31results about How to "Reduce redundant features" 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 the technical field of data processing, and discloses a service prediction method, device and equipment based on an LNM large numerical model, and the method comprises the steps: carrying out the preprocessing and multi-scale decomposition of internal historical service data and external macroscopic, mesoscopic and microscopic associated data, and obtaining long-term trend feature data and short-term fluctuation feature data; inputting a deep service association network for feature processing to obtain first comprehensive feature data; obtaining second comprehensive feature data through feature importance score and feature dependency relationship analysis; inputting an LNM large numerical model fused with multi-scale gating prediction to carry out service prediction, and obtaining a service prediction result; according to the method, distributed learning is carried out, iterative optimization is carried out through the local evaluation module and the local participation module, a target business prediction model is obtained, deep fusion of multi-dimensional features is realized, and the capture capability of an LNM large numerical model on data features is improved.

The invention provides a cancer patient gene methylation feature screening method based on DQN reinforcement learning, which converts a gene methylation feature screening problem into a reinforcement learning problem by constructing a deep Q network model, and dynamically selects the most representative methylation feature by taking maximization of prediction accuracy as a target. The method comprises the following steps: firstly, preprocessing gene methylation data of a cancer patient, then continuously optimizing a feature selection strategy in a multi-round training process through a DQN model, and finally screening out key methylation features closely related to cancer diagnosis and prognosis. The method can effectively reduce the feature dimension, improves the generalization ability of the model, and has wide application prospects in early diagnosis, personalized treatment and prognosis evaluation of cancers.

The invention belongs to the technical field of deep learning, and particularly discloses a data augmentation method for a visual language navigation task of an unmanned aerial vehicle, and the method comprises the steps: determining a path-action sequence-visual observation pair; obtaining a path-image feature-text feature pair; obtaining a combined path-image feature-text feature pair; constructing a training data set for training a projection layer according to the visual observation and the combined path-image feature-text feature pair; training the projection layer by using the training data set; obtaining a path-image feature embedding-text feature embedding-instruction description pair; and screening the path-image feature embedding-text feature embedding-instruction description pair to obtain augmented data. According to the method, the dual-scale visual representation capability of the dual-scale image Transform model and the large language model can be utilized at the same time, and the quality of the generated instruction is improved. The problem that in the prior art, generated natural language navigation instructions are poor in quality is solved.

The invention relates to the technical field of deep learning, in particular to a migration attack resisting method based on a regularization constraint feature layer, which comprises the following steps: data preprocessing and regularization constraint are carried out, an initial sample is prepared, and regularization constraint is applied to remove low-frequency components; generating an initial confrontation sample, and generating the initial confrontation sample by using an existing iterative attack algorithm; dividing a neural network feature layer, and identifying and dividing the feature layer of the neural network model; feature layer regularization loss function design: designing a regularization loss function for each feature layer; feature layer confrontation sample generation: generating a confrontation sample on each feature layer; evaluating the mobility of the adversarial samples, and evaluating the mobility of the adversarial samples of each feature layer on other models; according to the anti-migration attack algorithm based on the regularization constraint feature layer, disturbance is added to the optimal migration level of the statistical result, and then an anti-migration sample with better mobility is generated.

The invention discloses a hyperspectral image classification method and system. The method comprises the following steps: acquiring a hyperspectral image; preprocessing the hyperspectral image to obtain a preprocessed image; extracting spectrum, space and frequency domain features from the preprocessed image; superpixel segmentation is carried out on the preprocessed image, each superpixel of the image is regarded as a node, a graph structure is constructed, and global features of the graph structure are extracted; obtaining fusion features of the hyperspectral image according to the global features and the local features of the graph structure; and determining a classification result of the hyperspectral image according to the fusion features of the hyperspectral image. And the classification identification effect of the hyperspectral image is improved.

The invention belongs to the technical field of intelligent diagnosis of four-stroke diesel engines, and particularly relates to a PSO-RF intelligent diagnosis method and system for thermal faults of a marine diesel engine. The method comprises the following steps: designing a diesel engine simulation model by using simulation software; constructing a data set and dividing the data set; performing noise processing and normalization processing on the data in the data set; extracting characteristics which can best represent diesel engine faults in the data set; a random forest fault diagnosis model is constructed, parameter optimization is performed on the random forest model by using a particle swarm optimization algorithm, a PSO-RF intelligent diagnosis model is finally obtained, and fault diagnosis is performed by using the PSO-RF intelligent diagnosis model. The system comprises a diesel engine simulation unit, a data acquisition unit, a data processing unit, a feature extraction unit and a fault diagnosis unit. According to the method, automatic fault positioning and diagnosis can be realized, sailors are helped to quickly and accurately identify problems, the dependence on expert experience is reduced, and the diagnosis precision is remarkably improved.

The invention discloses an automatic driving vision multitask sensing method and system based on aerial view features, and belongs to the technical field of automatic driving. Comprising the following steps: acquiring images from a look-around camera of the automatic driving system and extracting features; a multi-task distillation module is adopted to cooperatively predict dense depth features, semantic features and content features, and a dense depth prediction result and a dense semantic prediction result are generated; splicing the content feature and the dense semantic prediction result, and generating a semantic enhanced aerial view feature by using the dense depth prediction result as a depth weighted value of the content feature; optimizing aerial view features; predicting the distribution probability of occupied grid categories, fusing the probability distribution of foreground categories and optimized aerial view features; and generating a target detection prediction result according to the fused features. According to the method, the relevance between the target detection task and the occupied grid prediction task is fully mined, the model performance is improved, deployment redundancy caused by multiple independent tasks is reduced, and therefore the training and operation and maintenance cost is reduced.

The invention discloses a short-term electricity price prediction method and system based on an MIC-TCN-Transform model, and the method comprises the steps: obtaining electricity price prediction data from different data sources, and integrating the electricity price prediction data into a unified format and structure; performing integrity, continuity, consistency and rationality check on the data, cleaning unqualified data, performing feature extraction and selection on a data set, performing normalization processing on all the data, and segmenting the data set into a training set, a verification set and a test set; the method comprises the following steps: defining a TCN model and a Transform model, and creating an MIC-TCN-Transform model; an optimizer and a loss function are configured, the MIC-TCN-Transform model is trained by using the training set, and the MIC-TCN-Transform model is evaluated by using the verification set; and the test set is inputted into the trained MIC-TCN-Transformer model, and a prediction result is obtained. Reasonable configuration and utilization of renewable energy sources can be effectively promoted, and the reliability and stability of a power system are improved.

A table grape detection method applied to agricultural robot picking comprises the following steps: step 1, loading data of a shot grape image data set into a model, performing mosaic data enhancement operation on an image to obtain richer data samples, and then performing subsequent feature extraction operation; 2, extracting features of the preprocessed image to generate a feature vector; 3, performing multi-scale fusion on the extracted features; step 4, performing target prediction on the multi-scale feature map output by the network and calculating a loss function; 5, judging whether the training achieves the optimal training effect or not, and carrying out a new round of iteration or outputting a model result; and step 6, performing target detection operation by using the trained model, and deleting redundant prediction frames by using a non-maximum suppression method. The method improves the recognition precision and recognition speed of table grapes in a complex scene.

The present invention belongs to the technical field of deep learning, and specifically discloses a data augmentation method for unmanned aerial vehicle visual language navigation tasks, including: determining a path-action sequence-visual observation pair; obtaining a path-image feature-text feature pair; obtaining a merged path-image feature-text feature pair; constructing a training data set for training a projection layer according to visual observations and the merged path-image feature-text feature pair; training the projection layer using the training data set; obtaining a path-image feature embedding-text feature embedding-instruction description pair; screening the path-image feature embedding-text feature embedding-instruction description pair to obtain augmented data. The present invention can simultaneously utilize the dual-scale visual representation capability and large language model of a dual-scale graph Transformer model to improve the quality of generated instructions. The problem of poor quality of natural language navigation instructions generated in the prior art is solved.

The invention relates to the technical field of text classification and grading evaluation, in particular to a specific text classification and grading evaluation system and method based on an information element extraction technology. The method comprises the steps that an extraction unit performs key element field extraction on text data based on a large model fine tuning technology; the standardization unit carries out standardization processing on each extracted element field; and the evaluation unit constructs a text classification and grading evaluation system through a decision tree based on the normalized fields. According to the technology, the accuracy and the coverage rate of key element field extraction can be improved through the large model fine tuning technology and the model, the model can be helped to better understand the context in a text, so that the situations of false report and missing report are reduced, the extraction quality is improved, the fine tuning technology can enable the model to better adapt to a new data set, and the extraction efficiency is improved. Even if the data set is slightly different from the pre-training data set, good performance can be kept, and the robustness of the model facing texts of different styles or formats can be enhanced.

The present invention relates to the technical field of charging load analysis, and specifically to a method and system for predicting the charging load of an electric vehicle based on data correlation, comprising the following steps: collecting multidimensional data from multiple data sources; associating and marking data with strong correlation; constructing a feature vector, and considering the data with associated marking in the process of constructing the feature vector; using the constructed feature vector as input to establish a charging load prediction model; inputting the multidimensional data collected in real time into an optimized charging load prediction model to obtain a charging load prediction result for the electric vehicle; and outputting the result. The present invention performs dimensionality reduction processing on the feature variables after comprehensive evaluation, constructs a feature vector, and in the process of dimensionality reduction, associates and marks the feature variables with strong correlation and assigns a higher weight, fully retains the key features, and improves the validity of the model input data.

The invention discloses a point cloud classification and segmentation method, and particularly relates to a point cloud classification and segmentation method based on adaptive convolution and self-attention, and the method comprises the following steps: S1, reading three-dimensional point cloud data, carrying out the sampling and preprocessing of the point cloud data, and inputting the preprocessed point cloud data into a model; s2, inputting the three-dimensional point cloud data into an adaptive residual convolution module, and performing feature extraction; s3, combining the plurality of local features obtained in the step S2, then passing through a multi-layer perceptron, and finally passing through a self-attention module to obtain advanced features; and S4, carrying out maximum pooling operation and average pooling operation on the advanced features, splicing the advanced features to obtain high-dimensional features, inputting the high-dimensional features to a full connection layer for dimension reduction, and outputting classification and segmentation results of the three-dimensional point cloud data.

The invention discloses a change detection method and system based on multi-scale fusion and integrated kernel adaptive learning. The method comprises the following steps: constructing a Siamese sensitivity enhanced feature extraction network for extracting dual-tense features; establishing a multi-scale dual-time fusion attention module to fuse dual-time images of all branches; defining that a nested decoder is formed by a jump connection mode to reserve high-dimensional semantic details and fine-grained positioning details; constructing an integrated kernel adaptive learning module to fuse output features of four levels, and fusing multi-scale semantic information; and extracting an optimization result, and outputting a change region. According to the method, the false change phenomenon caused by feature redundancy in the feature extraction process can be well relieved, important components in the dual-time features can be more effectively recognized, the feature expression diversity is enhanced, and the omission ratio in a small change area is reduced by capturing subtle changes in the image.

The application is suitable for the technical field of computer application, and provides a two-dimensional code image feature extraction method and device based on a CNN-MALT hybrid architecture, which comprises the following steps: inputting a target two-dimensional code image into a CNN preprocessing module of a preset lightweight image feature extraction network to generate first structure perception features; inputting the first structure perception features into a MALT encoder of the preset lightweight image feature extraction network, performing feature extraction on the first structure perception features based on a proxy attention mechanism to generate deep features; and inputting the deep features into a feature fusion module of the preset lightweight image feature extraction network to generate target features. Thus, the structure features of the two-dimensional code image are perceived by the CNN preprocessing module, the deep features are generated by the MALT encoder based on the proxy attention mechanism, and the deep features are fused by the feature fusion module, so as to ensure that the calculation complexity is reduced while the two-dimensional code image feature extraction accuracy is improved.

The embodiment of the invention provides an image segmentation model training method and device, an image segmentation method and device and electronic equipment, and relates to the technical field of image segmentation. The method comprises the following steps: inputting an annotated image and an unannotated image into an image segmentation model to respectively obtain an annotated prediction result and an unannotated prediction result, inputting an unannotated strong disturbance image into the image segmentation model, processing the unannotated strong disturbance image by a discarding module to obtain a strong disturbance prediction result, inputting an unannotated weak disturbance image into the image segmentation model, and obtaining a strong disturbance prediction result. The method comprises the following steps of: obtaining a strong disturbance prediction result, processing the strong disturbance prediction result through an attention enhancement module to obtain a weak disturbance prediction result, and carrying out iterative optimization according to a strong disturbance prediction result, the weak disturbance prediction result, an unlabeled prediction result and a labeled prediction result to obtain a mature image segmentation model. Through combination of the discarding module and the attention enhancement module, rich features and important features can be automatically learned under different data distributions, so that the segmentation accuracy of the model in a complex clinical scene can be improved.

A YOLOv5 marine fish target detection method based on attention mechanism and DropBlock, comprising the following steps: step 1: acquiring a target fish data set; step 2: labeling and dividing the acquired data set; step 3: constructing a YOLOv5 marine fish target detection model with attention mechanism and DropBlock; step 4: using the constructed marine fish target detection model to detect fish. The purpose of the present invention is to solve the technical problem that the existing fish detection method has a large range of marine fish activities and a variety of living environments, and the fish photographed underwater are deformed, and a clearer fish target cannot be captured, which affects the accuracy of fish detection, and the single convolutional neural network is easily disturbed by image noise when extracting fish features, and the fish cannot be quickly detected, thereby affecting the speed and accuracy of the target detection method.

The present invention discloses a deep learning wafer solder joint detection method based on improved YOLOV5, the method comprising: 1) preparing a wafer solder joint data set; 2) constructing an attention mechanism module CCANET; 3) constructing a YOLOV5 network incorporating the attention mechanism; 4) introducing a Ghost module; 5) training the improved YOLOV5 network. The present invention improves the accuracy of wafer solder joint detection while using fewer network parameters. Under the same conditions, not only more wafer solder joints can be detected, but also blocked wafer solder joints can be detected.

The invention discloses a DRF-Net-based small target detection method under a view angle of an unmanned aerial vehicle. Constructing an unmanned aerial vehicle aerial photography small target detection data set; a frequency domain guide feature enhancement module FCSA is introduced in the shallow layer stage of the backbone network; a multi-frequency reconstruction module MFRB is introduced in the high-level stage; constructing DRF-Neck in a neck structure, generating a saliency graph S based on MFRB output features, and introducing saliency-driven deformable convolution SAR-DCN, content aware recombination S-CARAFE and a dynamic feedback controller DFC to form a closed-loop optimization mechanism of saliency generation-self-routing convolution-dynamic feedback; and finally, the features are input into a decoupling detection head, network parameters are optimized through joint optimization of cross entropy loss and AHIoU loss, and a small target detection result in an unmanned aerial vehicle aerial photographing scene is output by using non-maximum suppression in a reasoning stage. Compared with the prior art, the method has the advantages that the detection precision on the unmanned aerial vehicle aerial small target data set is improved, and the method can be applied to the fields of unmanned aerial vehicle inspection, urban security monitoring, traffic target detection and the like.

The invention discloses a method, device and system for constructing a credit risk assessment model and a storage medium. In the scheme, a first training sample set is obtained, and the first training sample set comprises a plurality of first features and credit risk scores corresponding to corresponding samples. A Pearson correlation coefficient between each first feature and the credit risk score is calculated. And screening out a plurality of second features from the plurality of first features based on a Pearson correlation coefficient between each first feature and the credit risk score. And constructing an initial credit risk assessment model based on the plurality of second features. And iteratively optimizing the initial credit risk assessment model based on a simulated annealing algorithm to obtain a final credit risk assessment model. According to the method, feature selection is performed through the Pearson correlation coefficient, and parameter optimization is performed by using the simulated annealing algorithm, so that the selected features are more comprehensive and are prevented from falling into a local optimal solution at the same time, and the accuracy of the high credit risk assessment model is improved.

This invention discloses a method, apparatus, and medium for predicting lymph node metastasis. The method involves acquiring a three-dimensional image of a lymph node to be detected and determining the region of interest (ROI) within the image. Feature extraction is performed on the ROI to obtain local features corresponding to the three-dimensional lymph node image. The ROI data is then expanded to obtain pixel blocks to be processed, and feature extraction is performed on these pixel blocks to obtain global features corresponding to the three-dimensional lymph node image. Based on the local and global features, the lymph node metastasis prediction result for the three-dimensional lymph node image is determined. This invention achieves a comprehensive description of the features of the three-dimensional lymph node image through the complementarity of local and global features, solving the problem of low accuracy in lymph node metastasis prediction and improving its reliability. Furthermore, the local features only include the features of the lymph node itself, reducing redundant features.

The present invention discloses a method for semantic segmentation of indoor scenes based on an improved full convolutional neural network. First, a convolutional neural network is constructed, wherein the hidden layer includes 5 neural network blocks, 5 feature re-extraction convolutional layer blocks, 5 block attention convolutional blocks, 12 fusion layers, and 4 upsampling layers; the original indoor scene image is input into the convolutional neural network for training to obtain a corresponding semantic segmentation prediction map; then, the loss function value between the set consisting of the semantic segmentation prediction map corresponding to the original indoor scene image and the set consisting of the one-hot encoded image processed from the corresponding real semantic segmentation image is calculated to obtain the optimal weight vector and bias term of the convolutional neural network classification training model; the indoor scene image to be semantically segmented is input into the trained convolutional neural network classification training model to obtain a predicted semantic segmentation image. The advantage of the present invention is that the semantic segmentation efficiency and accuracy of indoor scene images are improved.

This invention discloses a lightweight asphalt road defect detection method and system with partial multi-scale attention, belonging to the field of computer vision and intelligent transportation technology. This invention is applicable to the real-time detection and precise localization of defects such as cracks and potholes. It employs PMA modules, SCConv modules, and the inner-AIoU loss function to construct the required asphalt road detection model, achieving accurate detection of the road image to be detected. The model constructed in this invention, while maintaining its lightweight advantage, achieves significant improvements in speed, substantial increases in defect detection accuracy, and diversification of detected defect categories, fully meeting the needs of real-time detection. Furthermore, the detection model constructed in this invention also possesses good portability and scalability, making its integration with hardware devices more convenient.

The invention discloses a contour missing target feature extraction and detection method and system in severe weather, and the method comprises the steps: firstly obtaining a low-quality video in severe weather, and extracting the high-dimensional features of a video image; searching robust information of a contour missing region; adaptively searching significant information of different-scale contour missing targets, and extracting multi-scale features; the extracted multi-scale features are processed, and cross-level information interaction is carried out on the features of different scales; then extracting relevance among the contour region, the contour missing region and the background region; then predicting the category and positioning information of a contour missing target; and finally, eliminating a redundant detection frame to obtain a final contour missing target detection result. According to the method, the target with the missing contour can be rapidly and accurately detected in severe weather, and intelligent monitoring in complex environments such as haze, rainstorm and low light can be realized or the robustness of an unmanned driving system can be enhanced.

The invention relates to the technical field of image recognition, and provides an electric power fitting shielding image recognition method and system, and the method comprises the steps: obtaining the historical image data of an electric power fitting; preprocessing the acquired historical image data of the electric power fittings, and constructing a first recognition model and a historical image data set; establishing a second identification model; inputting a historical image data set in the second identification model for training, and outputting the category and the shielding condition of the electric power fitting; real-time digital image data of the electric power fittings are acquired and preprocessed through image acquisition equipment, and the electric power fittings are automatically identified by using an electric power fitting shielding identification model. According to the method, through cooperative work of the double models, the identification accuracy of the electric power fitting is effectively improved, in addition, the shielding condition can be identified, the installation position and angle of the electric power fitting can be conveniently inferred subsequently, and therefore more accurate data support is provided for maintenance and management of an electric power system.

The invention relates to a sparrow search algorithm-based stereoscopic vision comfort feature extraction system and method, and belongs to the field of stereoscopic vision comfort feature extraction. The system comprises a data preprocessing module, a feature extraction module and a feature optimization module. According to the method, EMD and Hilbert transform in an HHT method are used for adaptively extracting time-frequency features of EEG signals, and then a spatial filter in a CSP method is used for obtaining time-frequency features of multiple spatial dimensions. The two combined methods not only can adaptively obtain the time-frequency characteristics, but also can reveal the interaction between different brain regions in the stereoscopic vision transmission process, and more comprehensively understand the stereoscopic vision comfort characteristics based on the EEG signals. The SSA algorithm is utilized to simulate foraging behaviors and anti-capturing behaviors of sparrows and an early warning mechanism, so that the features can be accurately and quickly optimized, the optimal feature combination can be selected, the time-frequency feature redundancy phenomenon is reduced, and the precision and the performance of the stereoscopic vision comfort classification model are improved.

The application discloses a bearing steel grinding burn identification method based on multiple magnetic parameters, and relates to the technical field of nondestructive testing.The method first constructs a training set and a test set containing multiple samples, and uses an X-ray stress meter to calibrate the burn state of the training set samples to obtain training set burn samples; the electromagnetic parameters of the training set burn samples are acquired through a micro-magnetic multi-parameter detector, and a judgment index feature based on a physical model is extracted; subsequently, the features are reduced in dimension by using a stacked auto-encoder, and the low-dimensional features are clustered by using a K-means clustering algorithm to automatically calibrate the state of the training set samples and determine a burn judgment threshold; finally, for the test set samples, whether burn occurs is judged by comparing the low-dimensional feature vectors thereof with the threshold value.The application can nondestructively and quickly acquire magnetic characteristic signals, effectively detect grinding burn and reduce misjudgment, and has the advantages of fast detection speed and high precision.

The invention belongs to the technical field of image processing, and particularly relates to a low-light image detection algorithm based on multi-scale feature extraction, which comprises a multi-scale feature extraction module, a loss function and a space-to-depth sampling module, and is characterized in that the multi-scale feature extraction module is added into a feature extraction network of an enhancement model; the loss function comprises an exposure control loss function, a color constancy loss function, an illumination smooth loss function and a mixed loss function, and the space-to-depth sampling module is composed of a space-to-depth layer and a non-convolution step length layer; multi-branch step-by-step pixel feature extraction is carried out on a multi-level feature extraction network, space and channel reconstruction convolution is used for replacing standard convolution operation in the last-level feature extraction stage, redundant features are reduced, calculation complexity is reduced, and meanwhile the correlation degree between pixels can be well improved.

The present invention discloses a method and system for extracting and detecting features of targets with missing contours in bad weather. First, low-quality videos in bad weather are obtained, and high-dimensional features of video images are extracted; then robust information of the missing contour area is searched; significant information of targets with missing contours of different scales is adaptively searched, and multi-scale features are extracted; the extracted multi-scale features are processed, and cross-level information interaction is performed on features of different scales; then the correlation between the contour area, the missing contour area, and the background area is extracted; then the category and positioning information of the missing contour target are predicted; finally, redundant detection frames are eliminated to obtain the final missing contour target detection result. The present invention can quickly and accurately detect targets with missing contours in bad weather, realize intelligent monitoring in complex environments such as haze, heavy rain, and low light, or be used to enhance the robustness of unmanned driving systems.