33 results about "Search computing" patented technology

A material search computing system generates a joint feature comparison space by combining joint image-text features of surface material data objects. The joint feature comparison space is a consistent comparison space. The material search computing system extracts a query joint feature set from a query data object that includes text data or image data. In addition, the material search computing system compares the query joint feature set to the joint image-text features included in the joint feature comparison space. Based on the comparison, the material search computing system identifies a result joint feature set and associated result surface material data objects. The material search computing system generates material query result data describing the result surface material data objects, and provides the material query result data to an additional computing system.

The invention relates to the technical field of trajectory planning, and provides a rapid evaluation and search method and system for an optimal trajectory and terminal equipment, and the method comprises the steps: sampling a plurality of state points in a pre-planning space, and carrying out the pre-collision detection of the state points to remove points conflicting with an obstacle, and obtaining effective sampling points; performing cost evaluation on the effective sampling points, wherein the cost evaluation comprises heuristic cost based on historical planning trajectory information and static cost based on motion features; and iteratively searching from the initial state point based on the result of the cost evaluation: calculating the trajectory cost of the neighborhood state point and updating the search path along the gradient direction until the optimal feasible trajectory is output. According to the method, invalid sampling points are screened out through pre-collision detection, a gradient iteration search mechanism is adopted, and the track search efficiency is improved.

A device, method, and computer readable medium for automating processes in digital environments. Illustratively, the method includes providing an instruction to implement at least one operation on a target computing platform. The method includes executing one or more commands to search for one or more locations of the target computing platform to determine a target property. The one or more commands search based on at least one pre-configured parameter of a plurality of pre-configured parameters for searching computing platforms. The method includes determining (e.g., using machine learning) at least one automated command to execute based on the determined target property of the target computing platform and the at least one operation. The at least one automated command is selected from an automated command repository for detecting anomalies in digital environments. The method includes executing the determined at least one automated command to detect anomalies.

A material search computing system generates a joint feature comparison space by combining joint image-text features of surface material data objects. The joint feature comparison space is a consistent comparison space. The material search computing system extracts a query joint feature set from a query data object that includes text data or image data. In addition, the material search computing system compares the query joint feature set to the joint image-text features included in the joint feature comparison space. Based on the comparison, the material search computing system identifies a result joint feature set and associated result surface material data objects. The material search computing system generates material query result data describing the result surface material data objects, and provides the material query result data to an additional computing system.

A computer system for indexing custom-made coatings within a database may include processors and computer-readable media having stored thereon executable instructions. In some examples, the executable instructions may, when executed by the processors, cause the computer system to maintain a primary database including characterization data for a set of coatings (e.g., first-party coatings). The system may receive, from a user, characterization data or identifier data of a custom-made coating (e.g., a third-party coating) and may store the data in a secondary database. The computer system may then index the secondary database with the primary database such that the user may search both the primary database and the secondary database simultaneously. Accordingly, if the user performs a search, the computer system may return both first-party and third-party coatings. Such a system may provide benefits such as a more efficient workflow or a reduced search time.

The application relates to the technical field of trajectory planning, and proposes a fast evaluation search method, a system and a terminal device of an optimal trajectory. The fast evaluation search method comprises the following steps: sampling a plurality of state points in a pre-planning space, performing pre-collision detection on the state points to remove points in conflict with obstacles, and obtaining effective sampling points; performing cost evaluation on the effective sampling points, wherein the cost evaluation comprises heuristic cost based on historical planning trajectory information and static cost based on motion characteristics; based on the result of the cost evaluation, iteratively searching from an initial state point: calculating the trajectory cost of a neighborhood state point and updating a search path in a gradient direction until an optimal feasible trajectory is output. The application improves the search efficiency of the trajectory by pre-collision detection to screen out invalid sampling points and a gradient iterative search mechanism.

The invention relates to the field of algorithms, in particular to a Viterbi algorithm searching method and device, a medium and a program product, and the method comprises the steps: respectively calculating the cost from each state at an n moment to a state 1 at an (n + 1) moment, and obtaining an optimal preorder state; calculating the cost from the state 1 at the n moment to the state 2 to k at the n + 1 moment; respectively calculating the cost from the state 2 at the n moment to the state 2 at the k to the n + 1 moment to obtain an optimal preorder state; respectively calculating the cost from the state 2 at the n moment to the state 3 at the n + 1 moment to the k; and so on, the optimal preorder state of all the states at the n + 1 moment is found. According to the search method, the number of times of search calculation is not increased, but the memory access mode and the cache efficiency are greatly improved, and an experimental structure shows that the calculation time of a new search order is reduced by 30%.

A device, method, and computer readable medium for automating processes in digital environments. Illustratively, the method includes providing an instruction to implement at least one operation on a target computing platform. The method includes executing one or more commands to search for one or more locations of the target computing platform to determine a target property. The one or more commands search based on at least one pre-configured parameter of a plurality of pre-configured parameters for searching computing platforms. The method includes determining (e.g., using machine learning) at least one automated command to execute based on the determined target property of the target computing platform and the at least one operation. The at least one automated command is selected from an automated command repository for detecting anomalies in digital environments. The method includes executing the determined at least one automated command to detect anomalies.

A cybersecurity service (“service”) obtains alerts indicating security policy violations for assets in a computing environment and the corresponding issue category(ies) associated with each violation. For each alert, the service determines which asset in the computing environment to target for remediation of the associated violation by searching a graph representation of the computing environment based on the affected asset's type and / or the issue category. The service identifies a target asset and other assets related to the affected asset and the target asset as a result of searching the graph and generates a remediation plan indicating actions to take on the target asset to remediate the security policy violation for the affected asset based on the target asset type and the corresponding issue category. The service indicates the remediation plan, the related assets identified due to the graph traversal, and their corresponding security policy violations within the same issue category.

The application discloses a satellite signal lock loss relocation method, system, device and storage medium, the method comprises the following steps: taking the carrier NCO value and code NCO value before lock loss as the center, a two-dimensional search range is constructed; searching in the two-dimensional search range, calculating the non-coherent accumulation value of each grid point, and finding the maximum non-coherent accumulation value; according to the size of the maximum non-coherent accumulation value, judging the peak interval to which the maximum non-coherent accumulation value belongs; when the maximum non-coherent accumulation value belongs to the first peak interval, a traditional frequency-locked loop reentry locking mode is adopted; when the maximum non-coherent accumulation value belongs to the second peak interval, a high-sensitivity reentry locking mode is adopted. The application improves the lock loss relocation sensitivity under weak signals, effectively guarantees the availability of the receiver under weak signals, and solves the problem of lock loss relocation under weak signals.

The invention discloses a step-by-step spatial scale search-based water surface vehicle route planning navigation method, which comprises the following steps: S1, constructing a rasterized electronic chart containing water depth information, and marking non-navigable data influencing the navigation of a water surface vehicle on the electronic chart; s2, a task target from the navigation starting point to the navigation ending point of the water surface vehicle is set in the electronic chart, m layers of sub-grid masks with sequentially reduced grid space scales are constructed on the electronic chart, the front m-1 layers of sub-grid masks sequentially execute grid route chain search of the front and rear layers of accumulated heuristic task targets, and the (m-1) th layer of sub-grid mask outputs a route chain; and S3, performing optimal path search and optimization on the mth layer of sub-grid mask by using the route chain to obtain a water surface vehicle planning route. According to the method, search guidance of front and rear layer accumulated heuristic task targets is realized, and a refined route chain is obtained through quick search of a multi-layer sub-grid mask, so that the search calculation amount and complexity are reduced, and the water surface navigation path planning efficiency is improved.

The application discloses a kind of quick rejection methods of isolated water point in marine numerical model grid, belong to marine numerical simulation field.For the problem of long computing time when rejecting isolated water point by traditional depth-first algorithm or breadth-first algorithm, the application finds the nearest connectability calculation area of starting calculation unit before calculating the connectability of starting calculation unit and final calculation unit, and completes the connectability test based on the core idea of greedy algorithm, so as to greatly reduce the search time.The method of the application can quickly locate all isolated water points in the terrain data matrix, and finds the optimal adjacent non-isolated water point as the search end point before searching the connectability of the calculation unit, avoiding the defect of excessive calculation amount and calculation time caused by the fixed search end point in the traditional search algorithm.The test algorithm is suitable for all marine numerical model grids, and the calculation time is greatly reduced.

The present invention discloses a method for compiling a station yard map with integrated driving instructions, comprising the following steps: S1, generating common control combinations and association patterns to form a control usage behavior model; S2, inputting the control usage behavior model into a graph neural network to generate intelligent layout suggestions; S3, automatically laying out the controls using a force-directed algorithm based on the intelligent layout suggestions to generate a preliminary station yard map layout; S4, constructing a finite state machine for the controls based on the preliminary station yard map layout to enable the controls to achieve dynamic behavior response; S5, generating a station yard map with dynamic response capabilities based on the dynamic control behavior; S6, using simulated annealing tree search to calculate the optimal route path, and combining it with a constraint propagation algorithm to perform path conflict detection and adjustment; S7, performing image processing and feature calibration in an augmented reality environment to provide an intuitive path and station yard view. The present invention utilizes graph neural networks and simulated annealing tree search, etc., to achieve path optimization and station yard map compilation.

A logistics network planning method and device, relating to the field of warehousing logistics technology. The method comprises: obtaining logistics node configuration information and neighborhood configuration information; establishing a target model according to the logistics node configuration information and setting constraints; the target model is used to characterize the correspondence between planning schemes and planning features, wherein any planning scheme includes part or all of a plurality of candidate logistics nodes, and the planning features include: investment cost and / or delivery delay; according to the domain configuration information, a variable neighborhood search algorithm is used to perform at least one local search calculation on the target model to obtain a target planning scheme; wherein the target planning scheme includes a target logistics node among a plurality of candidate logistics nodes, and the target investment cost in the target planning feature corresponding to the target planning scheme is lower than the cost threshold, and / or the target delivery delay is lower than the delay threshold. The method is helpful to reasonably plan the logistics network.

A method and system provide for searching a computer-aided design (CAD) drawing. A CAD drawing is obtained and includes vector based geometric entities. For each entity, primitives are extracted. A feature coordinate system is created for each of the entities using the primitives. The primitives are transformed from a world coordinate system to feature coordinates of a feature coordinate system. Geometry data of the transformed entities is encoded into index codes that are utilized in an index table as keys. A target geometric entity is identified and a target index code is determined and used to query the index table to identify instances of the target geometric entity in the CAD drawing.

The invention discloses a satellite signal lock loss repositioning method, system and device and a storage medium, and the method comprises the steps: taking a carrier NCO value and a code NCO value before lock loss as a center, and constructing a two-dimensional search range; searching in a two-dimensional searching range, calculating an incoherent accumulated value of each grid point, and finding out a maximum incoherent accumulated value; according to the size of the maximum incoherent accumulated value, judging the peak interval to which the maximum incoherent accumulated value belongs; when the maximum incoherent accumulated value belongs to a first peak value interval, a traditional frequency-locked loop re-locking mode is adopted; and when the maximum incoherent accumulated value belongs to a second peak value interval, adopting a high-sensitivity re-locking mode. According to the invention, the lock loss relocation sensitivity under the weak signal is improved, the availability of the receiver under the weak signal is effectively guaranteed, and the problem of lock loss relocation under the weak signal is solved.

The invention discloses a ticket price search configuration method and system, electronic equipment and a storage medium, and the method comprises the steps: obtaining a ticket price search request, analyzing the ticket price search request into a trial calculation request and a user request, accessing a database based on the trial calculation request and extracting first configuration data, accessing a local cache based on the user request and extracting second configuration data, and calling a search calculation engine based on the first configuration data and / or the second configuration data and returning a ticket price search result. According to the method and the device, the trial calculation request and the user request are distinguished, the trial calculation verification configuration data and the online search configuration data are distinguished, in the ticket price search process, online real-time access is provided by utilizing cache data, and trial calculation search is performed by utilizing data in a database, so that online real-time configuration of the search configuration data can be realized, and the search efficiency is improved. Meanwhile, real-time verification and trial calculation are conducted on the configuration effect, search of online users is not affected, consumption of search resources is reduced, and search efficiency and user experience are improved.

A solution arranged to build or train a machine learning model and to upload the machine learning model to a server arranged to deploy the machine learning model to a plurality of communicating devices. The solution can include a machine learning model builder arranged to build the machine learning model and a machine learning production pipeline. The machine learning production pipeline can be arranged to train the machine learning model, convert the machine learning model to a web browser compatible format, and upload the converted machine learning model to the server. The machine learning model can be arranged to receive as input a sequence of one or more prior activities on one communicating device in the plurality of communicating devices, analyze the sequence of one or more prior activities on said one communicating device, predict a next activity on said one communicating device based on the analysis of the sequence of one or more prior activities, preemptively search a computer network based on the predicted next activity to find a computer asset, and preload the found computer asset to a storage in said one communicating device.

The invention relates to the technical field of satellite network security, and discloses a satellite internet node risk association analysis-oriented mining method and device, and the method comprises the steps: S1, obtaining an access record of a satellite internet node; s2, scanning an access record to obtain an access operation attribute set, selecting a plurality of attributes from the access operation attribute set as particles of a particle swarm optimization algorithm, performing search calculation on the particles through the particle swarm optimization algorithm to obtain the particles with the maximum fitness value, and taking the positions of the particles as optimal threshold values; s3, performing descending sorting on the access operation attributes with the attribute frequency greater than or equal to the optimal threshold in the access operation attribute set according to the attribute frequency, constructing a frequent pattern tree by the access operations in the access operation attribute set according to the sorting sequence, and obtaining the frequent access operations through a confidence coefficient formula after obtaining all the frequent access operations by utilizing the constructed frequent pattern tree. And determining a risk association rule between the access operations. According to the method, the satellite internet node risk association analysis efficiency is improved.

The invention discloses an RSA password decomposition method and system based on dynamic search. The method comprises the steps that parameters are defined; constructing a rectangular area according to the defined parameters; constructing a dynamic distributed parallel search calculation network and implementing search calculation on the rectangular area; dividing the rectangular region into independent sub-regions; dividing independent calculation blocks according to the cores participating in calculation; dividing the independent calculation block into a plurality of independent calculation units, and performing random sampling on ice by adopting an LCG sampling method so as to generate a search calculation object; and when any core or process finds a required result, the task is completed. The system comprises a memory and a processor used for executing the RSA password decomposition method based on the dynamic search. By using the method and the device, the technical problem that the integer decomposition difficulty in the existing RSA password decomposition method is relatively high is solved. The method can be widely applied to the technical field of cryptography.

The invention relates to the technical field of point cloud registration and splicing, and discloses a rapid color point cloud feature histogram registration method. According to the technical scheme, a color point cloud to be registered is obtained, a calculation radius is set, and a point cloud normal is calculated; calculating geometric features of the point clouds, and obtaining key points of the source point clouds and the target point clouds; performing neighborhood search on the key points, calculating geometric and color features of the key points, and constructing a color fast feature histogram of the key points; performing feature matching to obtain an initial feature matching pair; obtaining an optimal transformation matrix by using a random sampling consistency algorithm; and the obtained optimal transformation matrix is used as an initial transformation matrix to complete coarse registration of the source point cloud and the target point cloud, and an iterative nearest point algorithm is used to complete fine registration. The method has the beneficial effects that through the steps, the purpose of point cloud fine registration can be achieved. The method has the advantages of high speed, small error, high efficiency and the like. And an important technical support is provided for dynamic data acquisition in a tunnel construction process.

The invention provides a vector search system for a vector database. A cluster search module and a vector search module receive query vectors input by an application; data transmission is carried out between the cluster search module and the vector capture module, the cluster search module is used for realizing cluster sampling calculation of the FPGA hardware processor, and the vector capture module is used for realizing end-to-end access of the FPGA hardware processor; data transmission is carried out between the vector search module and the global retrieval module, the vector search module is used for realizing vector search calculation of the FPGA hardware processor, and the global retrieval module is used for realizing sorting calculation of the FPGA hardware processor; and the software runtime module is used for assisting the vector capturing module to realize end-to-end access. The end-to-end approximate nearest neighbor search speed of the system is 1.41-171.28 times faster than that of an existing system, and meanwhile, more than 90% of CPU computing resources are saved.

The invention discloses a two-step mobile search reachability calculation method fusing aspect-level sentiment analysis, and relates to the technical field of geographic information. According to the technical scheme, the method comprises the steps of obtaining public facility data, population data, traffic data and a public facility evaluation text of a user, and preprocessing the data; performing fine tuning on the general sentiment analysis model, constructing a domain-specific model, performing intelligent selection between the general model and the specific model through a dynamic confidence routing algorithm, and outputting a sentiment score; the service capability of the public facilities is evaluated by adopting a range standardization method and an information entropy method in combination with the scale and other attributes of the public facilities; and calculating the accessibility of the public facilities in combination with traffic network data and population space distribution multi-source data. The method has the beneficial effects that the limitation that traditional reachability analysis only depends on spatial distance and facility scale is broken through, subjective feeling of a user is brought into a quantitative evaluation system, organic fusion of objective indexes and subjective evaluation is realized, and a new analysis framework is provided for urban public service facility planning.

Disclosed are methods and systems for: receiving raw geoscience or energy domain data associated with a resource site, such that the raw geoscience or energy domain data comprises a plurality of textual data and image data having a plurality of disparate file / document formats; receiving a user input associated with the raw geoscience or energy domain data; applying the raw geoscience and energy domain data and the user input to a configured embeddings model thereby generating one or more of: a text embedding associated with the raw geoscience or energy domain data, and an image embedding associated with the raw geoscience or energy domain data; implementing, based on the applying, one or more of: a semantic search computing operation and a classification or clustering computing operation associated with a multidimensional vector space; and generating a report, based at least on the semantic search / classification / clustering computing operation.

The application discloses a kind of layer-by-layer cumulative heuristic guidance ground vehicle path planning method and system, its method includes: S1, the path planning model of containing road network map is constructed, obstacle data that influences ground vehicle driving is marked on road network map;S2, m-layer subgrid mask is constructed according to upper and lower dimensions on the road network map of path planning model, the front m-1 layer subgrid mask of road network map is executed task target in turn and front and rear layer cumulative heuristic grid route chain search;S3, path planning model uses route chain to carry out optimal path search and optimization on road network map and obtains ground vehicle driving planning path.The path planning model of the application is from macro route to fine route layer-by-layer cumulative search guidance and obtains route chain, realizes the search guidance of front and rear layer cumulative heuristic task target and obtains fine route chain by multilayer subgrid mask fast search, reduces search calculation amount and complexity, improves vehicle driving path planning efficiency.

This document discloses devices and methods for implementing semantic search on SSDs using a computational SSD system that distributes computation across each NAND flash die of the SSD. The SSD controller handles result aggregation with novel on-die computational logic circuits to provide semantic file search on the device. The computational SSD system can read file feature vectors from multiple dies into the SSD controller. If necessary, these feature vectors can be buffered in DRAM, and the controller handles the distance calculation. Local, on-die AI / ML processing units can, for example, perform computational and comparison operations and pass the processing evaluations and results to the SSD controller. The SSD controller aggregates the results from all dies and returns the result to the host.The feature vector memory size, switching logic, and number of on-die AI / ML processing units can be configured as needed to adapt to different tasks, system constraints, and / or feature vector sizes.

The present disclosure provides a key information extraction method, apparatus, electronic device, and storage medium, relating to the fields of computer technology, particularly intelligent search, computer vision, deep learning, and other technical fields. The specific implementation scheme comprises: obtaining optical character recognition results for an image and obtaining target field information to be extracted; determining a target pipeline corresponding to the image file type from among multiple pre-configured pipelines corresponding to file types; and processing the optical character recognition results using the operators included in the target pipeline to obtain an extraction result corresponding to the target field information.

An approach is provided for browser based robotic process automation. A computing device accepts, via a user interface, a selection of a query and one or more services. The computing device sends a request identifying a user account and the one or more services. The computing device receives, in response to the request, automation code for accessing the one or more services. The computing device retrieves credentials for accessing the one or more services by the user account. The computing device executes the automation code to cause the user account to be authenticated at the one or more services using the credentials and to cause a search of the query to be performed at the one or more services. The computing device outputs a result of the automation code, the result corresponding to one or more matches of the search of the query at the one or more services.

The method belongs to the field of neural network hardware design, an instruction set without multiplication basic operation is constructed as a search space, and high-energy-consumption multiply-accumulate operation is avoided; secondly, designing an evolution search framework based on NSGA-III multi-objective optimization, and dynamically evaluating the comprehensive performance of candidate functions in the search process by taking precision, energy consumption and reliability as joint optimization objectives; then, introducing function equivalence detection and a two-stage precision threshold mechanism, accelerating search convergence and eliminating low-efficiency individuals; and finally, aiming at the forward function obtained by searching, designing a multi-dimensional evaluation method to carry out comprehensive performance quantification on the candidate function, and carrying out neural network operation by using the non-multiplication candidate function. According to the method, on the premise of keeping similar precision, energy consumption is greatly reduced, and the practical value of the neural network in high-reliability scenes such as spaceflight, military affairs and marginal medical treatment is remarkably improved.

The present invention relates to the technical field of communication signal processing, and in particular provides a parallel synchronous search method for low-frequency signals throughout the whole process, including the steps of: acquiring an input signal, acquiring a plurality of local synchronous codes with different rates, performing square loop search calculations, and acquiring corresponding correlation values; acquiring the maximum value of the correlation values greater than the synchronous threshold, and outputting the maximum correlation amplitude corresponding to the correlation value; calculating the synchronous credibility, acquiring the starting point of the received data based on the maximum synchronous credibility and determining the rate of the received data, and receiving data at this rate; judging whether the data reception time has ended, and if not, repeating steps S1-S3 to update the maximum synchronous credibility. The present invention realizes fast and accurate synchronization of a short synchronization header with a low signal-to-noise ratio and automatic rate identification by continuously searching for the synchronous signal through multiple loops and updating the synchronous information, solves the problem of false synchronization / missing synchronization existing in the existing methods, and can still ensure stable and reliable reception of data information in the case of false start.