77results about How to "Improve classification performance" patented technology

The application discloses a movie classification system based on multi-modal deep representation collaborative federated learning, and the method comprises the following steps: firstly, initializing local model parameters of each client with public model parameters; then, the client updates the local model parameters according to the total loss on the local movie multi-modal data, obtains the local multi-modal deep representation, and uploads the local multi-modal deep representation and the local model parameters to the server; the server calculates the global multi-modal deep representation, obtains the initialization local model parameters of each client in the next global round through personalized parameter aggregation, and distributes the local model parameters to each client; finally, each client replaces the local model parameters. Except for the initialization, the above training process is repeatedly executed until the local model parameters of each client converge. For the to-be-recognized movie data, only the category thereof is recognized in the local client. While protecting the privacy, the application effectively improves the classification accuracy of each client on the movie multi-modal data.

The application provides a printed circuit board defect detection method and system based on deep learning, wherein the method comprises the following steps: inputting a PCB original image into a deep neural network for target detection reasoning; before the target detection reasoning, performing data preprocessing and background and edge extraction on the PCB original image; wherein the deep learning method aligns the PCB original image with a standard board image; after the alignment of the PCB original image with the standard board image, a deep target detection model is used for defect detection, and the deep target detection model is a two-stage target detection model of a comprehensive single-stage target detection algorithm; based on the deep target detection model, the defect detection of the PCB original image is completed, and a detection result is output. The defect detection scheme and process are optimized in efficiency; while ensuring the recall rate of PCB appearance defect detection, the detection accuracy is improved, and the academic research and application prospect are very broad, and the application can truly contribute to the PCB industry.

This invention proposes a method and system for improving a classification model for imbalanced multi-class samples, comprising: a training step, which involves analyzing the acquired training data, combining and grouping categories according to the data volume of each category and preset data grouping and partitioning rules, and further using the group classification model and the sub-category classification model of each group in a hierarchical classification model for layer-by-layer classification training; and a prediction step, which involves first loading the group classification model and the sub-category classification model of each group, then inputting the prediction data into the group classification model for classification prediction to determine the group label of each group, and then selecting the corresponding sub-category classification model according to the group label for final sub-category prediction. By improving the framework structure of the classification model, the classification effect is improved, solving the multi-class problem under imbalanced samples, improving classification effect and accuracy, and simultaneously enhancing the robustness and generalization ability of the model.

The invention relates to a spatial analysis single cell state modeling method and system based on domain self-adaption and layered fine tuning, and the method comprises the steps: obtaining multiple immunofluorescence images and single cell segmentation masks, and constructing a no-label data set; constructing a mask auto-encoder composed of a ViT encoder and a linear decoder, adding a classification token in front of the image, carrying out field adaptive training on the mask auto-encoder based on the unlabeled data set, learning the classification token, and obtaining a field adaptive weight of the ViT encoder; obtaining a labeled data set; constructing a state embedding generation model, wherein the state embedding generation model comprises a shared ViT backbone network and a two-stage classifier; a classification token is added in front of an image feature sequence in the labeled data set, hierarchical training is carried out on the state embedding generation model, and the classification token is learned; and inputting the cell image blocks into the trained state embedding generation model, outputting a classification result and cell state embedding, and carrying out interpretability analysis. Compared with the prior art, the method has the advantages that accurate cell classification can be realized, and cell state representation with biological interpretability can be generated.

This solution discloses a new method for identifying RNA pseudouridine sites. This method proposes to use a variety of feature representation techniques to extract sequence features, and then uses the SVM-RFE method for feature selection to compress the feature space and optimize the feature subset. The best feature set after feature selection is input into the kernel method KeMRF based on polynomial random forest to identify pseudouridine sites in the sequence. As a newly proposed classification method, compared with the traditional random forest, KeMRF not only optimizes the discriminant criterion for node splitting, but also combines with an easy-to-interpret kernel method, making the classification performance more superior. This method reduces the training time of the model, improves the classification performance of the model, and further enhances the accuracy of identifying pseudouridine sites.

This invention relates to the technical field of sorting retired lithium-ion batteries, and provides a sorting method for the cascade utilization of retired lithium-ion batteries, comprising the following steps: safety inspection of the retired batteries to be sorted; electrochemical impedance spectroscopy testing under preset conditions to obtain raw impedance data and construct an initial feature vector; parsing the initial feature vector to convert it into relaxation time distribution data to construct a multimodal feature set containing real data and enhanced data; inputting the multimodal feature set into a Transformer-MLSAE network, performing deep feature distillation and EIS-DRT feature fusion to extract a deep fused feature vector; classifying the batteries into performance level labels according to cascade utilization standards; training a regularized logistic regression classification model based on the deep fused feature vector and performance level labels to obtain a retired battery sorting model; obtaining the deep fused feature vector and inputting it into the trained retired battery sorting model to obtain the predicted performance level of the battery for cascade utilization.

The application provides an RGB-D image feature collaborative fusion method based on transfer learning and width learning, and comprises the following steps: obtaining an RGB-D data set, performing preliminary training through a neural network, and performing retraining in the data set after modifying the structure; after feature extraction, performing correlation analysis and fusion on RGB image features and depth image features; and using width learning to classify and identify the fused features. The application can reasonably fuse the features of RGB images and depth images, ensure that the feature information of color images and depth images can complement each other, improve the running speed of the system by using width learning, and finally make the classification result have higher accuracy and reliability.

This invention discloses a method for identifying unipolar and bipolar depression based on deep learning-based multi-state EEG fusion, comprising: Step 1, performing continuous wavelet transform on EEG signals in open and closed states respectively to obtain open-eye time-frequency maps and closed-eye time-frequency maps; Step 2, using a deep learning model to extract features from the open-eye and closed-eye time-frequency maps obtained in Step 1 to obtain feature vectors for open-eye and closed-eye states; then, fusing the feature vectors in open-eye and closed-eye states to obtain a multi-state fused feature vector; finally, using a deep learning classification network to classify and identify the multi-state fused feature vector to obtain the identification result, thus completing the identification. This invention uses a deep learning model based on EEG signals to perform three-class classification identification of unipolar depression, bipolar disorder, and healthy individuals, improving classification performance.

The invention relates to the technical field of speech processing, and particularly provides a disguise speech recognition method based on a Transform model and mixed features, which is carried out from two aspects of feature extraction and model establishment. A resonance peak parameter is calculated by adopting a cepstrum method, a cepstrum coefficient (GFCC) is obtained through a Gammatone filter bank, then the resonance peak, the GFCC and a difference coefficient of the GFCC are combined into a mixed characteristic parameter, and complementary correlation between mixed characteristics is mined. From the perspective of model establishment, the mixed features are used as the input of the model, and the Transform network model is used as the acoustic model of the voiceprint recognition system, so that the feature distribution is better fitted, the classification effect is remarkably improved, and the performance of the camouflage voice voiceprint recognition system is effectively improved. The problem of performance degradation caused by feature redundancy and modal noise in a traditional method is solved.

The invention belongs to the technical field of artificial intelligence, and discloses a medical staff emergency capability assessment method and system based on a deep neural network, and the assessment method comprises the steps: collecting the character information, basic information and emergency work recognition description information of medical staff; constructing a neural network model by taking the character information as input to output a feature vector; constructing a neural network model by taking the basic information as input to output a feature vector; splicing the feature vectors as input to construct a neural network model and output feature vectors; inputting the emergency work cognition description information into a text classification model to output a feature vector; splicing the obtained feature vectors to serve as input to construct a neural network model to output feature vectors; and classifying the finally obtained feature vectors to obtain an emergency capability evaluation result. The emergency capability assessment result obtained by the invention can provide a scientific basis for discovering emergency capability weakness of medical personnel and proposing a targeted training scheme.

The invention discloses a small sample data-oriented fruit external quality image classification method, and relates to the technical field of image classification. According to the method, a fruit external quality classification model is constructed and is mainly divided into three layers: a fruit classifier, a fruit defect recognizer and a fruit quality classifier. Firstly, the fruit classifier improves a Faster R-CNN model, a multi-scale subnet is added in a feature extraction part to adapt to fruit feature extraction of different scales, and the accuracy of fruit classification is improved; secondly, the main function of the fruit recognizer is to detect the classification layer to output the defect category of the fruit, and in order to better fit the marginalization or centralization of the feature of the defect, the pooling strategy of the CNN model is changed, the pooling mode is selected and changed in a targeted manner, and the defect classification effect is better realized; and finally, a fruit classifier is established through fruit feature dependence maximization, and accurate classification of fruits under small sample data is realized.

The application provides a ship target identification method and device based on an RCS sequence and electronic equipment, a signal is transmitted through a transmitting device of a radar system, and a feedback signal is received through a receiving device; an improved OS-CNN model that is pre-trained is acquired; the feedback signal is input into the trained improved OS-CNN model, and the object type of the returned feedback signal is obtained. The improved OS-CNN network is used in the field of radar target identification, the classification recognition rate is high, and the application has practical popularization value and application value. By introducing the advantages of offset convolution, attention mechanism and full-scale convolution, the improved OS-CNN network model can effectively extract the characteristics of target and interference RCS sequence data while maintaining light weight and high efficiency, accurate classification is realized, and the improved low-resolution radar intelligent target identification classification capability has important practical significance.

The application provides a multi-branch graph adaptive network for individualized motor imagery electroencephalogram signal classification and a method thereof. Through an adaptive matching technology, a graph convolutional neural network (GCN) is used to learn a mapping relationship between an EEG signal and an optimal time-frequency domain processing method, and an optimal processing method is matched for each individual. The multi-branch network comprises an original data branch, a CWT branch and an STFT branch, and corresponding feature extraction networks are respectively designed. A data enhancement method based on a super-resolution generative adversarial network is used to expand a training data set and improve the generalization ability of the model. Experimental results show that the method significantly improves the classification accuracy of individualized EEG signals and has practical application value.

The invention discloses a full-section pathological image classification method based on diffusion model feature transformation, and relates to the field of computational pathology. The method comprises the following steps: acquiring a paired HE dyeing image and IHC dyeing image, and extracting image features thereof; building a dynamic feature conversion network model, wherein the model comprises a feature encoder and a cross-modal dynamic diffusion module; the feature encoder comprises an HE stream and an IHC stream which are respectively used for extracting HE features and IHC features; and the cross-modal dynamic diffusion module takes the HE features as conditions, ensures the consistency of diagnosis semantics by comparing semantic bridging strategies, adaptively processes cross-modal distribution differences by using the frequency domain expert hybrid module, and finally generates target IHC features through a conditional denoising diffusion process. According to the feature conversion method, the IHC features with high quality and consistent semantics can be generated, the classification performance of a multi-instance learning framework is remarkably improved, an intermediate pixel image does not need to be generated, the calculation efficiency is high, and a new normal form is provided for calculation of biomarker prediction.

The invention discloses a bearing fault identification method based on multi-scale space-time synchronization attention and space-time alignment. The method comprises the following steps: carrying out sliding window segmentation and normalization preprocessing on an original vibration signal; constructing an adjacent matrix by using a K-nearest neighbor algorithm, and extracting local spatial features by using a graph convolutional neural network; bidirectional sequence features are extracted through a bidirectional gating loop unit network, and weighted fusion is carried out in combination with a global attention mechanism; and after space and time sequence features are spliced, fault prediction is realized through adaptive average pooling and a full-connection classifier. According to the method, graph modeling, GCN, BiGRU and an attention mechanism are fused, cooperative extraction of space-time double-path features is realized, the problems of weak space modeling and lack of time sequence dependence in a traditional method are effectively solved, and the bearing fault recognition precision and generalization ability in a multi-working-condition and strong-noise environment are remarkably improved.

The invention discloses an image classification method and system based on ordinal regression, and relates to the technical field of image retrieval. The method comprises the following steps: acquiring a multi-view semi-supervised ordinal regression data set; based on the data set, establishing a similarity graph for fusing neighbor information of all view angles under each view angle; constructing a multi-view semi-supervised ordinal number regression target equation based on the similarity graph; solving the ordinal number regression target equation based on an alternative optimization strategy to obtain an optimized classifier; and obtaining a classification label of the unknown sample by using the optimized classifier. The method can improve the image classification capability.

The application discloses a kind of pipe-flower gentiana spectroscopy classification method, system and equipment based on deep learning, belong to medicinal material identification field, method includes: S1.average spectrum data of pipe-flower gentiana sample is collected, forms one-dimensional spectrum sequence, and carries out standardization preprocessing;S2.suitable for pipe-flower gentiana one-dimensional spectrum sequence is constructed Multi-scale deep pyramid convolutional neural network;S3.pipe-flower gentiana spectrum data set is used to train the multi-scale deep pyramid convolutional neural network;S4.to be classified pipe-flower gentiana spectrum data is input into the multi-scale deep pyramid convolutional neural network that has been trained, and outputs its class label.The application improves the accuracy and robustness of pipe-flower gentiana spectrum classification by multi-scale feature fusion mechanism and deep feature mining ability, provides a new technical scheme and research idea for traditional Chinese medicinal material spectrum qualitative analysis.

An adversarial training method based on joint label smoothing can train a neural network model capable of resisting different attack perturbation sizes during the training process. The model can identify the existence of adversarial samples under large perturbation attacks and reject their output results, and can correctly identify the true category of samples under small perturbation attacks.

The application discloses a kind of reparameterization and stereoscopic coding brain disease classification method and system, it is related to computer medical image analysis technical field, including: S1.data acquisition step;S2.data processing step;S3.Establish RepBoTNet network model;S4.Establish RepBoTNet-CESA network model;S5.model training step;S6.classification step.The RepBoTNet-CESA network model constructed in the application combines the advantages that CNN network is good at capturing local information and transformer network is good at integrating global information, while obtaining excellent performance indicators, reduces the calculation cost;By constructing RepBoTNet-CESA network model, more rich and effective features can be extracted, and the region of interest is less likely to be lost, and when analyzing the whole brain structure, the accuracy of disease classification can be effectively improved.

This invention discloses a multi-view few-sample Android malware classification method based on optimal guidance matching, belonging to the field of information security technology. The invention includes constructing multi-view grayscale images, training a backbone network, optimal guidance matching classification, and dynamic fusion of multiple views. First, the method extracts permissions, APIs, components, and intent features of the Android application to construct multi-view grayscale images. Then, it trains a backbone network that integrates attention mechanisms and self-supervised rotation prediction to extract discriminative features with geometric structure awareness. Valid guidance samples are identified through optimal guidance matching, and category similarity scores are calculated. Finally, adaptive weights are generated based on view confidence, and the multi-view scores are dynamically fused to complete the classification. This invention improves the accuracy and robustness of malware family classification in few-sample scenarios, reduces noise interference, and is suitable for rapid and accurate identification of malware.

The application discloses a marine organism classification method based on multiple hierarchies, and adds a multiple hierarchy classification system with biological prior knowledge to a network model. The method mainly comprises three parts of hierarchy division, a multiple hierarchy network model and multiple hierarchy classification. Marine organisms are processed in hierarchies through hierarchy division, details of features of the same hierarchy are extracted by using the multiple hierarchy network, and the extracted hierarchical information is combined with a risk minimization strategy to optimize a classification result. The multiple hierarchy network model mainly designs a C-MBConv module and a fine classification module, so that the network model is more suitable for a multiple hierarchy classification task. In order to reduce error transmission caused by high-order categories, a risk minimization strategy is designed in the multiple hierarchy classification part, so that the possibility of error transmission is greatly reduced. The method can effectively improve the classification precision and has good application effect on data with multiple hierarchical categories.

This invention relates to the field of interference source analysis technology, and provides a method and system for interference source analysis in partial discharge measurement of large oil-filled equipment, comprising: Step 1, acquiring the original acoustic signature signal of the large oil-filled equipment in operation; Step 2, preprocessing the original acoustic signature signal, including filtering, multi-stage downsampling, and acoustic signature data enhancement; Step 3, extracting the acoustic features of the preprocessed acoustic signature signal, including time-domain features, frequency-domain features, and time-frequency-domain features; and then performing multi-domain acoustic feature fusion; Step 4, based on the acoustic features, constructing a multi-domain feature-driven interference source classification model to identify and classify different types of interference sources; Step 5, based on the interference source identification results, suppressing interference in the original acoustic signature signal and extracting effective partial discharge signals. This invention can perform interference source analysis more effectively.

The application claims a neural network backdoor sample and backdoor model detection method, specifically comprising: performing data enhancement on the obtained original data set, respectively training to obtain an original model and an enhanced model, extracting features of the original data on the original model and features of the corresponding enhanced data on the enhanced model, calculating the feature stability score of the data, and distinguishing and separating the sample as a benign sample or a backdoor sample according to the feature stability score. Based on the benign sample and the backdoor sample, a shadow model is trained, an energy function idea is introduced to optimize the query noise used to extract the shadow model features, the shadow model features are extracted using the optimized query noise, a meta-classifier is trained, and whether the target model is a backdoor model is detected. The application can better separate the backdoor sample and the benign sample; the energy function idea is introduced to reduce the number of shadow models required for training the meta-classifier, meeting the low-cost demand of users.

The application belongs to the field of graph computing and time series data prediction, and discloses a multi-dimensional asynchronous time series data classification method based on spatial structure graph learning, which considers the interaction between sensors, constructs a graph for multi-dimensional data at each time point respectively, uses a graph neural network to realize information update of each sensor data, fills in the missing observation values in the process of information propagation, and learns the mutual relationship between different time intervals by using a time interval-based Transformer mechanism. In the process of classifying asynchronous time series data, the application realizes mutual propagation of information between different sensors, fills in the missing observation values, improves the robustness of the algorithm under a large amount of missing data, considers the correlation of time series data in time series, and improves the classification effect of multi-dimensional asynchronous time series data.

This invention discloses a land use classification method based on intelligent fusion of multi-source data. It acquires multi-source geospatial data of the study area, including remote sensing imagery, street view imagery, building vector data, road network vector data, points of interest (POIs), mobile phone signaling data, and Weibo check-in data, forming a land use classification experimental dataset. It constructs remote sensing imagery features, street view imagery features, built environment features, POIs, and population distribution features for each study unit, forming a feature set for the land use classification task. The constructed features are paired to generate feature pairs composed of different modalities. A contrastive learning method is used to align the features of different modalities, and the aligned features are added to the feature set for the land use classification task. Land use classification is performed based on an XGBoost ensemble learning model, and five-fold cross-validation is used to verify the model's effectiveness. This invention improves the accuracy of land use classification.

The application discloses a marine organism classification method based on a hierarchical neural network model, which automatically extracts high-quality features of data through the construction of the hierarchical neural network model, and solves the problem of marine organism recognition; the hierarchical neural network model comprises a visual tree construction module, a full-channel learning variable-dimension partial convolution module and a relative attention module; first, two hierarchical structures are constructed based on unsupervised learning and prior knowledge, and feature relationships and biological relationships are fitted respectively; a backbone network vertically designs a hierarchical neural network based on the full-channel learning variable-dimension partial convolution module and the relative attention module according to the complexity of features, coarse classification assists fine classification, marine organism classification is realized, and the scheme can improve the marine organism classification accuracy, and meanwhile, the balance of the accuracy and the parameter quantity and other network performances can be ensured.

The application belongs to the technical field of physiological signal analysis, and specifically discloses an eye movement representation construction system and method based on visual pattern evolution embedded learning. An individual representation construction model based on eye movement visual exploration pattern evolution embedded learning is used to process eye movement trajectories, so as to obtain an individual eye movement representation classification result. The individual representation construction model comprises a time sequence cyclic encoder for capturing time sequence dependent relationships in eye movement to obtain eye movement time sequence dynamic characteristics. A visual exploration pattern discovery module obtains an eye movement visual exploration pattern set from visual exploration patterns. A personalized pattern combination module uses a personalized router and an adaptation layer selector to select pattern combinations from the eye movement visual exploration pattern set, and optimizes the pattern combinations to obtain personalized pattern combination representations. A classifier receives the personalized pattern combination representations to obtain the individual representation classification result. The application solves the problem that existing methods cannot capture specific combinations of shared patterns in each individual, thereby limiting the generalization ability for new individuals.

The present application relates to the field of artificial intelligence and the field of brain-like intelligent chip, in particular to a synapse weight updating method in event-driven type chip, a chip and an electronic device, the method comprises the following steps: in the training of a deep convolutional neural network of an event-driven type chip, adopting a BP-STDP learning rule to update the weight of each synapse of the convolutional layer of the deep convolutional neural network, and adopting a DFA-STDP learning rule to update the weight of each synapse of the fully connected layer of the deep convolutional neural network. The method can make the neuromorphic chip realize maximum parallel computing, greatly improve the processing speed and enhance the performance of the chip.

The application provides an open-vocabulary aerial image target detection method based on collaborative quality perception, constructs a labeled basic category dataset and an unlabeled new category dataset of aerial images, constructs an SQAPN target detection model, the target detection model comprises a backbone network and an RPN network with a spatial alignment perception module, an alignment consistency loss function is introduced in the training process of the target detection model to supervise the training of the spatial alignment perception module, a confidence modulation module is introduced in the inference stage to obtain a comprehensive quality score, the target detection model is trained by using the dataset to obtain a trained target detection model, in the inference stage, the aerial image to be processed is input into the trained target detection model, and a comprehensive quality score and a candidate box are output, candidate box screening is completed through non-maximum suppression (NMS), and finally, the detection result is output. The application can effectively improve the open-vocabulary aerial image target detection precision and robustness.