehicle distribution is generated by using a large-field algorithm, a relaxation main problem is solved, dual variables and reduced cost defined by the dual variables are obtained, sub-problems are solved by using a label algorithm, and an optimal path is obtained. According to the method, the scheduling model can be effectively solved, the distance of the distribution path of the unmanned aerial vehicle is reduced, and the utilization rate of a single unmanned aerial vehicle is improved.

The present application discloses a blockchain-based textile industry supply chain information tamper-proof method and system, the method comprising: establishing a supply chain traceability blockchain with multiple links in the textile industry supply chain as nodes; wherein, block information is used between nodes of the traceability blockchain to transmit interactive data; block information is marked with a preset symbol marking algorithm for link relationships; it is determined whether the block information between two adjacent parent-child nodes is consistent; when the block information is inconsistent, it is determined that the parent node of the two adjacent parent-child nodes has undergone information tampering, and the link relationship mark in the block information is updated. The present invention constructs a traceability blockchain for the textile industry supply chain, realizes the transparency of supply chain data; uses a preset symbol marking algorithm to mark the link relationship of block information, and when tampering occurs in the blockchain, it can efficiently locate the location where the tampering occurred, thereby ensuring the information security of the supply chain.

The invention belongs to the field of intelligent manufacturing, and provides a multi-target flexible processing optimization method and device based on dynamic programming and an AND / OR network model for solving the problem of flexible collaborative optimization of processes, sequences and resources in multi-variety small-batch production. According to the method, an AND / OR network model integrating operation, sequence and resource flexibility is constructed, and a candidate state (process-resource tetrad) is introduced to simplify resource modeling; and establishing a time and cost dual-objective optimization function, iteratively expanding a process path by adopting a dynamic planning label algorithm, and pruning through a dominating rule to generate a non-dominating Pareto optimal solution set. According to the scheme, the limitation that a traditional MILP model is high in calculation complexity is broken through, unified modeling and global optimization of three types of flexibility are achieved, a solution set covering all the optimal schemes can be efficiently generated through example verification, the method is suitable for complex part machining scenes such as aerospace, the flexibility level of a production system is improved, and meanwhile the production efficiency is improved. Decision support is provided for dynamically balancing time and cost, and the method has remarkable technical progress and practical value.

The embodiment of the specification discloses a programmatic advertisement putting method and device based on big data, equipment and medium, which relates to the technical field of Internet, and the method comprises the following steps: receiving advertisement putting requests of a plurality of advertisement putting users through an advertisement transaction platform, collecting basic data and behavior data of a plurality of target users through the advertisement transaction platform and big data collection technology; generating attribute labels corresponding to each target user in advance according to the basic data through a user portrait generation module, and generating operation labels corresponding to each target user according to the behavior data and a preset real-time label algorithm, so as to generate a user portrait of each target user based on the attribute labels and the operation labels; determining a specified putting user among the plurality of target users according to the plurality of advertisement putting requests and the user portrait of each target user through the advertisement transaction platform, and determining a specified putting advertisement among a plurality of to-be-put advertisements, so as to put the specified putting advertisement to the specified putting user.

The application discloses a kind of based on single RGB image and realizes the fusion method of different human reconstruction model, method includes: first, the implicit expression network obtained by two different reconstruction methods is meshed by Marching Cubs algorithm, obtains the three-dimensional human reconstruction model of space and space resolution alignment, the pixel alignment depth map of each reconstruction model is obtained by rendering mode, and the thickness of space z is calculated, and the same area and different area of two models are obtained by thickness map comparison;Using the RGB image and depth image rendered by human model dataset, the depth map network of visible face and invisible face of person is trained, the predicted visible face and invisible face depth map are obtained by using RGB image input network, the z space position is determined and the z space thickness of different models is aligned using thickness scaling algorithm;Finally, the different areas of two models and the same area are interpolated and fused at boundary using edge distance marking algorithm.The application improves reconstruction accuracy.

The application discloses a truck-unmanned aerial vehicle cooperative path planning method and system. The method comprises the following steps: inputting parameter information of a vehicle path problem with an unmanned aerial vehicle considering an unmanned aerial vehicle exchange mode, calling a heuristic algorithm, and outputting an initial upper bound solution scheme; inputting the initial upper bound solution scheme, calling a column generation algorithm, solving a subproblem in the column generation algorithm by using a cooperative path label algorithm, finding a path column by using a heuristic algorithm, and outputting a linear solution scheme; inputting the linear solution scheme, calling a branching strategy and a cut constraint search strategy, adding a branch and a cut constraint after solving the column generation algorithm at each branching node, and outputting an integer solution scheme. According to the scheme, the unmanned aerial vehicle can be exchanged between different trucks without relying on an unmanned aerial vehicle parking point, and the solving efficiency of an accurate algorithm is improved.

The invention discloses a geostationary orbit target detection method based on minimum pixel superposition, and solves the problem of how to accurately and quickly detect a geostationary orbit target. Belongs to the field of space target detection. Comprising the following steps: superposing minimum pixel values of each frame of image shot by a telescope in a staring mode at the same pixel position to generate a superposed image; respectively calculating a pixel mean value and a standard deviation of the superposed image, setting a threshold value based on a Pauta criterion, carrying out threshold value segmentation on the superposed image according to the threshold value, and carrying out connected domain marking by using a connected domain marking algorithm to obtain a binary image without outlier noise; searching a target marking a connected domain in the binary image by using a connected domain marking algorithm; and calculating the pixel coordinates and pixel values of the targets by using a Gaussian centroid positioning method to obtain the centroid coordinates of each target so as to obtain the position of the geostationary orbit target. According to the invention, the apparent motion difference between the geostationary orbit target and the fixed star is amplified, and accurate and rapid detection of the geostationary orbit target is realized.

The application discloses a large-scale multi-label text classification method based on label adaptive text representation. The application firstly explores label adaptive representation of the text to effectively process classification performance of head labels and tail labels under large-scale multi-label classification; a pre-trained language model is used to learn a representation pool for the text, so that different labels can focus on different representations to complete correlation discrimination. Considering the characteristics of deep models and long texts, text representation enhancement is proposed to ensure the difference and comprehensiveness of the representations in the pool. Therefore, the application can provide effective discriminative text features for large-scale labels to improve the prediction performance. Compared with current large-scale multi-label algorithms, the application can guarantee the overall classification performance of large-scale multi-labels on the one hand, and guarantee that tail labels can better focus on detailed text features on the other hand, and the performance is better than that of the current most advanced large-scale multi-label algorithm.

The present application relates to a skeleton line intersection region redundancy repairing method based on distance constraint and center replacement, and belongs to the field of image processing.The repairing method comprises: for each foreground pixel point in the skeleton image, extracting its 3*3 neighborhood and temporarily setting the center pixel as a background value, and detecting the intersection point through an 8-neighborhood connected marking algorithm; based on the spatial proximity relationship of the intersection point, realizing the intersection point clustering through the Euclidean distance calculation and the breadth-first search algorithm; performing group screening, ignoring the group containing only a single intersection point, and performing topological reconstruction for the group containing multiple intersection points.The present application accurately identifies the redundant pixel cluster, improves the skeleton line quality, accurately identifies and removes the redundant pixels in the diagonal region through the template matching algorithm, ensures that the skeleton line strictly meets the single-pixel width requirement, and avoids the false end point and the line length measurement error.

The present invention belongs to the field of materials science and discloses a method for constructing a double-connected three-dimensional microstructure dataset with the same boundary. First, determine the boundary type and reconstruct the microstructure with the boundary type as the constraints for the six faces of the cube. Then, select relevant optimization objectives for each boundary type and create an initial dataset under the corresponding constraints. Next, use generative AI for structure generation to expand the attribute space of the microstructure, and use an active learning strategy for iterative expansion of the dataset. Finally, identify and retain the structures with double-connected features through the connected component labeling algorithm, and remove the disconnected structures, thereby constructing a dataset with the same boundary and double-connectedness. The LBBM dataset constructed by the present invention significantly improves the computational efficiency of multi-scale assembly, ensures the boundary connectivity between microstructures and the double-connectedness of the overall multi-scale structure. The material constructed based on this dataset has print manufacturing friendliness and can realize functions such as fluid flow, heat conduction, filtration, and gas exchange.

The application relates to a knowledge representation learning method based on group detection, wherein the method comprises the following steps: determining a target knowledge node and associated knowledge nodes, constructing a knowledge graph based on the target knowledge node and the associated knowledge nodes, processing the knowledge graph according to a preset rule to obtain group detection information corresponding to the target knowledge node, adding different negative sample labels to the associated knowledge nodes based on the group detection information, and applying different penalty weight coefficients to the associated knowledge nodes according to the negative sample labels in the process of knowledge representation learning of an algorithm model, and generating a knowledge representation vector. Through the application, the problem that the distinguishing degree of different entity vectors of the same type is low in knowledge representation learning is solved, and the distinguishing degree of the knowledge representation vector is improved.

The invention discloses a multi-target detection method based on a millimeter-wave radar, and the method comprises the steps: collecting the echo data of multiple targets in a complex scene through the millimeter-wave radar, and carrying out the preprocessing of the echo data, and obtaining the distance-Doppler two-dimensional data to be processed; based on the obtained two-dimensional data to be processed, adopting a CFAR detection algorithm to obtain a coarse estimation CFAR mark detection result; the maximum module value square of each column of pulses on the two-dimensional data to be processed is stored, and a false target interference mark is obtained through dynamic adaptive threshold estimation; and based on the obtained CFAR mark and the false target interference mark, accurate detection of the target is realized through a comprehensive mark. According to the method, a CFAR-based comprehensive marking algorithm is adopted, target detection is realized by dynamically estimating background noise and a self-adaptive threshold value, multi-dimensional information such as distance, speed and time-frequency characteristics is fused, a real target and clutter are effectively distinguished, and the target can be accurately identified and related information of the target can be extracted in a complex scene.

The invention belongs to the field of material science, and discloses a method for constructing a double-connected three-dimensional microstructure data set with the same boundary, which comprises the following steps of: firstly, determining a boundary type and reconstructing a microstructure by taking the boundary type as a constraint of six surfaces of a cube; then, a related optimization target is selected for each boundary type, and an initial data set is created under corresponding constraint conditions; thirdly, carrying out structure generation by adopting a generative AI to expand an attribute space of the microstructure, and carrying out iterative expansion on a data set by adopting an active learning strategy; and finally, identifying and retaining a structure with a dual-connectivity feature through a connected component marking algorithm, and removing a non-connectivity structure, so as to construct a data set with the same boundary and dual connectivity, and through the LBBM data set constructed by the method, the calculation efficiency of multi-scale assembly is remarkably improved, and the accuracy of the multi-scale assembly is improved. The boundary connectivity between the microstructures and the double connectivity of the whole multi-scale structure are ensured, the material constructed based on the data set has printing and manufacturing friendliness, and the functions of fluid flowing, heat conduction, filtering, gas exchange and the like can be achieved.

The invention discloses an image recognition management system and method based on data analysis, and relates to the technical field of image recognition. The system comprises an initial classification module, a labeling information analysis and storage module, an effective feedback index analysis module, an optimal labeling algorithm determination module and a real-time target image storage reminding module. The initial classification module is used for performing initial classification of target images on any image data set comprising a plurality of images; the annotation information analysis and storage module is used for storing the associated image into a database corresponding to the identification keyword based on the annotation information; the effective feedback index analysis module is used for analyzing an effective feedback index of each type of target image corresponding to each type of labeling algorithm; the optimal annotation algorithm determination module is used for determining the optimal annotation algorithm of each category of target images, and the real-time target image storage reminding module is used for analyzing the optimal image category of the real-time target images and carrying out annotation algorithm storage reminding according to the optimal annotation algorithm corresponding to the optimal image category.

The present application provides a message testing method, device, electronic device, and computer-readable storage medium, which relate to the field of network communication supervision. The method includes: configuring multiple types of test data corresponding to multiple expected marking results according to a marking algorithm; marking the test data based on the marking algorithm to obtain a test marking result; determining the test result of the marking algorithm according to the expected marking result and the test marking result; wherein the test data includes at least one tested message and token bucket quantity data used to mark the tested message. The present application can mark and verify multiple types of test data under multiple expected marking results of different colors corresponding to the marking algorithm from a verification perspective based on the implementation logic of the marking algorithm, thereby effectively testing the marking accuracy of the marking algorithm, improving the purposefulness and accuracy of the test, and further improving the accuracy of the marking algorithm in marking message traffic, thereby achieving effective supervision of message traffic in the network.

The invention relates to the field of image processing, provides a micro-expression analysis method and system based on an adaptive pseudo mark and an attention mechanism, and aims to solve the problem of large pseudo mark error in an existing method by designing an adaptive pseudo mark algorithm and a micro-expression analysis network based on a fusion attention mechanism. According to the self-adaptive pseudo-marking algorithm, self-adaptive marking is carried out by combining a sliding window based on the proportion of the micro-expressions in the whole video interval, so that marking errors are reduced, the micro-expression analysis accuracy is improved, and the method can be used for solving the problem that subtle changes of the micro-expressions are difficult to accurately capture in an existing method. According to the method, a micro-expression analysis network based on a fusion attention mechanism is designed to seamlessly complete micro-expression positioning and recognition tasks, the micro-expression analysis network adopts a double-layer convolutional neural network to extract features, a double attention mechanism is introduced, channel and space information is captured respectively, significant features are focused, and the positioning accuracy of the micro-expression is improved. The accuracy of micro-expression positioning and recognition is improved, and the accuracy of micro-expression analysis is improved.

A kind of region adaptive stripe projection three-dimensional measurement method of spectral separation, comprising: the sinusoidal phase shift stripe pattern of transverse and longitudinal is projected to the surface of multiple measured objects, and the stripe sequence is collected with camera;Respectively, object contour mask and saturation mask are obtained;The white region coordinates of contour mask are converted to projector coordinates, and projection mask is obtained;Single object projection mask image is obtained using connected domain marking algorithm;Using the region marked and saturation mask, set two groups of mask region to generate longitudinal sinusoidal phase shift stripe pattern of adaptive intensity, and two groups of region stripe are projected to the surface of measured object in turn, while collecting spectral separation region stripe image;The stripe pattern of different channel color corresponding to the object is taken independently phase demodulation and phase fusion, and three-dimensional reconstruction is carried out according to the phase result after fusion;The present application significantly improves the measurement efficiency under the premise of ensuring to eliminate stripe aliasing interference.

Systems, methods, and apparatuses are described for secure transfer of tokenized data from a sender to a recipient without disclosing tokenization schemas of either party. A computing device may receive a detokenization algorithm associated with a sender and tokenized data. The computing device may generate plain data by processing the first tokenized data using the detokenization algorithm. The computing device may clear memory, receive a tokenization algorithm associated with a recipient, and generate tokenized data by processing the plain data using the tokenization algorithm. The computing device may then send the tokenized data to the recipient.

Systems, methods, and apparatuses are described for secure transfer of tokenized data from a sender to a recipient without disclosing tokenization schemas of either party. A computing device may receive a detokenization algorithm associated with a sender and tokenized data. The computing device may generate plain data by processing the first tokenized data using the detokenization algorithm. The computing device may clear memory, receive a tokenization algorithm associated with a recipient, and generate tokenized data by processing the plain data using the tokenization algorithm. The computing device may then send the tokenized data to the recipient.