22results about How to "Improve classification results" patented technology

The invention relates to the technical field of small sample image classification, in particular to a multi-modal small sample image classification method and system based on multi-scale dynamic feature fusion. The method comprises the following steps: acquiring a small sample image data set, and dividing the small sample image data set into a support set and a query set; and inputting the two features into a feature extraction network to extract global features, and inputting the features into a regional proposal network (RPN) to obtain corresponding local features. Extracting semantic information according to the category label of the support set data, and inputting the semantic information into a text encoder to obtain text features; inputting the global feature, the local feature and the text feature of the input image into an adaptive weight network, and performing dynamic feature fusion to effectively integrate multi-modal information; and inputting the fused multi-modal features into a classification model which takes a prototype network as a learning device and takes an MAML algorithm as a meta-learning device for classification prediction, finally training a network model by using a composite loss function, and performing image classification by using the trained classifier.

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 a multi-agent cooperative control system based on multi-mode brain-computer interface-visual tracking, and the system comprises the following modules: a signal collection and processing module which employs an 8-lead electroencephalogram cap to collect electroencephalogram original data, selects occipital region parts (PO7, PO8, P3, P4, Pz, O1, O2 and Oz) as an electroencephalogram data collection region, and carries out the EEG data collection; the feature extraction and classification module is used for extracting SSVEP signal features by using a space-time convolutional network, optimizing classification precision in combination with an integrated learning strategy, extracting EOG signals based on a Transform model, integrating an LMS adaptive filter and denoising the EEG signals in combination with wavelet transform, expanding the frequency tracking range of the adaptive filter, and reducing signal loss to the greatest extent; the cooperative control strategy design module is used for mapping the electroencephalogram signals into control instructions of different intelligent agents, and cooperative control over the unmanned aerial vehicle and the trolley is achieved; and the target detection and tracking module is used for optimizing an unmanned aerial vehicle visual tracking system based on a PP-YOLOE model and realizing accurate tracking and automatic adjustment of the trolley.

The invention relates to the technical field of brain-computer interfaces, and discloses a multi-interface fusion dynamic brain-computer interaction method based on frequency phase information multiplexing and a terminal. Based on an offline electroencephalogram experiment of a user, the most brain electrical stimulation classification number U of the user for achieving the optimal electroencephalogram stimulation classification effect, frequency-phase information of the brain electrical stimulation of the U classification and a first corresponding relation between the brain electrical stimulation of the U classification and electroencephalogram signal characteristics are obtained, and an interactive interface after multi-interface fusion is dynamically partitioned; the method comprises the following steps: acquiring Z sub-regions and brain control instructions included in each sub-region, distributing frequency-phase information of U-class brain electrical stimulation to each brain control instruction, acquiring brain electrical stimulation classification based on a first corresponding relation on the basis of brain electrical signal characteristics induced by a flicker rule of brain electrical stimulation corresponding to a certain brain control instruction when a user watches the brain control instruction, and determining a sub-region based on the user sight drop point coordinate, and outputting a brain control instruction in combination with the brain electrical stimulation classification and the determined sub-region. According to the invention, the brain-computer interaction work efficiency can be improved.

The invention provides a multi-source remote sensing data classification method, which belongs to the field of hyperspectral image processing in the field of remote sensing, and comprises the following steps: acquiring remote sensing data of various target objects; extracting shallow layer features of each kind of remote sensing data; obtaining a plurality of frequency characteristics of each kind of remote sensing data; fusing the same frequency features of all remote sensing data to obtain a plurality of same frequency fusion features; splicing the plurality of same-frequency fusion features to obtain a multi-source fusion feature; enabling the multi-source fusion features to sequentially pass through a frequency modulation layer and an attention layer which are superposed to obtain fused global features and local features; and weighting the global features and the local features in the spectral dimension to obtain a prediction classification result of the target object. According to the method, directional frequency characteristic decomposition and fusion of the multi-source data are realized, the multi-source fusion characteristics are extracted based on the multi-source remote sensing data to serve as the basis of classification, more comprehensive information is provided for classification of target objects, complex ground object characteristics can be fully captured, and the classification result is more accurate.

The embodiment of the invention discloses a self-supervised emotion recognition method based on electroencephalogram signals in a virtual reality scene. One specific embodiment of the method comprises the following steps: collecting original full-channel electroencephalogram data, and generating preprocessed full-channel electroencephalogram data; obtaining each differential entropy feature; generating each embedded feature; obtaining a time visible frame feature and a time masking frame feature; obtaining each piece of time code embedded information; obtaining each real two-dimensional coordinate, and generating each spatial position code; obtaining each space visible frame feature and each space masking frame feature; each piece of space code embedding information is obtained; obtaining each space decoding feature; obtaining reconstructed differential entropy features; determining a loss value, and determining a pre-training model; obtaining an emotion predictor; and obtaining an emotion category. According to the embodiment, the emotion recognition capability of the few-channel electroencephalogram data can be enhanced, and the problem that the emotion recognition accuracy is low due to individual differences can be solved.

The invention relates to a nuclear magnetic resonance image classification diagnosis method for adult diffuse glioma. The method comprises the following specific steps: step 1, respectively acquiring T2-fluid attenuation inversion recovery T2-FLAIR and contrast enhancement T1 weighted imaging CE-T1WI magnetic resonance images of glioma labeled by a doctor according to histological and genomics information; step 2, carrying out data preprocessing operation on the T2-FLAIR image data and the CE-T1WI image data; step 3, each pair of T2-FLAIR and CE-T1WI images after preprocessing is sent to a decision tree diagnosis model, and an Attention-DenseNet model is used for learning and extracting the characteristics of the T2-FLAIR and CE-T1WI images; 4, obtaining a prediction result based on a plurality of independently trained Attention-DenseNet models, fusing the outputs of the plurality of models by adopting a voting mechanism or a weighted average strategy, and generating a final classification decision; and 5, sending the obtained final classification to a platform, and making final diagnosis based on a classification result.

The invention discloses a multi-time-sequence PolSAR image labeling method and system based on a local-to-global time polarization relation inference network. The method comprises the steps that a multi-time-sequence PolSAR image is acquired and preprocessed; constructing an inference network model based on the local-to-global time polarization relationship; training a constructed reasoning network model based on a local-to-global time polarization relationship by using the preprocessed multi-time-sequence PolSAR image; and inputting an original multi-time-sequence PolSAR image into the trained reasoning network model based on the local-to-global time polarization relationship, and labeling the multi-time-sequence PolSAR image to obtain a classification result. The multi-time-sequence PolSAR image classification method can be used for achieving multi-time-sequence PolSAR image classification, and meanwhile, compared with a traditional model, the multi-time-sequence PolSAR image classification method can better capture change features of multi-time-sequence PolSAR data so as to improve classification precision.

This invention discloses a fruit quality detection method and system based on a fruit quality detection model. First, the original hyperspectral image R of the fruit to be detected is acquired and preprocessed. Then, the preprocessed hyperspectral image R2 is input into the fruit quality detection model to obtain the surface defect results and internal quality index values ​​of the fruit. Compared with the prior art, this invention can achieve simultaneous detection of the internal and external quality of the fruit at low cost and high efficiency, and obtain the fruit quality grade result.

The invention relates to the technical field of image recognition, in particular to an image feature extraction and recognition method, electronic equipment and a storage medium. Inputting the first image into a plurality of first CBS modules which are connected in sequence, and performing first feature extraction; inputting the first feature into a first ELAN module, and taking the output of the first ELAN module as a second feature; and finally, respectively inputting the second features into four intermediate layers which are sequentially connected in series, and taking the output of each intermediate layer as an image feature so as to obtain a plurality of image features. According to the embodiment of the invention, through multiple times of convolution and normalization, the results of the multiple times of convolution and normalization are superposed, multiple features are obtained, the levels and the number of reserved features are large, the classification result of the classifier is improved, and compared with a deep learning training model in the prior art, the number of needed training samples is small.

The present application provides a method and device for identifying the microstructure of a 2.5-dimensional woven composite material based on machine vision. The method provided by the present application includes: identifying the target woven composite material based on the top view and the structural tensor information identification method of the target woven composite material to obtain a first material classification result; determining the pixel range of the warp and weft yarns of the target woven composite material from the main view based on the first material classification result of the top view, and determining the number of pixels in each independent area based on the pixel range, thereby optimizing the first material classification result and obtaining a second material classification result; obtaining a third material classification result based on the main view recognition result and the side view recognition of the overlapping warp yarns, and combining the second material classification result and the third material classification result to obtain the recognition result of the microstructure of the target woven composite material. The method and device provided by the present application are used to accurately and efficiently identify the microstructure of woven composite materials.

The application provides a building fine-grained edge extraction and classification method, device, equipment and medium, relates to the technical field of remote sensing detection, and includes size cropping on the obtained building remote sensing image, and constructing a training sample set, inputting the training set into a plurality of instance segmentation networks selected in advance for training, obtaining a plurality of first building identification models, and obtaining corresponding first identification results output after inputting a test set; the first identification result is used as a label of the test set to expand and optimize the training of the training set, obtain a second building identification model corresponding to each first building identification model, and obtain corresponding second identification results output after inputting a test set; the plurality of second identification results are subjected to fine-grained mask fusion processing to obtain a building category, a confidence score, a building detection frame and a building fine-grained mask. The application can effectively improve the building boundary extraction precision and the classification result.

This invention provides a data cleaning method, comprising: calculating a first distance between the feature vectors of every two data points in an initial dataset, wherein the data in the initial dataset includes feature vectors and original label vectors; constructing a neighborhood set corresponding one-to-one with each data point based on the first distance; training an original classification model based on the initial dataset to obtain an intermediate classification model; inputting each feature vector into the intermediate classification model to obtain a corresponding predicted label vector; constructing a sample set based on a second distance between the original label vector and the predicted label vector; updating the corresponding data based on the sample set and the neighborhood set; determining whether the number of data updates has reached a preset number; updating the data again if the number of data updates has not reached the preset number; and forming the target dataset from the current data when the number of data updates has reached the preset number. This invention effectively cleans the initial dataset, making the data labels more accurate.

The present invention relates to the technical field of database construction, and in particular to a knowledge base construction method, device and medium for children's neurodevelopmental problems. The method comprises: generating initial text data of children's neurodevelopmental problems based on an existing database, cleaning and preprocessing the initial text data to obtain word segmentation results; converting vocabulary vectors according to the word segmentation results to obtain candidate vector sequences and identify subject semantic vocabulary vectors; analyzing the distribution of subject semantic vocabulary vectors and obtaining corresponding description levels; obtaining and marking vocabulary corresponding to each description level in the initial text data to form text ranges corresponding to each description level, obtaining storage priorities of subject semantic vocabulary vectors in combination with level truncation evaluation indicators, marking corresponding storage priorities based on text ranges, and completing the construction of a knowledge base; preventing the occurrence of similar symptoms leading to confusion in classification results, and enhancing the accuracy of knowledge data in the corresponding field of the knowledge base.

The invention provides a special-shaped household garbage classification method and system based on a knowledge graph, and relates to the technical field of artificial intelligence, and the method comprises the steps: collecting multi-dimensional parameter data of special-shaped household garbage, executing preprocessing and feature extraction to generate visual and material features, integrating and generating structured feature vectors, and converting the structured feature vectors into semantic feature representations, semantic feature representation is generated; constructing a junk knowledge graph, performing feature matching on the semantic feature representation and the knowledge graph, calculating an entity comprehensive matching degree, and screening to generate an instance candidate entity set; based on the instance candidate entity set and the knowledge graph, analyzing and reasoning an association path of the candidate entity and the garbage category, and calculating a credibility score of the association path; according to the method, the entity comprehensive matching degree, the association path and the credibility score are evaluated, and the garbage category is output, so that accurate classification of the special-shaped household garbage can be realized, the limitation of traditional matching based on a shape template is broken through, and the accuracy and adaptability of garbage classification are improved.

The present invention specifically discloses a self-supervised self-distillation meta-learning method for small sample remote sensing image scene classification, the method comprising the following steps: S1, based on self-supervised learning and using remote sensing image data to train a preset deep neural network to obtain an embedded network model; S2, embedding the embedded network model into a meta-learning framework for training to obtain a small sample classification embedded network model; S3, iteratively optimizing the small sample classification embedded network model based on self-distillation training, thereby completing self-supervised self-distillation meta-learning for small sample remote sensing image classification and obtaining a trained SMSR model. The present invention applies the embedded network model with feature extraction capability obtained by self-supervised learning training to a meta-learning framework for small sample remote sensing image classification training, and then obtains a trained SMSR model through self-distillation simplification, thereby realizing self-supervised self-distillation meta-learning for small sample remote sensing image classification.

The application discloses a text classification model training method, a text generation method and related equipment, the text classification model training method comprises the following steps: converting a question text into an embedding sequence through an embedding layer, effectively extracting coding features in the text through an encoding layer, capturing text features of different granularities by using a multi-scale convolution, enhancing the understanding ability of long text, and introducing a self-attention mechanism into the weighted processing of convolution features and pooling features, effectively improving the weight of important features, reducing the influence of redundant features, optimizing the accuracy of intent classification, finally generating a classification result through an output layer, and realizing model training according to the classification result to obtain a trained text classification model, effectively improving the accuracy of text classification, and the intelligent customer service system can take the classification result output by the trained text classification model as the basis for reply, thereby effectively improving the reply efficiency and reply accuracy of the intelligent customer service system under the condition of limited resources.

The invention discloses a dynamic classification-based model training method and a big data prediction method. The method comprises the steps of obtaining a first training subset, a second training subset and a prediction set; a plurality of classification models are trained according to a first training subset and a second training subset obtained through division according to the initial division value, and an optimal classification model is obtained according to the score value of each classification model; and using the optimal classification model to classify the training set and the prediction set, and further performing redundancy training on the prediction model to obtain a data prediction result. According to the method, the problems of feature dependence and high and low numerical value mixing existing in a traditional single classification model are solved by dynamically adjusting division values and selecting an optimal classification model; a data set, used for training, of the prediction model is expanded through a redundancy training method, the utilization rate of edge data is increased, and therefore a good prediction result can still be obtained under the condition that the data classification condition is poor; the method is more flexible and reliable.

The application discloses a kind of pedestrian trajectory classification methods based on semi-supervised random neural network.The method includes constructing pedestrian trajectory data;Pedestrian trajectory data is preprocessed;Different trajectory patterns are established according to pedestrian trajectory data;Finally, the combination algorithm of quantifiable minimum error entropy criterion and semi-supervised random neural network algorithm is used to classify pedestrian behavior.The application uses quantifiable minimum error entropy criterion to replace traditional mean square error criterion, and is combined with semi-supervised random neural network algorithm, which can further improve the classification ability of existing semi-supervised learning model, and also improves the accuracy of pedestrian trajectory classification.

The application discloses a kind of based on feature augmentation aero-engine fault diagnosis method, belong to aero-engine fault diagnosis technical field.The present application is directed to the saliency of aero-engine fault signal feature gradually reduces, it is difficult to extract useful information in engine actual operation and maintenance data, and then influence fault diagnosis accuracy problem.Original sample is carried out high-dimensional feature augmentation, and feature augmented sample is obtained;Again normalization processing is carried out, and then training sample set is constructed from normalized sample;Normal state sample and fault sample in training sample set are set different labels respectively;Training sample set is used to train fault diagnosis network, when reaching preset iteration number, training after fault diagnosis network is obtained;The operation data of aero-engine is collected, and normalized after processing is obtained after diagnosis data;Normalized after diagnosis data is input into training after fault diagnosis network, and aero-engine fault diagnosis result is obtained.The present application is used for aero-engine fault diagnosis.

The present invention relates to the technical field of medical statistical data, and in particular to a statistical data processing method and system based on intelligent medicine, including: collecting medical record data, medical images, basic information of patients and symptom information of medical images; extracting lesion features to obtain pathological feature values ​​of each medical image; constructing a symptom label sequence according to the pathological feature values, symptom information and basic information; obtaining a keyword set of medical record data; obtaining the importance of keywords according to the distribution of each keyword in the medical record data and the difference between the distribution of other keywords in the medical record data; screening to obtain a medical label sequence of each medical record data; classifying medical images, classifying medical record data according to the distribution difference between medical label sequences of different medical record data, and obtaining classification results of medical images and medical record data. The present invention can effectively improve the efficiency and accuracy of medical information retrieval.

This application provides a fault prediction model training method, fault prediction method, device, storage medium, and computer program product. The method includes: converting text data samples corresponding to one or more second devices into text semantic vectors through a first model; constructing triple samples based on the text semantic vectors through the first model, and querying a second model based on the triple samples, wherein the second model is deployed on other devices besides the first device; receiving the text semantic evaluation results output by the second model for the triple samples, and adjusting the model parameters of the first model based on the text semantic evaluation results to obtain a trained first model; and determining a fault prediction model based on the trained first model and a classification model obtained after training an initial classification model based on text data samples.