39results about How to "Reduce redundant information" patented technology

The invention provides a riverbed evolution prediction method based on multi-source data fusion, and belongs to the technical field of hydrological resource monitoring. The method comprises the following steps: acquiring and fusing multi-source data by using different devices, performing data preprocessing to form unified three-dimensional grid data, and then generating a multi-scale feature map based on a designed feature extraction network; based on the designed multi-channel convolutional neural network, learning and extracting features on multiple channels at the same time, and performing hierarchical expression and fusion on different information sources; and finally, based on the designed time sequence regression network, extracting a complex dependency relationship in the time sequence through a double-layer LSTM structure, learning residual error information through an adaptive residual error correction module to improve prediction precision, and obtaining a final prediction result. The method shows relatively high applicability and robustness in a complex river channel environment, and can cope with various river bed form changes caused by natural factors and human activities through multi-source data fusion and an innovative feature extraction and regression method.

The invention provides a multi-modal sentiment analysis method based on unwrapping representation learning, which comprises the following steps: acquiring a multi-modal sentiment analysis data set, and extracting features of texts, audios and videos; constructing a representation learning network, extracting private features and shared features, and processing through adversarial learning optimization and time smoothness constraint; under the guidance of different language scales, the private features of the audio and the video are fused; further fusing text, audio and video features through cross-modal attention and a gating mechanism; and finally, inputting the fusion features into an emotion analysis module, and generating an emotion prediction result in combination with an emotion classification or regression task. According to the method, the consistency of the audio and video features in continuous time is considered, redundant information is effectively reduced, the time dimension consistency is ensured, deep fusion of the audio and video features is realized under text guidance, and the cross-modal fusion effect is remarkably improved.

The invention relates to an early warning event processing system of an oil and gas pipeline optical fiber early warning system, and the system comprises a preprocessing module which is used for receiving and preprocessing the real-time early warning data of a plurality of early warning events, and obtaining the preprocessed real-time early warning data; the initial merging module is used for determining a first similar early warning event in the plurality of early warning events according to the preprocessed real-time early warning data, and performing initial merging on the real-time early warning data of the first similar early warning event to obtain a merged early warning event; the dynamic upgrading and downgrading processing module is used for acquiring a multi-dimensional feature vector of the combined early warning event; using a deep learning model to adjust the initial early warning level of the combined early warning event according to the multi-dimensional feature vector to obtain an adjusted early warning level; and the output module is used for outputting the combined early warning event and the adjusted early warning level thereof. According to the system, repeated and redundant information can be reduced, and the early warning event processing efficiency, response speed and dynamic flexibility are improved.

The invention discloses a self-adaptive point cloud sampling adjustment method. The method comprises the following steps: training a feature analysis model; sampling a workpiece at an initial sampling rate to obtain point cloud data of the workpiece, inputting the point cloud data into the feature analysis model to obtain regional features, and dividing a local region based on the regional features; and adjusting the sampling rate of the local region based on the region features. The local structure of the point cloud can be reflected more finely based on the division of the regional features, so that the subsequent sampling rate adjustment is more accurate. The refined division mode is beneficial for reserving important features such as edges and angular points in the point cloud, removing redundant information at the same time, and providing a more valuable data basis for subsequent point cloud processing tasks.

The invention discloses an acne grading method, system and equipment based on multi-modal fusion and a medium, belongs to facial acne severity grading in the technical field of artificial intelligence, and aims to solve the technical problem of poor facial acne severity grading effect in the prior art. A constructed facial acne grading model comprises a VGG feature extraction module, a multi-modal feature fusion module, an acne feature sensing module and a comprehensive grading module, the VGG feature extraction module extracts facial features in different modals, and the multi-modal feature fusion module models the relation between the facial features in different modals through cross attention for fusion. The acne feature sensing module enhances the sensing of acne features through a cross attention mechanism, and the comprehensive grading module makes a comprehensive and accurate acne severity grading result. According to the method, comprehensive grading of the severity of the acnes is realized by combining the multi-modal face pictures of the patient, and the grading effect of the severity of the facial acnes is greatly improved.

The invention discloses an important node screening method for an active telemetering scene, and provides a comprehensive evaluation and optimization framework aiming at the problems of probe resource waste and overhigh system load caused by improper node selection in the existing telemetering technology. According to the method, network static topology features and dynamic state features are combined, a node importance scoring model is constructed through feature extraction, dimension reduction fusion and multi-task feature comprehensive analysis, and accurate selection of important network nodes is achieved. According to the method, topological information is analyzed through a weighted TOPSIS method to obtain spatial features, key dimensions of state features in stage historical telemetry data are extracted through principal component analysis (PCA), redundant data are remarkably reduced, and efficient state feature representation is generated. On the basis, a screening method combining spatial features and node state features is designed, and a screening strategy is dynamically adjusted through feature analysis of staged telemetering so as to optimize node priority ranking. The method is suitable for an active telemetering scene in a complex network environment, the capturing capacity of important events can be remarkably improved, system resource consumption is reduced, and the overall performance of an active telemetering system is comprehensively optimized.

The invention discloses a joint optimization method for a joint photographic experts group (JPEG) quantization table and a truncation position based on rate distortion, relates to the technical field of JPEG coding optimization, and solves the technical problems that in a JPEG compression scheme, improvement of coding efficiency caused by joint optimization by using the quantization table and the truncation position is neglected, and the optimal quantization table and the optimal truncation position are difficult to find. According to the method, joint optimization is carried out on a quantization table and a truncation position of the JPEG by combining a JPEG compression method from the perspective of evolutionary computation, a strategy that the truncation position is fixed when the quantization table is optimized, a strategy that the quantization table is unchanged when the truncation position is optimized, an estimation method for designing fitness evaluation and a method based on rate-distortion optimization are respectively carried out, and the quantization table and the truncation position are optimized. The optimal quantization table and the truncation position can be effectively found, and the corresponding truncation position is optimized while multiple groups of quantization tables are generated in the optimization process; through joint optimization of the quantization table and the truncation position, image data can be compressed more effectively, redundant information is reduced, and therefore coding efficiency is improved.

The invention relates to the technical field of crankshaft online detection, in particular to an intelligent crankshaft machining online detection system and method. The method comprises the steps that physical parameter data in the crankshaft machining process are obtained in real time to serve as original signals, and the original signals are converted into electronic signals through a multi-level self-adaptive signal conversion and enhancement algorithm; the electronic signals are preprocessed, and finally collected physical parameter data are obtained; analyzing the finally acquired physical parameter data by using a multi-level self-adaptive hybrid neural network analysis algorithm, and comparing standard specifications to obtain an evaluation result of the crankshaft processing quality; and an adjustment suggestion is generated based on the evaluation result, a machining parameter adjustment instruction of the machining equipment is generated based on the adjustment suggestion, and the machining parameter adjustment instruction is fed back to the machining equipment. The problems that in the crankshaft machining online detection process, crankshaft physical parameters are not accurately obtained, and the detection and analysis accuracy is not high are solved.

The invention provides a cross-modal livestock behavior retrieval method and device, and relates to the technical field of text-video retrieval. The method comprises the following steps: acquiring a livestock video and an input text input by a user, wherein the input text comprises a livestock behavior event; extracting a target sentence and nouns and verbs of the target sentence in the input text; based on a word set corresponding to the livestock behavior retrieval tag, determining a retrieval tag of a noun and a retrieval tag of a verb of the target sentence; searching video features matched with the retrieval tag of the noun, the retrieval tag of the verb and the sentence structure of the target sentence from the video-text multi-channel feature set; and based on the matched video features, retrieving livestock behavior events in the livestock video. Based on the scheme, the accuracy of retrieving the livestock behavior event from the livestock video based on the input text can be improved.

This invention discloses a method for multi-source data fusion and power prediction of wind turbine units, belonging to the field of power prediction technology. It addresses the technical problems of poor multi-source data fusion and power prediction analysis in existing solutions. By filtering and retaining key features through mutual information, redundant information can be effectively reduced. Dynamic weighting via an attention mechanism makes the fused features more adaptable to real-time operating conditions. Differential weight adjustments for wind speed, terrain, and equipment status ensure high relevance of the fused features even in complex scenarios. Spatial feature extraction captures local spatial correlations, temporal features capture temporal dependencies, and random forests handle nonlinear mappings, solving the problem of insufficient generalization ability of single models. The attention mechanism automatically assigns weights to different time steps and features, avoiding complex mathematical processes such as matrix operations and noise covariance estimation.

The invention discloses a 3D printing real model rendering method based on variable granularity, and belongs to the technical field of 3D rendering. The method comprises the steps that 3D printing parameters and the number of layers of a printed object are obtained, and the 3D printing parameters comprise 3D printing head parameters and 3D printer parameters; carrying out single-layer modeling on a printed object on the basis of 3D printing parameters and an interval adjusting strategy; based on a multi-layer fusion strategy, surface elements of each layer of the printed object are obtained; obtaining a body element set of the printed object based on the surface element of each layer of the printed object and the same region principle; based on the body element set of the printed object, constructing a 3D model library; based on the 3D model library and the secondary rendering strategy, the real model of the printed object is rendered, and the method improves the speed and authenticity of 3D rendering.

The application provides a model propagation method and device based on semantic ability, comprising: taking a node needing to update a model and an algorithm as a sending end, and broadcasting original semantic information thereof; each receiving end searches for similar information from a local knowledge base, if the similar information can be found, the original semantic information is input into a model of the receiving end for recovery, recovered semantic information is output and fed back to the sending end, otherwise the original semantic information is directly ignored; the sending end selects recovered semantic information closest to the original semantic information according to a preset evaluation standard, and takes a receiving end corresponding to the recovered semantic information as an optimal receiving end, and the sending end sends a model propagation request to the optimal receiving end; and the optimal receiving end broadcasts a model thereof to the network, and updates the model and the algorithm of other nodes in the network. The method provided by the application can update and optimize a node with long-term un-updated semantic ability, low efficiency and unable to meet user demand in time, and balance semantic ability among nodes in the network.

The invention discloses a low-dimensional vocabulary generation method and evaluation system combining reverse decoding and a GAN network, relates to the technical field of deep learning, and solves the technical problems of information loss, redundant feature retention and lack of flexibility of a dimensionality reduction result in a dimensionality reduction process. The GAN network can capture a complex nonlinear relationship through a deep neural network structure, the embedding expression ability after dimension reduction is significantly improved, and the limitation of directly using a full connection layer dimension reduction method is solved; the discriminator guides the generator to remove redundant features in the data through a feedback mechanism, so that the generated low-dimensional embedded information is more compact, unnecessary redundant information is effectively reduced, and the efficiency of the model is improved; and the reverse mapping network is used for embedding and re-mapping low-dimensional embedding back to high-dimensional embedding, so that the integrity of information in the dimension reduction process is ensured, and meanwhile, the interpretability of the model is enhanced, and the data after dimension reduction can be effectively reconstructed.

The application provides a train warehouse entering control method and device, electronic equipment and storage medium, and belongs to the technical field of rail transit. The method comprises the following steps: in the case that a train management center (TMC) and an object controller (OC) establish a communication connection, determining a target warehouse into which a train enters based on the train management center (TMC) and the object controller (OC), and controlling the operation of a signal device on a route of the train to the target warehouse; acquiring the state of the signal device; and in the case that the signal device is in a passing state, controlling the train to enter the target warehouse. The application can save redundant information of interlocking system interaction, improve the efficiency of train warehouse entering control, meanwhile, without the intervention of dispatchers, the labor cost is reduced, and the intelligent level of train warehouse entering control is further improved.

The application discloses a hybrid automatic repeat request feedback processing method and device and a terminal, and belongs to the technical field of communication. The hybrid automatic repeat request feedback processing method comprises the following steps: in the case that a first hybrid automatic repeat request feedback (HARQ-ACK) and a second HARQ-ACK are in one time unit, and a codebook used by the first HARQ-ACK is a first codebook, a terminal generates HARQ-ACK information of the first HARQ-ACK and the second HARQ-ACK according to the first codebook, or generates HARQ-ACK information according to the priority of the first HARQ-ACK and the priority of the second HARQ-ACK.

The present invention relates to the field of image enhancement technology, and specifically to an image enhancement method, device and electronic device for ultrasound contrast imaging. The present invention analyzes the grayscale of each pixel point at each position in each grayscale uniform window in different video frames, and the corresponding frame sequence distribution, to obtain the overall tissue beating cycle; obtain the periodic video frame in the ultrasound contrast imaging video; use the arterial phase contrast imaging video or the portal phase contrast imaging video as the video to be analyzed, and obtain the overall lesion area possibility of the pixel point at each position according to the grayscale change degree of the pixel point at each position in the corresponding video to be analyzed, and the instantaneousness of the grayscale change; obtain the enhanced grayscale value of the pixel point at each position on each video frame, obtain the enhanced video frame, and obtain the enhanced ultrasound contrast imaging video. The present invention improves the effect of image enhancement by obtaining the accurate grayscale value of each pixel point in each video frame.

The invention discloses a digital circuit mode mining method and system, and the method comprises the steps: S1, extracting logic gate unit information in a netlist, and constructing a directed acyclic graph; s2, traversing all nodes of the directed acyclic graph, and recording modes corresponding to the nodes and the number of the modes by using preset mode codes; s3, combining all the isomorphic modes, and performing descending sorting according to the number of the modes; s4, sequentially traversing the sorted isomorphic mode codes, traversing nodes, marking modes to which the nodes belong, updating logic performance functions of the modes, and rejecting the modes of which the mode node coverage rate is lower than a preset threshold value; s5, comparing the logic functions corresponding to the modes in the mode codes with the known unit logic performance function list one by one, checking whether the logic performance functions are equivalent or not, and recording the logic performance functions of the new modes and the mode codes of the logic performance functions; and S6, sequentially outputting new modes and the number thereof from small to large according to the output number of the modes. The method has the advantages of high digging efficiency and the like.

The application belongs to the technical field of aviation strength test, and particularly relates to an acoustic emission data preprocessing method and device for removing waveforms. The method comprises the following steps: step one, obtaining acoustic emission data files, selecting acoustic emission data files to be preprocessed from the acoustic emission data files, and loading configuration parameters in the acoustic emission data files to be preprocessed; step two, opening a memory space for storing processed data; step three, based on the configuration parameters, the acoustic emission data files to be preprocessed are read in a loop, it is judged whether the acoustic emission data files read in each loop contain waveform information, if yes, the next loop is directly continued, if not, the acoustic emission data file read in the current loop is stored in the memory space and the next loop is continued, until the number of the acoustic emission data files to be preprocessed is zero. The application improves the analysis quality and efficiency of acoustic emission data, and promotes the application of acoustic emission monitoring technology in full-machine fatigue test.

The present application belongs to the field of physiological signal processing and artificial intelligence technology, and specifically relates to an event camera pulse wave detection method based on deep learning, aiming to realize reliable and high-precision non-contact pulse wave detection in complex real scenes. The method comprises collecting event stream data of the radial artery region of the wrist of a subject by using an event camera; wherein the event stream data refers to a collection of event signals containing pixel coordinates, time stamps and polarity information of events; performing a preprocessing operation on the event stream data to obtain a preprocessed count signal; the preprocessing includes region of interest extraction, time binning, detrending and standardization; inputting the preprocessed count signal into a deep learning network to recover the pulse wave signal in an end-to-end manner; using a photoplethysmography signal to supervise the training of the deep learning network, and calculating physiological indicators based on the recovered pulse wave signal.

The invention discloses an object recognition method and device, and the method comprises the steps: collecting the multi-modal original data of a recognition object, and enabling the multi-modal original data to at least comprise image data and audio data; preprocessing the multi-modal original data according to a first preset rule to generate multi-modal standard data; performing feature extraction on the multi-modal standard data according to a second preset rule to generate multi-modal feature data; performing fusion operation on the multi-modal feature data according to a third preset rule to generate multi-modal fusion data; and determining an object recognition result according to the multi-modal fusion data. According to the method, the density of effective information in the multi-modal fusion data is improved, the object features are enriched and described, the object features can be quickly determined, and the recognition accuracy and the recognition efficiency are improved.

The invention provides a visualization method and system for self-adaptive matching of a display object resolution and a fusion processing mode. The method provided by the invention comprises the following steps: before to-be-displayed content reaches a video physical port of a visualization system, determining a display template according to a plurality of to-be-controlled visualization system display sizes; selecting a target visualization system according to the content density of the to-be-displayed content; determining a display object type in the to-be-displayed content, determining a fusion processing mode of the display object according to the display object type, determining a fused display position of each display object according to the display scene, supplementing a display template corresponding to the target visualization system, and obtaining a target display template; different types of display objects are obtained by utilizing a video physical port, each display object is processed by utilizing a video processor according to a fusion processing mode of each display object, and a plurality of processed to-be-fused display objects are displayed in the same canvas according to a fused display position in a target display template to obtain a fused canvas; and displaying in the target visualization system.

This invention discloses a method for rapid prediction of data center temperature fields based on convolutional neural networks, comprising: acquiring the geometric information of the data center structure; performing geometric modeling and mesh generation on the data center structure to obtain the mesh structure of the data center; acquiring the operating parameters of the equipment within the data center; acquiring the temperature field data of the data center and dividing the temperature field data according to a height plane to generate the output dataset for training the convolutional neural network model; converting the mesh structure of the data center at different heights into a spatial feature map structure and recording the corresponding height information and operating parameter information to generate the input dataset for training the convolutional neural network model; and using the trained convolutional neural network model for temperature field prediction. This method can achieve accurate and reliable rapid prediction of the temperature field of data center rooms.

The application relates to a ciphertext retrieval method for a localization database, which is based on hierarchical security targets of covering mode leakage and actual operation conditions, and is designed under the guidance of strong backward security to support an index structure with an immediate deletion function and a searchable encryption scheme meeting the strong backward security, and a trusted execution environment is applied to improve the scheme performance. The application effectively supports the immediate deletion of index information and document information, reduces the redundant information on the server, in the scheme design aspect, the trusted execution environment is adopted to optimize the searchable encryption scheme performance, reduce the client storage, and greatly improve the ciphertext retrieval efficiency.

Embodiments of the present application disclose a system anomaly determination method and device, electronic equipment and a storage medium. The method comprises: obtaining a target log generated in a system running process and a plurality of log templates stored in the system, wherein the target log is generated according to any one of the plurality of log templates; obtaining an attribute set of the target log and encoding information of each log template respectively; determining a log text vector of the target log according to the attribute set and all the encoding information; and determining whether an anomaly occurs in the system according to the log text vector. Embodiments of the present application obtain the target log generated in the system running process and the plurality of log templates stored in the system, then further obtain the attribute set of the target log and the encoding information of each log template, thereby determining the log text vector of the target log, finally realizing the judgment of the system anomaly, reducing the generation of redundant information, and improving the efficiency of determining the system anomaly.

The embodiment of the invention provides a data verification method and device, electronic equipment, a storage medium and a program product. Comprising the steps that state message information currently sent by a sending end is received and stored, a first verification data packet currently sent by the sending end is received, and the first verification data packet comprises transmission data, first verification data and target data; the target data comprises first low-order data of a first preset digit in the first counting data and second low-order data of a second preset digit in the second counting data. And determining target step data according to the first low-order data and stored first counting data, and determining second verification data according to the target step data, the first counting data, the second counting data and the first verification data packet. And if the second verification data is the same as the first verification data, the verification of the state message information is passed. According to the invention, the data processing amount in the verification calculation process is reduced, so that the communication resources occupied in the verification process are reduced.

The invention discloses a group behavior recognition method based on a Markov decision process and extension graph convolution, and relates to the technical field of computers. The method comprises the following steps: extracting a key frame of a video to be identified based on a Markov decision process, and sequentially extracting a convolution feature map of the key frame, bounding boxes of a plurality of target individuals in the convolution feature map and behavior features of each target individual; in each layer of graph convolutional network, calculating a cosine distance between behavior characteristics of a node and an adjacent node in the human body target relation graph so as to determine a connection weight and update the behavior characteristics of the node; calculating a relation feature vector between any two nodes based on the behavior features obtained by the last layer of graph convolutional network, and dividing a plurality of sub-target interaction graphs; classifying each sub-target interaction diagram to obtain a predicted behavior label; and determining the predicted behavior tag of the subgroup with the highest occurrence probability in all the key frames as a group behavior tag. According to the method, the group behavior identification precision is improved.

The present invention discloses a small target detection method based on deformable attention feature fusion, step S100, obtaining a data set and performing data preprocessing; step S200, constructing an enhanced small target lung nodule detection network model based on a deformable attention mechanism; step S300, training an enhanced small target lung nodule detection network model based on a deformable attention mechanism; step S400, inputting an image to be detected into the trained enhanced small target lung nodule detection network model based on a deformable attention mechanism, and outputting a detection result. Aiming at the characteristic that the features of small-sized targets are concentrated in shallow features, the present invention adds a four-fold down-sampling detection layer in the feature fusion pyramid, supplements the spatial position and other information of small targets omitted in the original detection layer, helps to extract and enhance the feature representation of small targets, and at the same time reduces the interference of background redundant features, thereby improving the detection accuracy of small targets.

The application discloses a multi-modal emotion recognition method based on intra-modal perception and inter-modal cross fusion of a Transformer mode, and steps include the following: firstly, encoding and extracting deep features of speech and text; then, based on the intra-modal perception module of the Transformer proposed by the application, long-distance dependency relationships in each mode are captured, local perception learning of emotion features is realized, and redundant information in the deep features is reduced; secondly, in order to fuse unaligned multi-modal sequence information and fully utilize the complementarity of different modal information, the inter-modal interactive fusion module based on the Transformer is proposed to capture information dependency relationships between different modes and obtain fused multi-modal global information; finally, an ablation experiment is conducted to verify the effectiveness of the method. The application realizes effective parallel computing for multi-modal emotion recognition, and further improves the recognition performance and generalization ability of a multi-modal emotion recognition system.

The present invention relates to a virtualization nested translation method, system and readable medium, comprising receiving a virtual address translation request, judging the address translation mode of the virtual address translation request, and determining the virtual address type; selecting a corresponding translation cache according to the determined virtual address type; performing address translation in a multi-level translation cache in a parallel or serial search mode under the selected translation cache to obtain a search result; judging whether the search result hits a leaf item of a page table; if the search result hits a leaf item, returning a corresponding physical address to complete the translation; if the leaf item is not hit, performing a page table traversal to obtain a physical address. The method can improve the speed of virtualization nested translation, reduce redundant virtual addresses of a translation cache and redundant PTEs of a translation path cache under a single technical condition, and improve the translation delay from a virtual address to a physical address in memory virtualization or peripheral virtualization.

The present invention discloses a sunshine exchange settlement model training method, device and transaction data classification method, which specifically include the following steps: S1: acquiring transaction data and dividing it into a training set and a test set; S2: acquiring the weight of the sunshine exchange settlement model and reshaping it into a second-order matrix; S3: performing full-rank decomposition on the sunshine exchange settlement model to obtain two layers of convolution or reconstruct a fully connected layer; S4: applying orthogonal regularization and funnel regularization on the left singular matrix and the right singular matrix after the full-rank decomposition; S5: adopting an energy-based threshold method to reduce model parameters, thereby obtaining a low-rank model; S6: adjusting the hyperparameters of the low-rank model to obtain a target network model; the target network model is used to classify the transaction data.