32results about How to "Improve feature extraction efficiency" patented technology

The invention discloses an equipment state sensing system based on peak shifting and danger situation sensing, which relates to the technical field of electric digital data processing and comprises a data acquisition unit, a model training and optimizing unit, a situation sensing unit, a peak shifting regulation and control unit and a system interface and user interaction unit. According to the method, the device state data feature similarity is dynamically calculated by using the Gaussian kernel function, the similarity matrix is adjusted in combination with the gradient update item, dynamic weighting is performed in feature extraction, and redundant features are reduced by adopting the adaptive discarding probability, so that the feature extraction efficiency of the model in a high-dimensional feature space is improved; the problem that the dimension reduction effect is affected by too many redundant features of the equipment state data is solved. Equipment state features are extracted layer by layer through a deep forest algorithm of a multi-layer random forest structure, a decision tree is optimized in combination with a dynamic adjustment weight mechanism and a reinforcement learning strategy, and the classification capability of the model for equipment state changes and abnormal states is enhanced.

The invention relates to a marine ranch power supply system line fault diagnosis method based on MS-2DResNet and ICBAM, and belongs to the field of power supply system line fault diagnosis. The fault diagnosis method is mainly divided into two stages, namely, multi-scale feature extraction and attention mechanism feature fusion. Firstly, a preprocessed fault data sample is converted into two-dimensional data to be input into a fault diagnosis model; in the first stage, local and global feature extraction is performed on the converted fault data by using multi-scale convolution, so that the robustness of the model for fault data features of different sizes and scales is improved. And then, entering an attention feature fusion stage, and carrying out weighted fusion on stacked data from different scale channels by utilizing the improved CBAM, so that the model pays more attention to useful fault features and neglects useless features, and further, the fault diagnosis accuracy of the model is improved. The method has the characteristics of high diagnosis accuracy and good anti-noise performance.

The invention discloses an unmanned aerial vehicle target detection method based on feature fusion and detection head collaborative optimization, and the method comprises the steps: obtaining a multi-scale feature map through employing a backbone network formed by stacking a convolution module CBS and a residual module C2f, sequentially transmitting the feature maps of various scales to a multi-level feature fusion module (ML-FFM), carrying out the feature fusion from a deep layer to a shallow layer, and carrying out the detection of a target target. Semantic information of shallow layer features is enhanced; inputting the fused feature map into a refined feature enhancement module (RFEM), eliminating position conflicts and redundant information based on a space-time attention mechanism, and optimizing details and positioning capability of deep features; and the enhanced feature map is sent to a prediction head based on Transform for detection, and target classification and positioning of the aerial image of the unmanned aerial vehicle are completed. According to the method, the problems that the unmanned aerial vehicle aerial image is large in scale change, and small targets are high in proportion and dense in distribution are solved, so that the precision of target detection in the unmanned aerial vehicle aerial image is improved. The modules proposed in the method can significantly reduce the calculation amount, and are suitable for the real-time detection scene of the unmanned aerial vehicle.

The invention discloses a grinding workpiece edge shape detection method and system based on machine learning, and the method comprises the steps: obtaining an original image of a grinding workpiece, and carrying out the preliminary image preprocessing; denoising the image and enhancing edge information by combining image filtering and mathematical morphology image processing; based on a second-order differential operator, edge detection is carried out in combination with the morphological gradient, and a workpiece edge contour is obtained; performing feature extraction on the image by using a convolutional neural network, automatically learning workpiece edge features, and constructing and training an edge detection model; acquiring a workpiece image in real time, inputting the workpiece image into the edge detection model for detection, and dynamically correcting the model; and S6, outputting a final edge detection result. According to the invention, by combining image preprocessing, composite image filtering, second-order differential operator and morphological gradient edge detection and feature extraction and model construction of the convolutional neural network, high-precision and robust detection of the edge of the grinding workpiece is realized.

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 belongs to the technical field of image analysis, and particularly relates to a mulberry growth period identification method and system based on image analysis, and the method comprises the steps: obtaining image data of each growth period of a mulberry, preprocessing the image data, and inputting the preprocessed image data into a feature extraction module; performing feature extraction on the image data in each growth period, and establishing a feature comparison table of the images in each growth period of the mulberry trees; mulberry image data of each mulberry planting area to be identified is collected in real time, and after preprocessing, the mulberry image data is input into the feature extraction module for real-time image feature extraction; inputting the real-time image features into a feature comparison module, calling a mulberry growth period image feature comparison table, and matching the real-time image features with the mulberry growth period image feature comparison table; obtaining the mulberry growth period of each mulberry planting area to be identified according to the matching result of the real-time image feature and the mulberry growth period image feature comparison table; a preset mulberry growth period management measure table is called, management measures of the mulberries in the mulberry planting areas are obtained, mulberry planting management is facilitated, and the economic benefits of mulberry planting are improved.

The invention relates to the technical field of image processing, in particular to a lightweight remote sensing scene classification method based on a neural network, and the method comprises the steps: firstly obtaining a remote sensing image data set, and preprocessing the remote sensing image data set to obtain a training set and a test set; then, a teacher model including ConvNext and a student model remote sensing scene classification network of a ShuffleNetv2 network embedded with a mixed attention structure are constructed; inputting the training set into a teacher model and a student model at the same time to train the remote sensing scene classification network to obtain a remote sensing scene classification model; and finally, inputting the test set into the remote sensing scene classification model for classification prediction, and comparing an obtained prediction label with a real label of the test set to complete the test of the remote sensing scene classification model. According to the invention, the feature extraction capability is improved by using a mixed attention structure; and a knowledge distillation method is used to transmit dark knowledge of the teacher model to the student model, so that the precision of the neural network is improved.

The invention relates to the technical field of audio decoding, and discloses an audio decoding performance optimization method and device for a multimode SoC main control chip, and the method comprises the steps: carrying out the framing processing of input audio data through a DSP processing unit, and extracting audio feature data; performing period detection and multi-frame smoothing processing on the audio characteristic data to obtain an audio period characteristic vector; performing principal component transformation and feature dimension reduction based on the audio period feature vector to obtain decoding prediction data, and inputting the decoding prediction data into a task division model to perform decoding processing sequence analysis to obtain decoding task data; carrying out hidden Markov modeling on the decoding task data, and calculating to obtain decoding optimization data; and mapping the decoding optimization data to a hardware acceleration unit of the multi-mode SoC main control chip, and executing parallel decoding operation to obtain target decoding data. According to the invention, adaptive optimization of the decoding process is realized, and the decoding strategy can be flexibly adjusted according to different audio data characteristics and processing requirements.

The invention discloses a flaw detection method and device based on FabricMosaic and an improved fine-grained feature extraction network, relates to the technical field of cloth flaw detection, and solves the problems that in the prior art, a flaw target is incomplete and small target flaws are prone to missing detection due to the fact that a fixed cutting mode is adopted. A dynamic FabricMosaic data enhancement strategy based on target foreground and cloth texture background features is adopted to carry out sample expansion, images in a training set serve as input of a YOLOv5s network model, feature extraction is carried out through a feature fine-grained enhancement extraction module based on a divergent path to obtain a global fine-grained feature map, the YOLOv5s network model is trained, and a target foreground and cloth texture background feature is obtained. The final YOLOv5s network model is used for cloth flaw detection, a dynamic FabricMosaic data enhancement strategy is used for sample expansion, the integrity of a flaw target can still be effectively reserved in the sample integration process, and the risk of missing detection during small target flaw detection is reduced.

The invention relates to the technical field of road traffic marking detection, and provides a road traffic marking detection method based on efficient visual transformation. According to the method, road traffic marking data sets are combined, screened, expanded and labeled; constructing a road traffic marking detection model composed of a feature extraction network, a multi-scale feature map enhancement network and a detection head; inputting into a feature extraction network for feature extraction; inputting into a multi-scale feature map enhancement network for feature fusion; and inputting into a detection head for detection and classification to obtain a final detection result. According to the scheme of the invention, the feature extraction network utilizes an efficient attention mechanism to improve the global context information capture capability and improve the complex road traffic marking detection precision; the multi-scale feature map enhancement network adopts a bottom-up path structure to enhance the capturing capability of small target features, and the recognition precision and robustness under a complex background are improved.

The invention provides a stacking model-based obesity genetic risk model construction method, electronic equipment and a storage medium, the method comprises the following steps: obtaining a data set, the data set comprising genetic feature data, and the genetic feature data comprising single nucleotide polymorphism; screening the single nucleotide polymorphisms by using a random forest to obtain a first genetic feature set related to obesity genetic risks; screening the single nucleotide polymorphism by using a distributed gradient enhanced library model to obtain a second genetic feature set related to the obesity genetic risk; obtaining a significant genetic feature set according to the first genetic feature set and the second genetic feature set; constructing an obesity genetic risk model; according to the method, the model evaluation result of the obesity genetic risk model is obtained, the parameters of the obesity genetic risk model are optimized according to the model evaluation result, stacking training is carried out by integrating multiple basic learners and combining meta-classifier logistic regression, the advantages of different algorithms are effectively fused, and the method is suitable for large-scale popularization and application. And the accuracy and the stability of obesity genetic risk prediction are improved.

The invention relates to an insulator material defect detection method, system, device, medium and program, and the method comprises the steps: carrying out the denoising filtering and image standardization operation of an insulator image obtained in advance, and obtaining a standard insulator image; performing multi-channel convolution and residual connection on the standard insulator image to obtain an insulator image feature set; performing global attention coding on the insulator image feature set to obtain an attention image feature set; extracting global channel image features from the insulator image feature set, and performing channel fusion and spatial fusion on the global channel image features to obtain perception image features; and performing layer-by-layer feature decoding on the attention image feature set by using the perception image features to obtain a defect detection result, thereby improving the insulator defect detection efficiency.

The invention discloses a gas boiler operation state determination method, system and device, a medium and a product, and relates to the technical field of gas boiler state monitoring, the method comprises the following steps: carrying out word segmentation processing on a historical inspection report text set to obtain candidate new words; performing feature extraction on the candidate new words by using a TF-IDF weighting method to obtain a plurality of risk feature words; calculating a correlation coefficient between every two risk feature words, and combining the risk feature words of which the correlation coefficients exceed a first preset threshold value to obtain a risk expression statement; determining a first-level evaluation index, and determining a second-level evaluation index according to each risk expression statement; based on the first-level evaluation indexes, the second-level evaluation indexes and preset expert weights, determining the weight of each first-level evaluation index and the weight of each second-level evaluation index by adopting a group analytic hierarchy process; and determining the operation state of the gas boiler based on the weight of each primary evaluation index and the weight of each secondary evaluation index. According to the method, the operation state determination efficiency of the gas-fired boiler is improved.

The invention relates to the technical field of intelligent driving, in particular to an efficient real-time vehicle intelligent detection and recognition method and system based on YOLO-BOS, and the method comprises the steps: training a YOLO-BOS model through a vehicle sample image, and obtaining a vehicle detection and recognition model; the YOLO-BOS model comprises an input end used for preprocessing an input image, a backbone network used for carrying out feature extraction on the preprocessed image, a head network used for fusing the extracted image features, and a prediction output end used for carrying out classification prediction on the fused and output image features. The backbone network uses double-layer routing in a BRSA attention mechanism to adjust each spatial position weight in the image to screen and strengthen key features, and the head network uses full-dimensional dynamic convolution to expand a model receptive field and a perception range to capture related features in feature fusion; and extracting the category and bounding box information of the target vehicle in the to-be-detected image by using the vehicle detection and identification model. The method can adapt to complex and changeable conditions of road targets, and improves the vehicle detection and recognition precision.

The invention provides a quantum computing auxiliary analysis and marketing prediction method for private domain traffic pool data, which comprises the following steps of: acquiring and processing user behavior data, coding the user behavior data into a quantum state by using a quantum algorithm, extracting user behavior characteristics by using a quantum support vector machine (QSVM), and analyzing a periodic mode through quantum Fourier transform, so as to realize the quantum computing auxiliary analysis and marketing prediction of the private domain traffic pool data. According to the method, a marketing prediction model is constructed based on a quantum algorithm, model parameters are optimized by using a quantum annealing technology, the prediction accuracy is improved, and the method further comprises the step of carrying out classical verification and optimization on a quantum calculation result so as to realize personalized marketing. According to the invention, efficient analysis and precise marketing prediction of the private domain traffic pool are realized, and the pertinence and effectiveness of enterprise marketing strategies are improved.

The application provides a facial expression recognition method based on a double attention mechanism, named Adaptive Spatio-Excitation Transformer (ASET). This method introduces Adaptive Squeeze-and-Excitation (ASE) attention mechanism and spatial attention mechanism, and combines with Transformer network, which significantly improves the accuracy and efficiency of facial expression recognition. The core of the application is to recalibrate the channels of feature maps through ASE mechanism, and focus on the key areas of the face through spatial attention mechanism, so as to more efficiently extract expression features. In addition, the application also introduces visualization technology to highlight the key areas in the face image, helping to understand the decision-making process of the model. Experimental results show that the application has achieved excellent performance on multiple standard datasets such as FER2013, CK+ and JAFFE, and has significantly improved performance compared with existing technology. The application can be widely used in human-computer interaction, sentiment analysis and other fields, providing an efficient and accurate solution for facial expression recognition.

This invention provides an automatic SAR target detection method and apparatus based on adaptive feature focusing, belonging to the field of SAR image detection. The method includes: constructing a lightweight EOD-Net; after the backbone network extracts multi-scale features, the latent target perception branch PTBranch outputs a mask to coarsely locate the target region; and using a no-overlap feature map cropping NOFC to iteratively remove overlapping blocks, retaining only non-overlapping target feature blocks for the detection head to regress. During the training phase, a Gaussian-binary hybrid mask supervises PTBranch, and the focus loss and DICE loss are jointly optimized. This invention outperforms comparable methods such as YOLOv11 and Faster RCNN, achieving high-precision and efficient SAR image detection.

The invention discloses a voxel-oriented end-to-end 3D target detection method. The voxel-oriented end-to-end 3D target detection method comprises the following steps of A1, point cloud data input and data enhancement; a2, backbone network feature extraction: after the point cloud data is voxelized, the point cloud data is processed by two IDW-MSA modules, and rich feature information is extracted; and A3, detecting and classifying. According to the method, the IDW module is introduced, the large kernel convolution is used, the receptive field of the model is greatly increased, the MSA module is introduced, the attention mechanism is perceived by using neurons, additional parameters do not need to be added, the calculation efficiency is remarkably improved, and meanwhile high detection precision is kept; compared with the prior art, the method provided by the invention has the advantages that the target detection accuracy and speed under a complex background are remarkably improved.

The invention relates to the technical field of financial data analysis, and discloses a big data-based finance and taxation data analysis method, which comprises the following steps: firstly, collecting data, then cleaning, formatting and standardizing the collected data, and establishing a database so as to facilitate subsequent analysis and finance and taxation deep learning network establishment; step 2, respectively establishing a pre-training model and a deep learning model, respectively inputting the data of the database in the step 2 into the pre-training model and the deep learning model, performing knowledge distillation, extracting characteristics of finance and taxation data, and identifying a mode, thereby training and perfecting the pre-training model, generating a decision model, and performing decision making on the finance and taxation data. And carrying out finance and tax analysis and prediction by using the trained decision model. According to the big data-based finance and taxation data analysis method, various previous or public finance and taxation-related data, including financial statements, tax records, transaction records and the like, of other companies are stored by a finance and taxation data acquisition module through a data storage module, and a database is established.

The invention provides a corn early leaf recognition method, device and equipment and a storage medium, and belongs to the technical field of image recognition, and the method comprises the steps: integrating a FasterNet module in a C2F module of a backbone network of a YOLOv8 model, the FasterNet module comprising a partial convolution module, integrating an LSKNet network in a neck network, constructing an improved YOLOv8 model, and through a historical image data set comprising corn early leaves, carrying out the recognition of the corn early leaves. Training the improved YOLOv8 model to obtain a target detection network; and identifying the early leaf of the corn through the target detection network. According to the method, the Faster Net module is integrated in the C2F module of the backbone network of the YOLOv8 model, the LSKNet network is integrated in the neck network, and the backbone network and the neck network of the original YOLOv8 model are improved to identify the early corn leaves, so that the accuracy of identifying the early corn leaves is improved.

The invention discloses an ultra-high voltage power transmission line channel unmanned aerial vehicle inspection optimization and improvement method. The method comprises the following steps: S1, carrying out real-time image acquisition on a power transmission line channel based on an unmanned aerial vehicle aerial survey technology; s2, an adaptive median filtering and bilateral filtering combined algorithm is adopted to pre-process the image; s3, performing brightness equalization processing on the image and adjusting the image contrast; s4, performing multi-level feature extraction on the power transmission line channel image; s5, targets in the transmission line channel are detected based on the stable features and the dynamic features; according to the invention, optimization processing can be carried out on the unmanned aerial vehicle inspection image, multi-level feature extraction is carried out, a detection model is constructed, target detection and change detection of a power transmission line channel are realized, and hidden dangers are rapidly identified.

The application discloses a tomato disease and pest identification method based on an improved ControlNeXt lightweight classification model, and comprises the following steps: 1) model structure optimization; 2) lightweight module integration; 3) training strategy optimization; 4) performance verification; and 5) disease and pest identification. The tomato disease and pest identification method solves the outstanding problems in model complexity, calculation efficiency and actual deployment adaptability by improving the YOLOv8s-CLS model, and realizes the balance between accuracy and lightweight.

The application discloses a pipeline leakage detection method and device, the method comprises the following steps: acquiring video images and pressure data of multiple continuous time steps in a pipeline, fusing the video images and the pressure data to obtain fusion information; performing space-time feature extraction on the fusion information by using a space-time feature extraction network to obtain a feature vector; performing classification prediction on the feature vector by using each classification head in a trained multi-heterogeneous structure classification model to obtain pipeline state categories output by each classification head; and determining a final pipeline state category by using an entropy balance decision function according to the pipeline state categories output by each classification head and weights of each classification head, so that the accuracy of pipeline leakage detection is improved.

The application discloses a game performance analysis and running stability prediction system based on artificial intelligence, relates to the cross technical field of machine learning and game technology, and comprises a data acquisition and preprocessing module, a causal feature extraction module, a stability prediction module and a causal root cause positioning module; the data acquisition and preprocessing module collects game performance data in multiple channels and completes standardized processing to generate frame-level feature vectors, the causal feature extraction module builds a causal graph by improving a PC algorithm, filters features, and adaptively adjusts feature weights, the stability prediction module predicts abnormal probability by means of a causal time series convolution network and triggers early warning, and the causal root cause positioning module completes root cause searching and hierarchical judgment by relying on virtual causal intervention, wherein the application replaces the traditional experience-based feature filtering and correlation analysis mode by means of causal logic, saves the complicated operation of manual index filtering, can position multi-level performance causes layer by layer, and captures game lag and crash hidden dangers early.

The invention provides a gate slot sill foreign matter detection method based on machine vision. The method comprises the following steps: acquiring the last three frames of a real-time video of a gate slot sill of a target gate at a certain frequency each time; performing bilateral filtering preprocessing on the last three frames of the real-time video to relieve image noise caused by insufficient underwater light and turbid water; an improved detection model insensitive to a target position is adopted to perform foreign matter detection on the last three frames of a real-time video, and the influence of underwater foreign matter shaking and foreign matter position real-time change on the detection precision is relieved; whether foreign matters exist in the last three frames of the real-time video or not is judged, if no foreign matters exist, other gate detection work is carried out, and if foreign matters exist in more than one frame of image, foreign matter position information is transmitted back, and follow-up gate groove sill cleaning work is carried out; the foreign matter detection effect on the gate number and the gate groove sill is more remarkable in severe scenes such as insufficient underwater light and turbid water.

The invention discloses a model training method and device and a target equipment identification method. The method comprises the following steps: acquiring a plurality of image frames in a video stream; performing convolution operation on the image frame, and performing batch normalization processing on features obtained by the convolution operation to obtain a first target feature map; performing convolution operation on the first target feature map by using convolution kernels of different scales, and fusing output feature maps of the convolution kernels of different scales to obtain n layers of second target feature maps output by the convolution layers; performing multi-scale feature fusion on the second target feature map of each layer in the n layers of second target feature maps to obtain a first feature corresponding to the second target feature map of each layer, and performing weighted fusion on the first features to obtain a third target feature map; and training the deep learning model by using the third target feature map to obtain a target equipment identification model. The technical problem that the recognition precision of a related target equipment recognition model is low is solved.

This invention relates to a feature extraction model training and feature extraction method and apparatus, comprising: inputting a training image set into an initial model; extracting several local features from the training images by a local feature extraction layer; enhancing the several local features using an attention mechanism by a feature enhancement layer to obtain several enhanced local features; a global feature extraction layer aggregating features based on the several enhanced local features to obtain global features; and a classifier obtaining predicted classification parameters for the training images based on the global features. The model parameters of the initial model are adjusted according to the label parameters and predicted classification parameters of the training images to obtain a trained classification model. A feature extraction model for extracting global features is constructed based on the local feature extraction layer, feature enhancement layer, and global feature extraction layer in the classification model. This provides an end-to-end global feature extraction scheme, reducing algorithm complexity.

The invention provides a security equipment detection method and system based on an SGR-YOLO model, and belongs to the technical field of security equipment detection.The method comprises the steps that a security equipment image is obtained, the security equipment image is preprocessed, and the preprocessed security equipment image is divided into a training set, a test set and a verification set; training the SGR-YOLO model by using the training set and the verification set to obtain the weight of the trained SGR-YOLO model; and evaluating the trained SGR-YOLO model by using the evaluation index, and inputting the test set into the trained SGR-YOLO model to obtain a detection result of whether the constructor wears the safety equipment or not. According to the invention, the problems of poor lightweighting and low detection precision of the yolov5 algorithm during landing deployment in the embedded equipment are solved.

The application discloses a kind of diagnostic methods and systems for glenoid and humeral head defect area, belong to image processing technical field, the method includes: obtaining the medical image data of the glenoid and humeral head to be processed;The medical image data is input into the segmentation model of pre-setting, and the segmentation result of glenoid and humeral head is obtained;Based on the segmentation result, the defect area and defect proportion of glenoid are calculated;Based on the defect proportion, the defect degree of humeral head is determined.The diagnostic method and system for glenoid and humeral head defect area of the application provide a kind of efficient, accurate intelligent diagnostic auxiliary tool for clinic, with good application prospect and popularization value.

The application discloses a precoding matrix indication selection method based on a neural network. In the method, a user end first acquires a channel matrix of a downlink; then, the channel matrix is preprocessed by downsampling, segmented discrete Fourier transform and number domain conversion, so that input features suitable for neural network processing are obtained; then, the neural network is used for feature extraction and code word mapping on the preprocessed channel features, corresponding code word numbers are output for each subband, and the code word numbers are converted into precoding matrix indication (PMI) parameters; subsequently, the user end feeds back the PMI to a base station; and the base station maps corresponding precoding matrices according to the same predefined codebook and the received PMI, and the precoding matrices are used for downlink beamforming transmission. The application uses the neural network to replace a conventional traversal search type PMI selection process, reduces the calculation complexity of the user end under the premise of keeping compatibility with an existing codebook feedback mechanism, and improves the PMI selection efficiency.