299 results about "Information redundancy" patented technology

The invention discloses an environment modeling method and system for realizing intelligent security and protection, and relates to the technical field of intelligent security and protection, and the method comprises the steps: collecting a visual image and depth data of a monitoring region, and generating a monitoring scene feature matrix; performing hierarchical feature extraction on the monitoring scene feature matrix by using a deep convolutional neural network, and constructing a scene semantic feature map; projecting the scene semantic feature map to a three-dimensional monitoring space according to the depth data, and constructing a three-dimensional scene grid model; extracting spatial topological characteristics of each grid node in the three-dimensional scene grid model, and calculating association weights among the grid nodes; and constructing a scene dynamic knowledge graph based on the spatial topological features and the associated weights, and carrying out modeling analysis on abnormal events. According to the method, through cooperative collection of the multi-view vision sensor and the depth sensor, in combination with information entropy weighted standardization and principal component analysis, accurate generation of the monitoring scene feature matrix is realized, and the problems of information redundancy and feature coupling in traditional vision data processing are effectively solved.

The invention relates to the technical field of power grid equipment state monitoring, in particular to a transformer substation equipment health state assessment method, system and equipment based on few-sample multi-modal fusion and a storage medium. The method comprises the following steps: acquiring image, sound, text, structuralization and other multi-modal operation data of transformer substation power distribution equipment, preprocessing the data, and establishing a comprehensive equipment state data basis; cLIP, BERT and Wave2vec pre-training models are selected and finely adjusted, multi-modal features are deeply fused through a cross attention mechanism by using the transfer learning ability of large-scale pre-training knowledge, pairwise interaction and information complementation among different modals are realized, information redundancy is effectively eliminated, and a complex association relationship among the modals is mined; a hierarchical structured model is constructed, a structured data health state result is calculated in combination with a discrimination matrix, and traditional power system expert experience and quantitative analysis are organically combined; and carrying out weighted fusion on the multi-modal health state result and the structured data health state result.

The invention relates to the technical field of street cleanliness evaluation, and discloses a street cleanliness real-time evaluation method based on multi-modal data fusion. The method comprises the following steps: firstly, acquiring initial acquisition parameters and original monitoring data of the multi-modal sensing equipment, wherein the data comprises overlapped region environment information and is acquired according to a preset direction; then performing multi-modal preprocessing on the original data, extracting heterogeneous feature difference points and generating an incidence relation graph; dividing an effective monitoring range based on the atlas, and adjusting acquisition parameters to form an optimized parameter set; performing space-time alignment on the adjusted data to generate a multi-source data sequence; and finally, performing fusion processing on the overlapped region, eliminating feature conflicts and information redundancy, and generating a real-time evaluation result. In addition, the incidence relation graph can be corrected in combination with manual inspection records. According to the method, multi-modal data can be effectively fused, real-time and accurate evaluation of the street cleanliness is realized, and the evaluation reliability and adaptability are improved.

The invention discloses a sentiment analysis method based on prototype guide mode fusion and prompt enhancement, and constructs a multi-mode sentiment analysis network which comprises a multi-mode coding module, a prototype guide mode fusion module, a dynamic mode weight adjustment mechanism and a context prompt generation module. The method comprises the following steps: firstly, extracting semantic features of each mode by using a multi-mode encoder, and constructing a prototype feature library based on a labeled sample to describe typical representations of different modes under each category; and then, dynamically evaluating modal contribution through prototype similarity to realize modal adaptive fusion. Furthermore, a context prompt is generated according to a similarity retrieval result of the input sample and the prototype library, and the pre-training language model is guided to complete sentiment classification. According to the method, the problems of modal inconsistency, information redundancy, weak small sample generalization and the like can be effectively relieved, and the accuracy and robustness of sentiment analysis are improved.

The invention discloses a panchromatic sharpening method based on multi-resolution panchromatic feature guidance, which comprises the following steps: firstly, designing a double-branch feature decomposition module for respectively performing feature decomposition on a panchromatic image and a multi-spectral image to extract low-frequency basic features and high-frequency detail features, and enabling the two branches to share the same shallow feature extraction module; correlation constraint is applied to the extracted basic features; then, aiming at the problem of information redundancy caused by direct serial connection and stacking of the extracted features, performing interactive fusion on the features by using an interactive feature recombination module, and enhancing a long-distance dependency relationship and a global context relationship among different modal features; and finally, an INN-based image reconstruction module is provided, and efficient fusion of the basic features and the detail features is realized through multi-level transformation and mapping. According to the method, cross-modal features can be effectively extracted and recombined and fused, spectrum distortion is remarkably reduced, and the spectrum distortion phenomenon caused by improper feature fusion is effectively avoided.

The invention provides an influence maximization method and system based on deep reinforcement learning and a graph neural network, and belongs to the field of influence maximization in social network analysis. The method aims to optimize selection of seed nodes through a reinforcement learning technology so as to realize maximum propagation of information in a network, and the graph neural network can accurately capture an interaction relationship between nodes in a graph structure by virtue of a message transmission mechanism of the graph neural network, which is highly matched with an information propagation process in a cascade propagation model. Therefore, in the influence maximization model based on the deep reinforcement learning and the graph neural network, the graph neural network is combined with the deep reinforcement learning, the embedded representations of the nodes are learned by using the graph neural network, and the embedded representations are used as the input of the deep Q network so as to fully capture the complex structure and node relationship in the network. In addition, a multi-head attention mechanism is introduced to aggregate information of each round, expression of key features is enhanced, and information redundancy is effectively reduced.

The invention discloses a training method of a controllable and credible official document generation model, which relates to the technical field of natural language processing, and comprises the following steps: S1, acquiring original official document data based on a government agency, a public database and a legal document library, establishing a multi-source and multi-type corpus, the multi-source and multi-type corpus comprises a request report, a conference summary, a notification announcement and a policy document. According to the method for training the document generation model, the natural language processing technology and document generation specifications are deeply integrated, the document generation quality and efficiency are remarkably improved, the richness and diversity of document content are ensured by constructing a multi-source and multi-type corpus, and meanwhile, the method has the advantages of being high in practicability and easy to popularize. The problems of information redundancy and different formats are effectively avoided through deduplication and standardization processing, data quality is strictly controlled through introduction of an evaluation system, reliable guarantee is provided for model training, and dynamic constraint and real-time monitoring of the official document generation process are achieved.

The invention discloses an intelligent and automatic transformer substation information acquisition system, which relates to the technical field of information acquisition, and is characterized in that an information change coefficient and an information priority coefficient are acquired and substituted into a logistic regression model to obtain an information potential evaluation coefficient, the information potential evaluation coefficient is analyzed with a preset potential threshold to obtain a comparison result, and the transformer substation information is acquired according to the comparison result. Dividing to obtain a real-time acquisition potential information set and a non-real-time acquisition potential information set, substituting into a default score combination mechanism, determining an actual classification acquisition result of each piece of information, obtaining a network state when each piece of information is uploaded, obtaining a network delay degree and a bandwidth utilization rate from the network state, and substituting into fuzzy logic for fuzzy reasoning to obtain a real-time acquisition potential information set and a non-real-time acquisition potential information set; according to the method, the reasonability result of the default score combination mechanism is determined, then the network delay degree and the bandwidth utilization rate when each piece of information is uploaded are substituted into set parameters in the default score combination mechanism, the score combination mechanism is determined and adjusted, and the information redundancy and the system burden are reduced.

The invention discloses a lung cancer gene mutation classification method based on frequency domain multi-scale fusion guidance, and relates to the technical field of medical image processing and gene detection. According to the MFHA mechanism provided by the invention, the pathological image is decoupled into low-frequency global and high-frequency detail sub-bands through wavelet transform, and extraction of key high-frequency features such as cell nucleus morphology and local texture is enhanced by combining multi-scale convolution and up-sampling guided by high-frequency information; the problems of insufficient feature detail mining and low feature fusion efficiency in a traditional pathological image analysis method are solved; key features are screened and focused through a channel, frequency domain-space feature deep fusion is realized through up-sampling, robust representation is constructed by combining space attention with cosine similarity and multi-dimensional statistical features, a frequency domain analysis-space focusing collaborative optimization mechanism is formed, information redundancy caused by simple feature splicing is avoided, and the robustness of the system is improved. And the classification stability of the model in a complex pathological scene is improved.

The invention discloses a reservoir multi-target optimization scheduling decision-making method based on digital twinning and AI simulation, and relates to the technical field of reservoir scheduling, the method effectively solves the problems of data fragmentation and information redundancy in traditional scheduling by constructing a multi-source data fused reservoir digital twinning body, realizes accurate virtual-real mapping of reservoir working conditions, and improves the scheduling efficiency. A high-reliability simulation basis is provided for subsequent optimization scheduling, and decision errors caused by data errors or model deviation are avoided; four core targets of flood control, power generation, ecology and water supply are considered, an engineering safety bottom line and rigidity requirements are defined through a layered constraint system, constraint satisfaction degree screening is enhanced by means of an improved multi-target optimization algorithm, and target weights can be dynamically adjusted according to a real-time scene. And after the non-dominated scheme set is optimized and output, candidate schemes are screened in combination with real-time demand quantitative scores.

The invention discloses a query statement generation method and electronic equipment, and relates to the technical field of query statement generation, and the query statement generation method comprises the following steps: obtaining a historical query record in response to a received natural language query request of a user; judging whether query relevance exists between the natural language query request and a historical query record or not, and if yes, determining a candidate data table set based on a historical query data table; calling a pre-trained language model, and performing first reasoning on the natural language query request and the candidate data table to generate an initial query statement; executing the initial query statement on a first data table set constructed based on the candidate data table set, verifying whether a query score of an initial query result meets a preset availability standard threshold value or not, and if yes, taking the query score as a target query statement. According to the method and the device, the problems of input information redundancy, high overhead loss, low statement verification efficiency and insufficient correction capability in related technologies are solved, and efficient and accurate generation from the natural language query request to the query statement is realized.

The invention discloses a multi-modal ship classification method based on cross-modal multi-stage fusion, and the method comprises the following steps: building a cross-modal multi-stage fusion ship classification network for a visible light image and an infrared image of a ship; training the cross-modal multi-stage fusion ship classification network; and inputting a visible light image and an infrared image to be classified into the trained cross-modal multi-stage fusion ship classification network to obtain a classification result. According to the method, a plurality of cross-modal feature interaction modules are integrated in two feature extraction networks with unshared weights, interaction features between two modals are extracted by using cross attention, and redundancy is removed by using low-rank decomposition, so that the extraction of the cross-modal interaction features is effectively enhanced, and the accuracy of ship identification is improved; and at the later stage of fusion, different modal confidence coefficients are introduced by calculating the conditional information entropy, so that the later fusion effect of the multi-modal features is improved, information redundancy and loss are reduced to the greatest extent, and the reliability of an identification result is improved.

The invention relates to the technical field of computer intelligent operation and maintenance and artificial intelligence, in particular to an operation and maintenance scheme generation method, system and device based on cooperation of a large language model and a knowledge graph and a medium. The method comprises the following steps: denoising and abstracting multi-source heterogeneous operation and maintenance data by using a large language model, extracting key fact dimensions, further identifying a fault entity, mapping the fault entity to an operation and maintenance knowledge graph to position an initial anchor point, executing two-stage collaborative reasoning based on the anchor point, and generating a plurality of candidate traceability paths by alternately performing relationship exploration and entity exploration; and finally, reordering the candidate paths based on path length perception and semantic correlation, dynamically controlling the termination and understanding strategy of reasoning by using a comprehensive confidence score, outputting a final solution, effectively overcoming the problems of information redundancy interference reasoning and large model illusion in a complex operation and maintenance scene through the cooperation of a large model and a knowledge graph, and improving the reliability of the system. And the accuracy of root cause positioning and the executable performance of the solution are obviously improved.

The invention discloses an intelligent manufacturing factory production line environment on-line real-time monitoring management and control system, and relates to the technical field of environment monitoring, and the system comprises a system architecture including a sensor sensing module, a credible verification module, a self-repairing module and a management and control module. And full-flow closed-loop monitoring from data acquisition, credibility analysis to repair feedback is realized. The sensor sensing module distributes multiple environmental monitoring acquisition nodes on a production line, and performs zoning distribution according to the process layout and the characteristics of an air flow field. The acquisition nodes in each area form a local trust cluster, and local redundancy and data mutual correction among the nodes are realized, so that a distributed environment sensing structure based on spatial correlation is established. According to the arrangement, the space coverage density of environmental parameters is remarkably improved, and information redundancy and structural stability can be achieved through neighborhood nodes under the condition that any node breaks down, drifts or loses data.

The invention provides a partial discharge mode recognition method based on multi-channel sound signal time-frequency space spectrogram fusion, and belongs to the technical field of partial discharge mode recognition. The problem that information redundancy and difference between existing multi-channel sound signals cannot be effectively fused and discriminant features cannot be extracted is solved. The method comprises the following steps: collecting and summarizing multichannel sound signal data corresponding to different partial discharge models as a data set; performing normalization processing on the original multichannel partial discharge sound signal, and mapping a time domain signal into time-frequency joint distribution to obtain a gray scale time-frequency spectrogram; performing multi-scale and multi-direction decomposition on the grayscale time-frequency spectrogram to obtain each channel sub-graph; different fusion strategies are used for fusing the channel sub-graphs obtained through decomposition; reconstructing a time-frequency space fusion spectrogram from the fused sub-graphs of each channel through inverse transformation; extracting features; identifying a partial discharge mode; the method is applied to partial discharge detection.

The invention relates to the technical field of Internet of Things data processing, in particular to a multi-modal data real-time processing method for an Internet of Things gateway, and the method comprises the steps: obtaining multi-source operation data, and packaging the multi-source operation data into a standard operation data set; obtaining high-bandwidth modal data in the standard operation data set; when the maximum link delay time is greater than a preset delay abnormal threshold value, obtaining a version effective fingerprint of the target model and a local model fingerprint; when a model version abnormity diagnosis result is obtained, calculating information redundancy; obtaining correction delay time, and comparing the correction delay time with a preset normal reference delay interval; and when the correction delay time is greater than or equal to the upper limit value of the normal reference delay interval, adjusting the current decision period. By monitoring delay in real time, diagnosing consistency of model versions, eliminating redundant information and executing link adjustment, accurate processing of high-bandwidth modal data is realized, and real-time performance and reliability of Internet of Things gateway data processing are improved.

The invention provides a method and device for controlling vehicle functions, a vehicle and a storage medium, and relates to the technical field of vehicle function control. According to the method, function control information is determined, an analysis object can be clearly analyzed, and introduction of irrelevant data in the initial stage is avoided. Information redundancy needing to be processed by the preset large model is reduced from the source, and a foundation is laid for subsequent efficient analysis. And further performing semantic analysis on the function control information according to a layer-by-layer screening mode through a preset large model and a vehicle function formed based on a vertical progressive layered structure. According to the hierarchical structure, the preset large model does not need to traverse all vehicle functions, and each layer only focuses on the function of the corresponding layer, so that the number of functions needing to be processed by the preset large model can be greatly reduced, and the consumption of lexical elements is directly reduced. Meanwhile, layer-by-layer screening replaces one-by-one reasoning, the target range of each layer is rapidly narrowed, and all vehicle functions do not need to be analyzed one by one. The path for determining the function of the target vehicle can be remarkably shortened, and the determination efficiency is improved.

The invention discloses a vibration measurement information enhancement method based on fusion data, and relates to the technical field of vibration measurement information enhancement. Edge nodes splice reasoning probability, frozen layer activation statistics, weak spectrum gradient direction and frozen layer gradient variance to generate a multi-source confidence tensor, and a heat peak list is formed through adaptive time-frequency kernel mapping; the information is uploaded after being subjected to redundancy elimination screening; a central node distributes domain credibility to each domain and then constructs a global heat map by adopting self-attention weight fusion heat peaks, and links hierarchy, frequency bands, links and working condition nodes in a mechanism attribution map network to calculate a causal path; and an attribution result triggers a strategy rule engine to select a corresponding maintenance script or a network self-checking instruction, and the maintenance script or the network self-checking instruction is pushed to a monitoring large screen, a mobile terminal and a mail in a unified message packet format along with a global heat map and a reason link, so that display and explanation of weak fault abnormity and operation and maintenance instruction closed loop are completed.

The invention discloses an ultra-lightweight low-resolution dark light face enhancement method and system, and belongs to the technical field of computer vision and artificial intelligence, and the method comprises the steps: obtaining a degraded image, and processing the degraded image into a shallow feature map; inputting the shallow feature map into a hierarchical network structure; setting a module group in each level of the encoder and the decoder; only the first Transform module is used for carrying out complete self-attention calculation; the other Transform modules except the first Transform module in the module group execute attention sharing operation; inputting the deep feature map into an image reconstruction module formed by a convolutional layer to generate a residual image; and carrying out pixel-level addition on the residual image and the input degraded image to obtain a recovered clear image. According to the method, the problems of high calculation complexity and information redundancy in the prior art are solved. According to the method, the calculation overhead is remarkably reduced through an attention sharing strategy, and the excellent recovery quality is ensured through wavelet enhancement.

The invention relates to a cross-modal hash retrieval method based on label-driven semantic perception learning. The method comprises the following steps: S1, acquiring a data set; s2, constructing a CLIP, and inputting samples in the data set into the CLIP for initial feature extraction; s3, constructing a teacher model, wherein the teacher model comprises a feature extraction module, GLoVe, an information supplement module, an information filtering module and a feature hash module; s4, constructing a student model, wherein the student model comprises an image MLP and a text MLP; s5, constructing target loss functions corresponding to the teacher model and the student model; s6, parameters of the teacher model are updated according to the target loss function, and training of the student model is guided after training of the teacher model is completed; and S7, generating respective corresponding hash codes for the data in the database and the query data by using the trained student model, and outputting a retrieval result according to the distance between the hash codes. The information filtering module reduces the information redundancy of the sample, and the completeness of the semantic content of the sample is improved through the information supplementing module.

Embodiments of the invention provide a data packet transmission method and apparatus, an electronic device and a readable storage medium. The method comprises the steps of obtaining a target data packet; extracting a target field from the target data packet; acquiring historical communication data of the network equipment, and extracting a field in the protocol frame from the historical communication data to determine information redundancy of the field; according to the importance degree corresponding to the information redundancy evaluation field, determining the importance degree corresponding to the field matched with the target field as the importance degree corresponding to the target field; splitting the target data packet into sub-data packets according to the importance degree corresponding to the target field and the network state of the network equipment, and determining a transmission strategy of the sub-data packets; the transmission strategy at least comprises a window size and / or a bandwidth proportion when the sub-data packets are transmitted; and transmitting the sub-data packets according to the transmission strategy. According to the embodiment of the invention, the target data packet is transmitted according to the importance degree of the field in the target data packet and the network state, so that the transmission efficiency and reliability are ensured.

The invention provides a vibration acceleration monitoring point clustering optimization method and system for a satellite mechanical test, and the method comprises the steps: obtaining initial data of a frequency response function of each vibration acceleration monitoring point of a satellite through a dynamic test, carrying out the standardization and merging, and obtaining the final data of the frequency response function; dividing the vibration acceleration monitoring points with similar dynamic characteristics into a cluster by taking data correlation as a clustering distance based on a k-means clustering algorithm; classifying the clusters according to the number of the measuring points in the clusters, and determining reserved measuring points of each cluster and cutting measuring points of each cluster according to the correlation of each vibration acceleration monitoring point in each cluster; the number of the reserved measuring points is determined by calculating the redundant correlation coefficient between the cutting measuring points of each cluster and the reserved measuring points in the same cluster and the independent correlation coefficient between the reserved measuring points of each cluster, so that the information independence of the reserved measuring points and the information redundancy of the cutting measuring points are ensured, the cutting method is quantified, the measuring point cutting result is more reliable, and the measurement accuracy is improved. And the efficiency of a ground environment test link is improved.

The invention relates to a reservoir level prediction method and device based on multi-modal feature fusion, and belongs to the technical field of hydrological prediction. The method comprises the following steps: acquiring meteorological data, hydrological data and image data of an area where a reservoir is located; constructing a feature extraction and conversion module, and processing the meteorological data through the feature extraction and conversion module to obtain evaporation capacity, reservoir precipitation and water flow; constructing a feature fusion module, wherein the feature fusion module comprises a mapping unit and a gating attention module; the evaporation capacity, the water flow and the reservoir precipitation are mapped to the same dimension through a mapping unit, and then fusion features are obtained through a gating attention module; the fused features are subjected to water level prediction through a reservoir water level prediction module to obtain a water level prediction value. According to the method, multi-source features such as evaporation capacity, water flow and precipitation are dynamically fused through the improved gating attention unit, the information redundancy and loss risk of high-dimensional features can be effectively reduced, and the prediction precision and the model robustness are improved.

The invention relates to an unsupervised speech recognition modeling method based on phoneme segment-level characterization discretization, and belongs to the field of speech recognition. According to the method, phoneme features of original audio are extracted by using an IFMF model through a voice feature discretization module, then a K-Means clustering algorithm is trained to cluster audio representation, and a clustering index is acquired as a discrete mark; training a generator network and a discriminator through an adversarial learning module; and decoding the phoneme discretization-based model output obtained by unsupervised training by using a language model through a voice discrete representation decoder module. According to the method, multiple speech features are fused, and phoneme segment-level characterization discretization is adopted, so that the influence of tone information on Lao is considered, and the influence of information redundancy caused by fine-grained features on cross-modal modeling is reduced as much as possible. Compared with a traditional speech recognition method, the method provided by the invention obtains a competitive result.

The invention discloses a content abstract generation method based on chapter structure analysis, and belongs to the technical field of natural language processing. The method comprises the steps that firstly, an original text is preprocessed, then text structure deep analysis is carried out, the text type of the text is recognized, an explicit / implicit text relation is extracted, and special symbols are introduced through a Prompt normal form for implicit text relation extraction to strengthen logic semantics; secondly, scoring sentences by adopting a double-path scoring mechanism in combination with a deep neural network model of chapter structure features and an optimized text sorting algorithm, and fusing scores through a logistic regression model; then, screening target sentences based on a chapter relation weighted secondary modulus function and a greedy algorithm, and finally, carrying out post-processing to generate an abstract. According to the abstract generation method, the chapter structure logic is deeply utilized, so that the problems of logic unsmoothness, information redundancy or key relation missing in the existing abstract generation are solved, and the semantic coherence and information integrity of the abstract are improved.

The invention discloses a fusion voxel-based global mask auto-encoder point cloud classification segmentation method and system, and belongs to the technical field of three-dimensional point semantic segmentation and deep learning, and the method comprises the steps: obtaining and processing a non-labeled point cloud data set and a labeled point cloud data set, constructing a fusion voxel-based global mask auto-encoder model, and obtaining a fusion voxel-based global mask auto-encoder model; parameters obtained through pre-training are migrated, a downstream task model is constructed, then the model is finely adjusted through a labeled data set, and an optimal deep network model is obtained and used for classification or segmentation of three-dimensional point clouds. According to the method, a self-supervised learning method is adopted, the problem of insufficient mark data sets is solved to a certain extent, the problem of low efficiency of a model caused by dividing local blocks based on a mask auto-encoder in the past is effectively improved, no farthest point sampling and K-nearest neighbor operation exist, global voxel features are used as prediction targets, and the prediction efficiency is improved. And the problem of information redundancy caused by block division is also avoided.

The invention discloses a collaborative awareness and recognition method and system based on an unmanned aerial vehicle cluster, and relates to the field of unmanned aerial vehicle cluster control and computer vision. The method comprises the steps of starting ground control software to load a route path file, constructing a sequence image optical scene and evaluating the effect, dynamically fusing and evaluating the effect of a wide-area optical scene, intelligently and collaboratively recognizing an optical target, confirming and judging collaborative recognition data, and counting and evaluating recognition accuracy. The system comprises a central node unmanned aerial vehicle subsystem, an edge node unmanned aerial vehicle subsystem and ground control software. According to the method, the central node and the edge nodes cooperatively work, the global scheduling decision-making capability of the central node and the real-time computing capability of the edge nodes are exerted, efficient coverage of a wide-area scene, high-resolution image splicing and cooperative recognition and tracking of a dynamic target are achieved, the problems of unbalanced computing load, information redundancy and the like of a traditional architecture are solved, and the method is suitable for large-scale popularization and application. The image quality and the target recognition accuracy are improved, the reliability is high, and the expandability is high.

The invention discloses an information display frequency control method based on user behavior characteristics, relates to the technical field of information display control, is used for solving the problem of excessive pushing of display contents, and obtains similar potential groups by setting a group division mechanism for initial users. Selecting target information according to the information display duration of each group, extracting word frequency, screening representative vocabularies, screening a target group according to the representative vocabularies, displaying the information of the target group on a user interface, setting a display frequency, monitoring the browsing duration and the browsing amount of a user in real time, performing statistics, and screening out the target group according to a result. Generating an update ratio according to the screening quantity to enter a frequency adjustment mechanism, adjusting a display frequency by using the update ratio to obtain a precise frequency, monitoring a screening result in real time to judge whether to quit the frequency adjustment mechanism, and collecting user click search and active refresh times during quit to determine whether to quit the group division mechanism; user fatigue caused by information redundancy is avoided, and information utilization efficiency and resource allocation rationality are improved.

The invention discloses an image semantic segmentation method based on gating attention and multi-scale enhancement, and belongs to the field of image semantic segmentation, and the method comprises the steps: building a channel space attention gating sequence building module; constructing a multi-scale feature enhancement block; constructing a gating fusion block; an encoder-decoder network is improved based on a channel space attention gating sequence building module, a multi-scale feature enhancement block and a gating fusion block, and a semantic segmentation neural network is obtained; obtaining an image training set, and training the semantic segmentation neural network by using image training to obtain a trained semantic segmentation neural network; and obtaining a to-be-classified image, and classifying the to-be-classified image by using the trained semantic segmentation neural network. According to the method, the problems that the context information capturing capability is limited, the global modeling capability is relatively weak, the multi-scale fusion lacks a dynamic adaptation mechanism and the jump connection information is redundant in the existing method are solved.

The present application relates to the technical field of communications, and provides a communication method and apparatus. An access network device receives at least two pieces of first information, each piece of first information comprising an operation identifier, and the operation identifier being used for indicating information about a corresponding service operation. If the operation identifiers in the at least two pieces of first information are the same, because in the present application, each operation identifier indicates the information about the corresponding service operation, the access network device receives at least two identical operation identifiers, and it is considered that a plurality of identical operation instructions are received; in this case, the access network device only executes the service operation corresponding to the operation identifiers once. On this basis, the situation of executing identical operation instructions multiple times and reporting multiple identical operation results can be avoided, so that the problem of information redundancy can be avoided and interaction signaling between access network devices can be reduced, thereby saving signaling resources and improving data processing efficiency.