158results about How to "Enhanced Feature Representation" patented technology

The invention discloses an editable condition printing image generation method, and belongs to the field of image generation. The method comprises the following steps: constructing a printing image generation model, and training the model to obtain a generation model with editable conditions; and generating a printing design image meeting the condition requirements by using the model. According to the method, different advanced features of each element of the printed plane image are extracted through downsampling, element and style information in the printed plane image is captured through multi-scale feature extraction and information exchange, then hidden space vectors corresponding to different printed plane image features are analyzed, fusion of specific styles is achieved in the hidden space, and the image quality is improved. And the decoding of the printing design image is realized through the decoder, the decoupling of the element information is realized, the personalized design and diversified requirements can be met, and the expansibility is realized.

The invention discloses a low-light image denoising method based on a local-global interactive Transform, and the method is characterized in that the method comprises the following steps: 1, carrying out the denoising of a low-light image; the method comprises the following steps: step 1, constructing an overall structure of LGIT for recovering a given degraded image; 2, a Top-N Patches selection module is used for selecting N patches most related to the target patch and providing feature input for subsequent modules; step 3, the Patch selects an attention Transform module, and the Transform module is used for carrying out global feature aggregation based on the information of Top-N Patches; and step 4, mixing the double-gate feed-forward network MDGFF for further processing the features output by the PSATB and generating a final de-noised image. According to the method, the self-attention inside the patch and between the patches is effectively calculated on different scales, and accurate removal and detail recovery of low-light image noise are realized.

The invention provides a network node classification method based on a multilayer graph neural network model, and the method comprises the steps: constructing a graph according to network nodes, carrying out the preprocessing of initial node features in the graph, and obtaining an initial embedded vector; according to the initial embedding vectors of the node and the neighbor node thereof, updating the node representation by adopting a layered message passing mechanism; according to the initial embedded vectors of the nodes and the neighbor nodes thereof, the graph attention network uses an attention mechanism to calculate weights for the neighbor nodes of each node and then dynamically updates feature vectors of the nodes; a multi-head attention mechanism is adopted to splice and merge the feature vectors of the nodes output by the graph attention network and the updated node representation, and the comprehensive feature vector of each node is obtained; the comprehensive feature vectors are further optimized through graph contrast learning and a node embedding module, and optimized features are obtained; and obtaining predicted node classification according to the optimization features. According to the method, complex features of heterogeneous graph data can be effectively captured, the method has good robustness for an over-smoothing problem, more accurate heterogeneous network representation learning is realized, and the method has advanced performance and robustness in a node classification task.

The invention discloses an intelligent eye-protection illumination sitting posture correction method and system, particularly relates to the technical field of illumination sitting posture correction, and is used for solving the problem of asynchronous illumination sitting posture analysis. According to the method, data from different sensors are ensured to be synchronized on the same time reference through a dynamic time warping algorithm, errors caused by asynchronization are eliminated, a self-attention mechanism is combined with position coding, the complex relation and time sequence dependence of the data are accurately extracted, the feature representation capability of the model is improved, and finally, the model is optimized. Time sequence data fusion is carried out on the multi-modal features according to the long-short-term memory network, the user state and demand are predicted in real time, accurate light adjustment and sitting posture monitoring are achieved, the robustness and adaptability of data analysis are improved, the user experience and health management efficiency can be remarkably improved in intelligent eye-protection illumination sitting posture correction, and the user experience is improved. And a more comfortable and healthy working and living environment is provided for the user.

The invention discloses an SAR sea surface wind speed inversion method based on multiple attention-global enhancement UNet, and the method comprises the following steps: (1), obtaining wind speed data observed by a buoy, carrying out the cleaning and preprocessing, and making a label data set; (2) acquiring dual-polarization ground distance multi-view SAR data collected in an interference width mode IW type; (3) preprocessing the SAR data; and selecting SAR data corresponding to the buoy data information. Dividing the preprocessed SAR data set into a training set, a test set and a verification set; (4) constructing a multiple attention-global enhancement UNet model, namely a CPA-KiTUNet model, training to obtain a sea surface wind speed inversion model, and selecting a test set to test on the trained inversion model to obtain an inversion wind speed result; according to the invention, richness of inversion data is improved, and wind speed inversion can be accurately carried out in a complex scene.

The invention discloses a single-frame infrared weak and small target detection method based on multi-scale feature enhancement, and the method comprises the following steps: 1, obtaining a single-frame infrared image, constructing a single-frame infrared image data set, and dividing the single-frame infrared image data set into a training set and a test set; 2, an infrared weak and small target detection network model is constructed, the infrared weak and small target detection network takes a YOLOv5 network model as a basic network, and the YOLOv5 network model comprises a trunk feature extraction network, a feature fusion network and a detection head; a multi-scale feature enhancement module is introduced into the trunk feature extraction network, and the multi-scale feature enhancement module comprises a context feature fusion module and a feature detail enhancement module which are sequentially connected in series; 3, designing a training strategy, and training the infrared weak and small target detection network model by taking the training set as input according to the training strategy; and step 4, after training is completed, inputting a test set to test and evaluate the infrared weak and small target detection network model.

The invention relates to the technical field of pharmacology and artificial intelligence, and provides a compound-target interaction prediction method and system, and the method comprises the steps: constructing a knowledge graph containing compounds, target entries and disease entities, and the correlation of the compounds, the target entries and the disease entities; inputting the knowledge graph into a multi-modal feature extraction model, and extracting topological information and internal semantic information in the knowledge graph; wherein the multi-modal feature extraction model comprises an ICD2Vec layer, a graph Transformer, a gene ontology embedding layer and an attention mechanism-based graph neural network, and the ICD2Vec layer, the graph Transformer, the gene ontology embedding layer and the attention mechanism-based graph neural network are respectively used for extracting disease semantic features, compound semantic features, protein target semantic features and topological features of a knowledge graph; performing knowledge fusion on the extracted semantic features and topological features of the knowledge graph through a feature fusion module, and obtaining a compound-target prediction probability matrix through a knowledge graph embedding method; and according to the compound-target prediction probability, performing descending sorting on the input compound-target interaction, and outputting a target recommendation list.

The invention relates to the technical field of big data, in particular to a real-time commodity sales prediction system and method based on edge calculation. The system comprises a data acquisition module, an edge calculation module, a commodity sales prediction module, an optimization feedback module and an intelligent scheduling module. The system collects sales, inventory, customer behaviors and environment data in real time through the Internet of Things, and utilizes a graph embedding technology to construct a dynamic relation graph and enhance features. Modeling is carried out based on a space-time diagram network algorithm, and accurate prediction of commodity sales volume is realized. Model parameters are optimized through reinforcement learning, inventory replenishment, a distribution path, a production schedule and a promotion strategy are dynamically adjusted in combination with a discrete differential evolution algorithm, and a real-time resource scheduling scheme is generated. According to the invention, the accuracy and real-time performance of commodity sales volume prediction are improved, optimal configuration of resources is realized through intelligent scheduling, and the operation efficiency and market competitiveness of enterprises are effectively improved.

The invention discloses an electroencephalogram attention classification model, an electroencephalogram attention classification method and a computer system, and solves the problems of low electroencephalogram signal recognition accuracy, insufficient feature extraction and difficulty in processing complex electroencephalogram signals in the prior art. The model comprises a multi-scale separable convolutional network which performs convolution operation on electroencephalogram signals of each input channel and extracts a low-level feature map; the Squeeze-and-Excitation module is used for carrying out inter-channel dynamic adjustment on the low-level feature map so as to generate a re-weighted feature map, and the Squeeze-and-Excitation module is used for carrying out inter-channel dynamic adjustment on the low-level feature map; the lightweight Transform is used for calculating attention weights between time steps corresponding to feature vectors in the re-weighted feature map from a time dimension based on an attention mechanism so as to capture a dependency relationship between the time steps; the MobileNet is used for extracting an advanced feature map from the feature map in which the dependency relationship between the time steps is determined; and the classification module maps the feature vectors in the advanced feature map to vectors of the category number through a full connection layer, and calculates the category probability through a Softmax function to obtain an attention classification result.

The invention relates to the technical field of building damage identification, and provides a damage degree identification method, system and equipment based on image data and a medium, which can realize cross-disaster double-temporal building damage degree evaluation and is used for building damage degree evaluation and building positioning through an end-to-end convolutional neural network. And experiments are carried out on an xBD data set, so that the network provided by the method provided by the invention is proved to have better performance. According to the method, dual-time-phase feature fusion is adopted to effectively fuse pre-disaster and post-disaster feature maps, and feature fusion is carried out on dual-time-phase feature maps to enhance feature representation; and meanwhile, style unification is carried out by adopting fast Fourier transform, so that the problem of different dual-time-phase data imaging styles caused by different imaging conditions is reduced.

The invention discloses a breast tumor benign and malignant identification system based on a double-branch attention distillation network, belongs to the field of medical image processing, and particularly relates to the breast tumor benign and malignant identification system based on the double-branch attention distillation network. The invention aims to solve the problem that the existing method based on deep learning is insufficient in feature extraction capability and cannot accurately focus a lesion area. The system comprises an ultrasonic image preprocessing main module used for preprocessing collected breast ultrasonic images, obtaining ultrasonic imaging areas, unifying the image size of the obtained ultrasonic imaging areas, obtaining preprocessed breast ultrasonic images, and taking the preprocessed breast ultrasonic images as a training set; the double-branch convolutional neural network main module is used for building a double-branch attention distillation network model; the model training main module is used for obtaining a trained double-branch attention distillation network model; and the model test main module is used for loading the trained double-branch attention distillation network model and classifying breast ultrasound images to be diagnosed.

The invention relates to an Internet of Vehicles multi-agent cooperative sensing method and system. The method comprises the following steps: constructing a collaborative sensing network, wherein the collaborative sensing network comprises a self-agent and a plurality of collaborative agents; acquiring and processing sensing data through the collaborative sensing network; performing feature extraction to obtain intermediate features; self-adaptive sparsification is carried out to obtain sparse features, and the sparse features are compressed and transmitted to a self-agent; performing time sequence feature enhancement on the features of all the agents at the self-agent end; fusing the features to obtain fused features; and constructing an Internet of Vehicles perception model, and realizing perception by the detection model according to the fused features. According to the method, the calculation complexity of traditional global attention is reduced from the square level to the linear level through an adaptive sparsification mechanism, the calculation overhead is remarkably reduced while the multi-agent feature interaction precision is kept, and the method is more suitable for real-time operation on the vehicle-mounted edge equipment with limited resources.

The invention discloses an aerial image camouflage target detection method and device based on a multi-scale attention and boundary sensing network, equipment and a medium. A pyramid vision converter (PVT) is adopted as a trunk network, multi-scale features are extracted as attention-based feature extraction, and a high-resolution feature map is generated; a boundary sensing module (BAM) is developed, and the module aggregates local boundary information and global position information so as to fully mine boundary semantic information of the camouflage target; designing a boundary guidance module (BGM) which integrates the boundary features of the boundary sensing module with the multi-scale features of the backbone network to guide feature learning; a context aggregation module (CAM) is constructed, the module aggregates multi-layer fusion features, context semantic information of a camouflage target is fully mined by adopting extended convolution, and feature representation is further enhanced. According to the method, the problem of detecting the aerial image camouflage target is solved.

The invention discloses a road and bridge deck disease identification method based on deep learning. The method comprises the following steps: acquiring a road and bridge deck disease image data set; constructing a road and bridge deck disease recognition model, and training the road and bridge deck disease recognition model based on the road and bridge deck disease image data set to obtain a trained road and bridge deck disease recognition model; the road and bridge floor disease identification model is constructed and obtained on the basis of a YOLOv9 network framework in combination with a CARAFE up-sampling module and a Triplet Attention mechanism; and obtaining a to-be-identified road and bridge surface image, and inputting the to-be-identified road and bridge surface image into the trained road and bridge surface disease identification model to obtain a disease identification result of the to-be-identified road and bridge surface image. According to the method provided by the invention, the problem of low recognition precision of an existing algorithm in different application scenes is effectively solved, and multiple disease types in road surface and bridge environments can be effectively and accurately recognized.

The invention relates to the field of machine vision and image recognition, and particularly discloses a smoke detection system, which comprises a data preprocessing module used for carrying out normalization, data enhancement and clipping and zooming on an input image to obtain preprocessed image information; the feature extraction module is used for performing feature extraction by using a convolutional neural network structure, generating multi-scale feature representation, obtaining a feature extraction image, and obtaining a multi-scale feature fusion image through a multi-scale feature fusion technology in combination with feature maps of different levels; and the CBAM attention mechanism module comprises a space attention sub-module and a channel attention sub-module, and the space attention sub-module is used for enhancing features of an important space region in the multi-scale feature fusion image according to a space attention mechanism. By adopting the technical scheme of the invention, real-time detection can be realized on resource-limited equipment in a lightweight and efficient manner, and the method has high recognition accuracy for small-target and dynamically changing smoke in a complex scene.

The invention discloses a consumer clustering method based on multi-modal graph comparative learning, and the method comprises the specific steps: (1) firstly processing a discrete multi-modal data set, obtaining neighbor nodes of each sample similarity top-k through employing a KNN method, and constructing an undirected graph structure; (2) obtaining two enhanced views from the obtained graph structure by using an edge disturbance data enhancement method, and then inputting the two enhanced views into a graph neural network to encode node feature representations of the two enhanced views; (3) performing comparative learning pre-training on the encoded feature representation by using an InfoNCE loss function, and finally training to obtain an optimized encoder; and (4) finally, inputting an originally constructed graph structure into a trained encoder to encode a final node feature representation, and then inputting the final node feature representation into a spectral clustering algorithm module for clustering to realize a clustering target. According to the method provided by the invention, discrete data is converted into a graph structure for processing, and a local structure between sample data can be captured; in addition, characteristics with higher discrimination can be learned by using comparative learning pre-training, so that a subsequent clustering algorithm can obtain more accurate clustering division.

The invention discloses a molten steel element content on-line detection system based on a laser-induced breakdown spectroscopy technology, and relates to the technical field of molten steel element content on-line detection, the molten steel element content on-line detection system comprises an excitation system, an acquisition system, a data preprocessing module and a component detection module; the excitation system is used for emitting laser to the molten steel surface to generate plasma; the acquisition system is used for collecting signal light radiated by the plasma; the data preprocessing module is used for carrying out baseline correction and noise reduction processing on the collected original spectral signals; and the component detection module comprises an element content prediction model based on a gating specific expert attention network LSEA-Net and is used for carrying out quantitative analysis on the preprocessed spectral data to obtain the element content of the molten steel. The technology has the advantages that sample preparation is not needed, online rapid analysis and multi-element synchronous detection are achieved, and the technology is suitable for the high-temperature environment and the like, and the requirement of a steelmaking site for real-time component detection is met very well.

The invention provides an infrared weak and small target detection method based on a semantic guidance fusion network. An encoder extracts image semantic features through a convolutional neural network; shallow semantic features are guided by using spatial distribution of deep semantic features through a semantic-guided adaptive fusion module, and target pixels are segmented through a segmentation detection head. According to the method, the problem of deep semantic feature target space information loss caused by downsampling and the problem that target backgrounds are similar and difficult to distinguish due to complex backgrounds can be relieved, so that model false alarms are reduced while space detail information is enhanced. The feature maps of different semantic levels are effectively fused, and it is ensured that the fused feature maps have rich semantic information and accurate spatial positioning information. And the detection performance of the model in a complex scene is improved.

The invention discloses a ship target detection method and system based on a large-kernel multi-scale attention mechanism. The ship target detection method comprises the steps of obtaining a to-be-detected ship target image; performing target detection on the to-be-detected ship target image by using a target detection neural network based on a multi-scale fusion strategy; in the target detection neural network, a detection task is divided into bounding box prediction, category prediction and object prediction, a large-kernel multi-scale attention module is integrated, the large-kernel multi-scale attention module firstly adopts 5 * 5 depth separable convolution to integrate local features, and then a two-dimensional large convolution kernel (k, k) is decomposed into cascaded horizontal depth separable convolution (1, 2); the method comprises the following steps of: extracting multi-scale information (k, k) and vertical depth separable convolution (k, 1), and weighting channel local features extracted by Ghost Module by utilizing the extracted multi-scale information; and the weighted features are deeply fused through the Ghost Module again, so that the accuracy of ship detection is further improved.

The invention discloses a COD (Chemical Oxygen Demand) prediction method based on a two-channel convolution time sequence mechanism, which comprises the following steps of: acquiring COD original data for preprocessing to obtain COD input characteristics; constructing a prediction model, and importing data into a dual-channel convolution module; importing the first double-channel convolution output feature vector and the second double-channel convolution output feature vector into a double attention memory network to obtain a time sequence dependence feature; the time sequence dependence characteristics are imported into an output module, and predicted COD output is obtained; training the model, and in the training process, optimizing hyper-parameters of the model based on an improved pelican algorithm; and importing real-time COD original data into the trained model, and obtaining real-time COD output. According to the method, a high-efficiency effluent COD content prediction model of the sewage treatment plant is constructed by creatively fusing the two-channel convolution and the double-attention memory network.

The invention provides an automobile driver driving behavior identification method and device. The method comprises the following steps: constructing an original identification model comprising an encoder based on a depth separable convolutional network and a decoder based on a compression excitation network; training the original recognition model through the driving behavior sample data and the training configuration data to obtain a driver distraction recognition model; deploying the driver distraction identification model to a vehicle-mounted hardware platform of a target vehicle; in the process that a driver drives the target vehicle, video stream data collected by a cabin camera on the target vehicle is obtained; performing driving behavior recognition processing according to the driver distraction recognition model and the video stream data to obtain a recognition result; and determining the current driving behavior of the driver according to the identification result. Therefore, the method and the device can identify the driving behavior of the driver in real time, are good in accuracy and stability, do not need the driver to wear detection equipment, and ensure the driving experience and comfort of the driver.

The invention relates to the technical field of computer vision, and discloses a visual saliency target detection method based on an RGB-D image, and the method comprises the following steps: S1, carrying out the channel alignment and spatial calibration of a mutual relation between corresponding features of different modals through an MI module, and obtaining a cross-modal representation; s2, then, an IGF module is added to the tail end of the Swin Transform encoder, and the IGF module is added to the tail end of the Swin Transform encoder; s3, the cross-modal fusion feature # imgabs0 # is subjected to channel cross fusion through a CCT module, and relevant information about significance is further obtained; and S4, finally, carrying out modeling on the remote dependency of the fusion features from a global perspective by using a decoding block based on Swin-Transform. According to the visual saliency target detection method based on the RGB-D image, by designing a modal interaction module and a channel cross fusion mechanism, the fusion effect of multi-modal features and the fine modeling capability of multi-scale features are fundamentally improved, and therefore the problems of saliency target detection precision and robustness which are difficult to overcome in the prior art are effectively solved.

The invention discloses a time series prediction method of a multi-time scale channel mixed attention mechanism, and relates to the technical field of time series prediction and deep learning. Specifically, weighted processing is carried out on extracted features of time sequence data through a channel mixed attention mechanism. The mechanism can dynamically adjust the weight of the feature according to the importance of the feature, and ensures that the key feature plays a greater role in prediction. Inputting the time sequence data into a multi-time scale convolutional neural network to obtain feature vectors under different time scales; and fusing the weighted features to obtain comprehensive multi-time scale features, and completing the training process of the time sequence prediction model. According to the method, the accuracy and adaptability of time sequence prediction are improved, and complex modes and trends in time sequence data can be captured more accurately. Through the method, the method can play a role in various application scenes needing high-precision prediction.

The invention belongs to the technical field of computers and information science, and relates to a power network asset key feature extraction method and device, equipment and a medium. According to the method, an improved k-means clustering algorithm is adopted to perform clustering analysis on power network asset data to obtain key features of power network assets; wherein the k-means clustering algorithm is improved in that in the clustering process, the mahalanobis distance between each data point and each clustering center is calculated, and the data points are distributed to the clustering center with the shortest distance; updating a clustering center based on dynamic weight updating and an adaptive learning rate; performing feature weighted summation and nonlinear transformation processing on the key features obtained by clustering by using an attention mechanism to obtain an optimized feature vector matrix; and powerful support is provided for subsequent asset management, fault diagnosis and predictive maintenance.

The invention relates to the technical field of variable subtitles, and particularly provides a multi-variable subtitle method for adaptive balanced perception. The method comprises the following steps: obtaining training set visual features of changed subtitles; constructing a vocabulary and a tag file; in the first stage of the encoder, image features are obtained; in the second stage of the encoder, the obtained common characteristics are as follows; in the third stage of the encoder, extracting difference features of the image; splicing the two groups of image features, and constructing a comparison expression of two differences; finally, modeling the overall difference and the original image respectively; in the first stage of a decoder, grammar correlation is established for words in text data one by one through a text dependence model; in the second stage of the decoder, a group of subtitle information describing image differences is obtained, according to grammar rules and context information, the method ensures that sentences are complete and smooth, visual information and language information are effectively combined, and the accuracy of semantic information is improved.

The invention relates to a defogging network based on dynamic routing and three-dimensional fusion. The network comprises a dual-branch network Twood-branch network, a dynamic routing feature module DRFM and a clear image reconstruction module CIRM, the input end of the double-branch network Twood-branch network is used for inputting a foggy image I, the output of the double-branch network Twood-branch network is connected with the input of the dynamic routing feature module DRFM, the output of the dynamic routing feature module DRFM is connected with the input of the clear image reconstruction module CIRM, and the output of the clear image reconstruction module CIRM is used for outputting a defogged image.

The invention discloses an image super-resolution method and device based on a multi-cascade single look-up table and a storage medium, and relates to the field of image processing, the image super-resolution method comprises an N-cascade look-up table LUT structure and an inference model, and the LUT structure stores a spatial mapping relation of searching a high-resolution HR image by using a low-resolution LR image; in the previous N-1 layers of cascaded LUT structures, the previous layer of LUT structure stores an index value for querying a mapping relation in the next layer of LUT structure, and the Nth layer of LUT structure stores a pixel value of an HR image; when an LR image is input into the reasoning model for reasoning query, the reasoning model performs step-by-step index query based on the N-layer LUT structure, and a corresponding HR pixel value is obtained from the Nth-layer LUT structure. According to the scheme, layer-by-layer searching is carried out in a multi-layer index mode, so that the LUT storage pressure is relieved, the weight of the channel is dynamically adjusted by using the RCAB, the network identification capability is improved, and the SR method indexing efficiency is improved.

The invention provides an RGB-D semantic segmentation method based on feature collaboration and multi-scale feature fusion, and belongs to the field of computer vision and computer graphics. According to the method, semantic segmentation prediction is performed on an image based on an encoder-decoder network structure, multi-modal feature correction and fusion are realized by using multiple attention mechanisms at each stage of a network, feature representation of input data is optimized and enhanced, and a top-down multi-scale feature fusion strategy is used in a decoder. Through multiple attention mechanisms, the model can effectively capture global features and local features, fusion of different modal information under the same scale is realized, and fine aggregation of features under different scales is promoted. According to the method, cooperation of multi-modal features and fusion of multi-scale features are realized by applying multiple attention mechanisms, and the problems of difference between RGB and depth information, noise interference in data and loss of detail information in a network extraction process are effectively solved.

The invention discloses a remote sensing image segmentation method based on multi-scale wavelet transform and Mama. The method comprises the following four steps: firstly, carrying out preprocessing and division on an ISPRS Potsdam data set and a Vaihingen data set; then constructing a segmentation network, wherein the network comprises a local detail extraction branch, a spatial semantic extraction branch, a cross-branch feature fusion part and a decoder; training and optimizing the network by using the training set; and finally, performing segmentation reasoning on a test image by using the trained model. According to the method, the high-frequency detail extraction capability of the image is enhanced through multi-scale wavelet transform, the long-distance dependency relationship is modeled by using the visual state space block, and effective fusion of the features is realized through the double-branch fusion module, so that the feature extraction capability and the semantic segmentation precision are improved, and meanwhile, the training efficiency and the stability are optimized.

The invention belongs to the technical field of stereoscopic vision, and discloses a stereoscopic matching method based on unsupervised online learning. The method is realized through a stereo matching module based on a multi-level cost body, a non-occlusion region confidence estimation module and a Mann-Kendall trend detection module. Compared with the prior art, the method has a better stereo matching effect. The model provided by the invention has the best result in both the initial stage and the later stage, and due to the good generalization of the model, the model can always keep a relatively accurate prediction result in a sheltered region. The model has a good prediction result in the occlusion area due to high generalization, and the tendency of parameter error update is relieved due to the fact that the self-adaptive adjustment strategy provided by the invention eliminates the influence of error loss of most of the occlusion areas, so that the network can maintain the optimal performance all the time.