63 results about "Pair matching" patented technology

The invention discloses a method for extracting a landmark scene abstract, which comprises the following steps of: extracting color moment and ripple lamination of each landmark scene image as global feature of the landmark scene image, and extracting a SIFT descriptor as the local feature of the landmark scene image; performing initial clustering to a landmark scene image library through the global feature according to the two dimension feature of the image; selecting a plurality of representative images closest to the clustering center from each cluster to the initial clustering result through the local feature, and performing pair matching for the representative images for saliency-based geometric verification; after the geometric verification; extracting the landmark image from each cluster aggregate; and performing geometric matching and screening to the landmark image of the area interested by the user in the selected landmark scene images through the geometric verification of clusters, and gathering the landmark images with same or similar visual angles, thus realizing the mixing of the similar clusters, and extracting the landmark scene abstract.

The invention discloses a text interaction matching method and device for financial knowledge questions and answers, belongs to the field of natural language processing, and provides a method for mapping user questions to standard questions in order to accurately judge the matching degree of the financial knowledge questions of a user and the standard financial knowledge questions, and the methodcomprises the following steps: 1, constructing a question pair knowledge base; s2, constructing a question pair matching model training data set; s3, constructing a question pair matching model, comprising the following steps: S301, constructing a character mapping conversion table; s302, constructing an input layer; s303, constructing a character vector mapping layer; s304, constructing a text coding model of an attention mechanism; s305, constructing a text interaction matching layer; and S4, training a question pair matching model and question selections with the intention. (2) The devicecomprises a question pair knowledge base construction unit, a question pair matching model training data set generation unit, a question pair matching model construction unit and a question pair matching model training unit.

The invention discloses a duplex frequency pair matching method based on dynamic mapping. The method comprises the following steps: in the frequency division duplex system for the dynamic distribution of frequency resource, when an uplink and a downlink frequency points are matched into a duplex frequency pair through dynamic selection, a frequency pair with the interval of the central frequency of the selected uplink and downlink frequency points equal to or larger than the minimum duplex interval is selected and listed into the distribution range of the dynamic frequency. By the method, the problems of the prior duplex technology in dynamically dividing frequency, such as the mutual interference of the uplink and the downlink frequency points caused by the use of the duplex interval as the standard for measuring the isolation of the uplink and the downlink, are solved. The method effectively controls the width of a protective band between the uplink and the downlink, realizes relatively flexible matching of frequency pairs, and ensures the dynamically selected uplink and downlink can satisfy the required frequency interval, thus providing more duplex frequency pairs that meet the requirement for isolation.

The invention discloses a question pair matching method and device based on a deep feature fusion neural network, belongs to the field of natural language processing, and aims to solve the technical problem of how to accurately judge the matching degree of a user question and a standard question and sort out a complete question pair matching model. The technical scheme is that the method comprisesthe following steps: S1, constructing a question pair knowledge base; S2, constructing a question pair matching model training data set; S3, constructing a question pair matching model, comprising the following steps: S301, constructing a character mapping conversion table; S302, constructing an input layer; S303, constructing a character vector mapping layer; S304, constructing a neural networkcoding layer based on depth feature fusion; S305, constructing a text similarity matching layer; and S4, training the question pair matching model and selecting a standard problem. The device comprises a question pair knowledge base construction unit, a question pair matching model training data set generation unit, a question pair matching model construction unit and a question pair matching model training unit.

The invention discloses a method suitable for compressing sequence images of an unmanned aerial vehicle. The method comprises the following steps: according to airborne auxiliary data, carrying out intelligent dynamic coding frame group division of the sequence images of the unmanned aerial vehicle; within a frame group, carrying out high-precision quick motion estimation of the first frame and the last frame to the current frame, thereby acquiring a number pair matching point; according to corresponding matching relation, obtaining projection transformation matrixes of the first frame and the last frame to the current frame so as to acquire a forecast image and residual error; combining residual error images of the first frame and the other frames of the same frame group into a singe image; and compressing the singe image by JPEG2000 to acquire a high-performance compression result by a global optimization pruning algorithm. The method takes into account the characteristics of the sequence images of the unmanned aerial vehicle and makes full use of the characteristics of airborne auxiliary information and the JPEG2000 compression algorithm; therefore, the method has the characteristics of low algorithm complexity, good compression performance, strictly controlled compression code rate, easily realized hardware, and the like.

An ontology-based query method and apparatus include acquiring a to-be-queried triple input by the user, where a known element is a query condition. One or more unknown elements in the to-be-queried triple is a query object and searching is performed, in the key-value pairs stored in each of the plurality of computing nodes, for a key-value pair matching the query condition. An element corresponding to the query object is determined from three elements included in a key value of the matched key-value pair, to acquire elements corresponding to the query objects determined in each of the plurality of computing nodes. A query result is acquired according to the elements corresponding to the query objects determined in each of the plurality of computing nodes.

The invention provides a resource allocation method for Multiuser-Multiple Input Multiple Output (MU-MIMO) user pairing. The method comprises the following steps of: A, sorting users to be scheduled according to user priority; B, sequentially pairing the users in the order of user priority from high to low, defining a higher-priority user in each user pair as a paired main user and defining a lower-priority user in each user pair as a paired auxiliary user; C, judging whether paired auxiliary users of the paired main users to be scheduled exist, and putting the paired auxiliary users in an available pairing user queue in the order of pairing matching degree of the paired auxiliary users from high to low if the paired auxiliary users of the paired main users to be scheduled exist; D, carrying out resource allocation for the paired main users and defining resource obtained from allocation as available pairing resource; and E, sequentially allocating the available pairing resource to all the paired auxiliary users in the available pairing user queue in a segmented way.

The invention discloses a visible light broadband spectrum image registration method. The method includes the steps that graying, denoising and downsampling are carried out on a reference image and an image to be registered, feature points of the images are extracted by using an SIFT algorithm, and feature description operators are constructed; the feature points are matched according to a k-d tree nearest neighbor method, and coordinate upsampling is carried out on matched feature points; according to an image degradation threshold T, wrong matching points are restrained by means of an Euclidean distance, and restrained matching point sets omegan are obtained; m pair matching point space uniform sampling is conducted on the restrained matching point sets omegan to obtain matching point sets omegam uniformly distributed; image radiation conversion parameter matrixes Tm are solved by using the least square method, and coordinate conversion and bilinear interpolation calculation are conducted on the image to be registered through the Tm; mutual information MIm of the reference image and a registered image serves as an evaluation index, the registered image corresponding to maximum mutual information MImax is locked, and a visible light broadband spectrum image is registered. According to the method, the number of the wrong matching points can be greatly reduced and image registration precision can be improved.

The invention discloses a block-based Web record linkage system which comprises a Web crawler, a Sample database, a Web record database, a block attribute analysis module, a blocking module, a block balancing module, a paired matching module, a matching determination module and a record linkage result set. According to a block-based Web record linkage method, data from various data sources are quickly blocked by a Mapreduce model, and the data are compared and recorded in the blocks, therefore, the record matching efficiency is improved to a large extent; on the basis, the sizes of blocks are balanced, so that the record matching efficiency is further improved. The recalling rate of record linkage is also improved by adopting a method for blocking a data set with a multi-block function from multiple angles.

The invention provides a text similarity calculation system and method based on multi-keyword pair matching, and a storage medium, for texts P and Q, text similarity calculation is completed by meansof Word2vec, a bidirectional long short-term memory network (BiLSTM), an attention mechanism and a dual-task architecture. According to the method, the influence of noise and redundant data on the model performance is avoided through the WP-Attention and the dual-task architecture, and the model is simple in structure, easy to expand, high in robustness and easy to popularize and use in practice.

The invention discloses a probe vehicle map matching acceleration method for limiting the traveling speed. The shortest path upper bound of an adjacent tracing point traveling in a road network is estimated according to a road segment traveling speed threshold value and a probe vehicle sampling interval in the urban road network, and the shortest path of all node pairs in the road network is previously calculated through the shortest path upper bound; when the candidate road segment pair matching probability of a front tracing point and a rear tracing point in the map matching process of a probe vehicle is estimated, the analysis of the shortest path can be obtained only by inquiring the preprocessing result, so that the time expenditure of the probe vehicle map matching can be greatly reduced; in addition, the node scale of the road network is controlled through the grid segmentation, and the preprocessing time can be completed within a short time.

The invention relates to the field of liquid crystal displaying, in particular to a classified liquid crystal displaying component. The high-privacy liquid crystal displaying component comprises a liquid crystal displaying screen and screen-watching glasses. The liquid crystal displaying screen comprises a backlight module, a lower glass substrate, a TFT liquid crystal module, an upper glass substrate and a horizontal polarizing film which are arranged in the light-moving direction in an overlapping mode. The horizontal polarizing film is attached on a light-out face of the upper glass substrate. The screen-watching glasses comprise vertical polarizing films and spectacle lenses which are arranged on a glasses frame, and the vertical polarizing films are attached on the spectacle lenses and are matched with the horizontal polarizing film. According to the high-privacy liquid crystal displaying component, the polarizing films which are originally matched with the liquid crystal displaying screen are detached and are attached on the spectacle lenses, so that the spectacle lenses and the liquid crystal displaying screen become one set of component, the content of the liquid crystal displaying screen can be seen only during pair matching, displaying screen content revealing can be effectively avoided even in public places, and safety is improved.

The invention discloses a question and answer model training method, computer equipment and a computer readable storage medium. The question and answer model training method comprises the steps of establishing a text matching model and a meta-learning model, wherein the text matching model and the meta-learning model have the same deep neural network structure; obtaining a test task and a plurality of different training tasks; updating meta-network parameters of the meta-learning model based on respective first matching losses of the plurality of different training tasks, wherein the first matching losses are used for representing matching errors of respective text matching models of the plurality of different training tasks; and generating initial parameters of a text matching model by utilizing the updated meta-network parameters, training the text matching model by using the test task to update the initial parameters of the text matching model, and taking the text matching model obtained after training as a question and answer model. By means of the mode, the question and answer model can be applied to the question and answer field with few samples, and the question and answer pair matching accuracy is improved.

The invention discloses a knowledge graph construction method and a related device. The method comprises the following steps: obtaining corpus data; carrying out entity pair matching on a plurality ofstatements in the corpus data by utilizing a preset entity pair relation to obtain a matching result; utilizing the training learning model to train statements containing entity positive examples andstatements containing entity negative examples to obtain a training result, and the training result at least comprises weight values of words in the statements in the statements; generating a relation template based on the words with the weight values meeting a first preset condition; based on the positive example statement and the probability value thereof, and the negative example statement andthe probability value thereof, calculating a probability value of the relation template; determining a target relationship template of which the probability value meets a third preset condition in the relationship template, and calculating a proportion value of a corresponding entity pair relationship for the entity pair in the statement by utilizing the target relationship template; and selecting an entity pair with the proportional value of the entity pair relationship meeting a fourth preset condition to construct a knowledge graph.

The invention discloses an information pair matching method and system on a small sample set, and the method comprises the steps: inputting a text, and recognizing attributes and attribute corresponding values in the text; constructing attributes appearing in the statements and features of attribute corresponding values in pairs; constructing a training sample; training a model based on the training sample; and constructing an information pair matching model. According to the method, feature construction is carried out on the annotation set of the small sample based on the syntactic path, datamatching is carried out through a machine learning method, a model with high matching accuracy can be efficiently obtained, and structural information in the document is obtained.

The invention provides a prefabricated part point cloud automatic mold closing method based on particle swarm optimization variable domain search matching points. The prefabricated part point cloud automatic mold closing method comprises the steps: inputting prefabricated part point cloud, retrieving a corresponding BIM model, discretely extracting contour point cloud data by the BIM model, re-sampling, roughly matching and judging, grading important parts of the BIM model, searching and pairing matching points, optimizing a spatial position and the like. The invention provides a technical method for automatically carrying out mold closing by the prefabricated part point cloud and the BIM design model, effectively solving the problem that a large amount of human intervention is needed forassembling the prefabricated part point cloud and the BIM design model during quality detection, and providing method support for later size data estimation.

The invention provides a short text matching method and system based on global and local matching, being characterized by constructing a matching graph, representing the whole short text pair as a graph node based on a pre-trained language model, carrying out cross-pair matching learning on the basis of graph attention network node classification, statically combining local matching and global matching through connection weights of graph attention network nodes, and dynamically combining the local matching and the global matching through joint training and iterative updating, so that local andglobal advantages are better exerted, and the short text matching method and system are suitable for different short text matching tasks.

The invention relates to application of a laser heat treatment technology in a cylinder sleeve of a marine diesel engine, in particular to application of a laser heat treatment technology in the configuration of a large-scale ocean marine diesel engine, that is a laser irradiation quenching technology is adopted to carry out heat treatment on the inner furnace of the cylinder sleeve of the marine diesel engineer so that the surface of the inner wall of the cylinder sleeve obtains a hardening layer with certain hardness and depth and improves the wearing resistance and the anti-cavitation property of the inner wall of an oil cylinder; the deformation amount of laser treatment is between 1/20 and 1/30 of the deformation amount of the prior heat treatment; the deformation amount is small, basically keeps the original state of the cylinder sleeve before quenching and has good sealing and pair matching performances with a piston ring; the largest hardness value of laser quenching is HRC66 and is higher than the hardness value HRC54 after the prior quenching; the depth of a quenching layer reaches between 3 and 3.5 millimeters, and the width is between 2 and 2.5 millimeters; therefore, the cylinder sleeve reaches the requirements on high wearing resistance and anti-cavitation property.

A data storage system a process including:

• • storing an array of registered entity property-value pairs representing the registered entity where each of the pairs includes a property label representing a property and a value range of the property;
  • storing an entry of one identifier property-value pair where each pair includes an identifier property label uniquely identifying the registered entity, and an identifier value representing a value of the identifier property;
  • acquiring a dataset having acquired dataset property-value pairs for each of a first set of acquired dataset entities, where each pair includes a property label representing a property and a value representing a value range of the property;
  • identifying an identifier property-value pair matching an acquired dataset property-value pair for the acquired dataset entity; and
  • consolidating the acquired dataset property-value pairs into the array of registered entity property-value pairs.

The invention belongs to the technical field of aviation mechanical machining technology devices, and provides a floating supporting and pressing device. The floating supporting and pressing device ischaracterized in that a spherical high-pair matching is formed by a universal inner spherical surface floating head and a thread-adjustable outer spherical surface threaded supporting rod, and angleadjustment in a 5-degree range is achieved. The floating supporting and pressing device has the beneficial effects that by using the mode of changing the technology state of products, the floating supporting and pressing device can achieve pressing of the products on the technology devices only through one bolt, and the machining surface is guaranteed to be completely opened; and the problems thatthe pressing position of a conventional pressing plate is not machined and the pressing plate is moved in the machining process are solved, cutting of a machine tool is continuous carried out, so that continuous machining is realized, and the utilization rate of the machine tool is improved.

The invention discloses a multiple-master-with-multiple-slave Bluetooth communication method for a portable health examination box. The communication method comprises the following steps that: 1, mater communication equipment and slave communication equipment are designated and preset; 2, a master equipment Bluetooth module and slave equipment Bluetooth modules are powered up simultaneously and the master equipment Bluetooth module and the slave equipment Bluetooth modules enter a pair matching mode; 3, the master equipment Bluetooth module searches for the slave equipment Bluetooth modules and connects with the slave equipment Bluetooth modules; 4, the connection in the step 3 is interrupted and a far-end Bluetooth address is obtained; 5, the step 3 and the step 4 are repeated till Bluetooth addresses of all the slave equipment Bluetooth modules are obtained and backuped; 6, the master equipment Bluetooth module is connected with the slave equipment Bluetooth modules according to the addresses; 7, the steps from 1 to 6 are repeated, the master equipment Bluetooth module is set into a slave equipment Bluetooth module, a new master equipment Bluetooth module is designated till the master equipment Bluetooth module designated for each time acquires and saves corresponding addresses of the slave equipment Bluetooth modules in need of connections, and the multiple-master-with-multiple-slave Bluetooth communication mode is achieved. The multiple-master-with-multiple-slave Bluetooth communication method for the portable health examination box has the advantages that the usage is convenient, accurate and efficient, and no connecting wire is needed.

The invention provides an entity pair matching method and device in a database, electronic equipment and a storage medium, and the method comprises the steps: obtaining two to-be-matched target entities in the database, and respectively removing attribute tags from the two entities to obtain corresponding text sequences; inputting the text sequence into a vector representation learning model to obtain vector representation of the to-be-matched entity; calculating the similarity of the vector representation; and determining whether the two target entities are matched according to a difference value between the similarity between the vector representations of the two target entities and a threshold value. The BERT model is trained according to the comparison loss function of the entity pair and the comparison loss function of the entity attribute, so that the problem that the vector representation cannot reflect the characteristics of different attributes is avoided, the accuracy of the vector representation pair is improved, and the accuracy of a matching result is improved. The process of repeated judgment in the prior art is avoided, and the entity matching efficiency is improved.

The invention provides a pair matching device used for a rolling friction experiment of a ceramic ball, and a use method thereof. A left side retainer, a right side retainer, the ceramic ball, a block test piece and a disk test sample form one group of block-ball-disk pair matching devices, and meanwhile, a distance between the left side retainer and an installation chuck, a distance between the left side retainer and the right side retainer and the length of the block test sample are regulated to meet the clamping of the ceramic balls of different diameters in a certain diameter range. Meanwhile, according to V-shaped grooves arranged on the left side retainer and the right side retainer, an axial dimension is prevented from being regulated. The pair matching device has the characteristics of compact structure, simpleness in operation, stability and reliability in operation and wide applicable range, and has good applicability for researching the tribology behavior of a material under the mixed state of rolling and sliding.

The invention discloses a movable wrench convenient to use. The movable wrench comprises a handle and a wrench head. The wrench head which is integrally formed is arranged at the front end of the handle. A fixed clamp is arranged at the front end of the wrench head. A movable clamp is installed in the corresponding position on the inner side of the fixed clamp. A movable channel is formed in the position, corresponding to the movable clamp, of the wrench head. The tail end of the movable clamp is meshed with a driven gear through a rack. The other end of the driven gear is meshed with a driving gear. According to the movable wrench convenient to use, due to the gear pair matching mode, the aim of pushing the movable clamp is achieved. Meanwhile, the rotary handle is installed on the top of the driving gear so that the movable wrench can be operated more easily during use, and objects needing to be fixed can be stably clamped. After the wrench rusts, the situation that a rotary screw rusts and is locked in the prior art can be avoided by adding lubricating oil, and an anti-slip rubber sleeve can prevent accidents from being generated during work.

The invention relates to a translation auxiliary system based on a mobile communication network. The system comprises a device for converting to-be-translated information of a target language sent byan opposite-end user into a translated text of a source language and converting to-be-translated information of the source language input by a local-end user into a translated text of the target language by at least two instant messaging terminals which are separated from each other at a geographic position in an instant messaging state. According to the device for graphically displaying the translation of the source language, the translation of the target language and the response information of the source language for responding to the translation of the source language, not only can instanttranslation of different languages be realized, but also automatic reply can be realized, and the translation efficiency and accuracy are improved in a statement pair matching manner.