Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

25results about How to "Address imbalances" patented technology

Optimization method based on active reasoning in MEC enabled low-altitude element cosmic system

The invention belongs to the technical field of wireless communication, and discloses an optimization method based on active reasoning in an MEC-enabled low-altitude element cosmic system, which realizes low energy consumption and low delay of Internet of Things equipment on the premise of ensuring user service quality through joint optimization of a flight path and a resource allocation strategy of an unmanned aerial vehicle, and improves user experience. And meanwhile, an auxiliary means is provided for improving the economic benefits of operators. Aiming at the dynamic nature of the network, the uncertainty of information acquisition and the incompleteness of data use, an optimization problem is converted into a Markov decision process, an improved deep learning algorithm based on active reasoning is adopted, factors except the reward dimension are additionally considered on the guidance of dynamic learning, and the dynamic learning algorithm is optimized. And a personalized strategy is explored according to the preference of the intelligent agent.
Owner:XIAN UNIV OF POSTS & TELECOMM

Training method and system for cataract recognition model and storage medium

The invention relates to the technical field of medical image analysis, in particular to a training method and system for a cataract recognition model and a storage medium. The method comprises the following steps: acquiring cataract eye image samples and non-cataract eye image samples in different stages; lesion region segmentation and image synchronous simulation enhancement are carried out on the cataract eye image samples and the non-cataract eye image samples in different stages to obtain cataract lesion enhanced image samples and non-cataract enhanced image samples; extracting local detail features of the lesion and extracting structural features of a mask pair, and constructing a multi-task convolutional neural network learning framework to train a cataract recognition model and optimize focus weighted loss; and obtaining a cataract related key feature set and carrying out online incremental learning updating to generate a cataract recognition online incremental updating model. According to the invention, the accuracy and robustness of cataract recognition model training can be improved.
Owner:THE FIRST AFFILIATED HOSPITAL OF HEBEI NORTH UNIV

Unbalanced trajectory data anonymity removal method based on Granger causality

The embodiment of the invention provides an unbalanced trajectory data anonymity removal method based on Granger causality, which belongs to the technical field of data processing and specifically comprises the following steps of: 1, performing data equalization processing on anonymous and non-anonymous trajectory data sets aiming at the problem of imbalance of multi-source trajectory data; step 2, inputting the trajectory data set subjected to equalization processing into an improved bidirectional long-short-term memory model to respectively obtain anonymous trajectory representation and non-anonymous trajectory representation; 3, inputting the non-anonymous trajectory representation into a causal learning module, and extracting a causal structure existing in the trajectory data; 4, inputting the causal structure into a user ID balance prediction module, and training under the optimization of a loss function; and step 5, inputting anonymous trajectory characterization into the trained model to obtain a user and trajectory link result of the anonymous trajectory, and realizing trajectory de-anonymization. Through the scheme of the invention, the accuracy of the anonymity removal effect is improved.
Owner:湖南工商大学

A duplex stainless steel material and its application in preparing seawater heat exchanger

The present invention discloses a duplex stainless steel material and its application in the preparation of a seawater heat exchanger. The chemical composition of the duplex stainless steel material is, by mass percentage, C≤0.08%, Si≤0.8%, Mn 3‑5%, Ni 7‑10%, Cr 21‑23%, Mo 2.50‑3.50%, N 0.08‑0.20%, and Fe remainder. The duplex stainless steel material directly obtains a phase composition and microstructure with a ratio of ferrite to austenite close to 1:1 after laser rapid melting and solidification molding. Also disclosed is the application of the laser additively manufactured duplex stainless steel on complex pipe fittings of seawater heat exchangers. The present invention solves the problem that traditional composition duplex stainless steel is mainly ferrite after laser molding, and obtains the best phase composition and comprehensive performance; it also realizes the integrated molding of complex duplex stainless steel seawater heat exchangers, avoiding the problems of cumbersome assembly and errors of traditional split heat exchangers.
Owner:GUANGDONG INST OF NEW MATERIALS

Negative sample intelligent mixed training target detection method based on positive sample thermodynamic diagram guidance

The invention relates to the technical field of images, and provides a negative sample intelligent mixed training target detection method based on positive sample thermodynamic diagram guidance. The core technical problem of the invention is how to improve the accuracy and generalization ability of a target detection model under the condition of processing imbalance of positive and negative samples. The main scheme comprises the following steps: 1, preparing positive and negative sample data sets; step 2, generating a thermodynamic diagram, converting an abnormal region detected in the positive sample into the thermodynamic diagram, and generating the thermodynamic diagram by using a Gaussian blurred image processing method to highlight the abnormal region and smooth the edge; 3, processing data, designing an intelligent mixed training strategy, and taking a positive sample thermodynamic diagram as guidance information; and 4, evaluating and optimizing a result.
Owner:SICHUAN SHUJU INTELLIGENT MFG TECH CO LTD

Target detection method and device based on fusion mode, computer equipment and medium

The embodiment of the invention provides a target detection method and device based on a fusion mode, computer equipment and a medium, and relates to the technical field of automatic driving, and the method comprises the steps: obtaining self-collected visual data and first point cloud data; extracting second point cloud data corresponding to the category in the first point cloud data in the open source data set; integrating the first point cloud data and the second point cloud data to obtain third point cloud data; performing feature extraction on the visual data and the third point cloud data to obtain visual modal BEV features and point cloud modal BEV features; obtaining a modal offset matrix based on the visual modal BEV features and the point cloud modal BEV features; performing feature fusion on the visual modal BEV features and the point cloud modal BEV features based on a modal offset matrix to obtain fusion features; and performing target detection based on the fusion features. According to the invention, the problem of sample imbalance of the self-collected data set is solved, and the robustness and precision of the target detection method are improved.
Owner:WUHAN JIMU INTELLIGENT TECH CO LTD

Energy pile structure based on shaping phase change material backfilling and operation method thereof

The invention provides an energy pile structure based on shaping phase change material backfilling and an operation method of the energy pile structure. The energy pile structure sequentially comprises a pile core heat injection base material A, a heat removal heat shield base material B and a concrete shell from inside to outside. A heat injection pipe is arranged in the pile core heat injection base material A, and a heat removal pipe is arranged in the heat removal heat shield base material B; the heat extraction heat shield base material B is internally connected with the pile core heat injection base material A, and is externally connected with the concrete shell; the pile core heat injection base material A and the heat extraction heat shield base material B are both made of shaped phase change materials; and the heat removal heat shield base material B adopts asymptotic phase change, so that the problems that underground cold and heat loads are unbalanced, a ground temperature field is unbalanced, inter-pile heat interference is serious, underground heat transfer efficiency is low, and the heat and cold supply capacity and energy efficiency of a system are not expected and are reduced year by year when an existing shallow soil source heat pump energy pile supplies heat and cold are solved.
Owner:LANZHOU UNIVERSITY OF TECHNOLOGY

Diabetes prediction model construction method and prediction system based on ensemble learning

The invention belongs to the technical field of diabetes diagnosis, and particularly relates to a diabetes prediction model construction method based on ensemble learning and a prediction system. The method comprises the following steps: step 1, carrying out preprocessing and feature screening on a data set; 2, dividing a training set and a test set; 3, searching and extracting positive samples in the training set, randomly generating positive samples obeying normal distribution according to the mean vector and the covariance matrix of the positive samples in the training set, adding the newly generated positive samples into the training set, and continuously repeating until the proportion of the negative samples to the positive samples is balanced; and 4, learning the model by using the processed training set machine. The invention also provides a diabetes prediction system constructed by using the machine learning model. The method can effectively improve the prediction performance of the prediction model through the balance of the sample data and the integrated learning of the single classifier, and has a good application prospect.
Owner:WEST CHINA HOSPITAL SICHUAN UNIV

Traffic scene image small target detection method, device and equipment and storage medium

The application relates to the technical field of deep learning, in particular to a traffic scene image small target detection method and device, equipment and a storage medium. The method comprises the following steps: acquiring a traffic scene image and a detection result of a small target in the traffic scene image, generating a training data set according to the traffic scene image and the detection result of the small target in the traffic scene image; constructing a convolutional neural network based on a feature extraction module, a feature fusion module, a sample distribution module and a positioning classification module, and optimizing the sample distribution module and the positioning classification module of the convolutional neural network to obtain an optimized convolutional neural network; training the optimized convolutional neural network by using the training data set, updating network parameters of the optimized convolutional neural network based on gradient back propagation, and stopping the training until a preset training stop condition is met; and detecting small targets in the traffic scene image by using the trained convolutional neural network. Thus, the problems of small target sample imbalance and poor positioning capability in the related art are solved.
Owner:WUHAN UNIV

Near-infrared nonlinear discrimination method for fruit sorting

The invention discloses a near-infrared nonlinear discrimination method for fruit sorting, and relates to the technical field of artificial intelligence, and the training process of a classification discrimination model comprises the steps: obtaining a spectral data set of fruits, carrying out the preprocessing, carrying out the moving smoothing of the preprocessed spectral data set, and carrying out the classification discrimination. Mixing the preprocessed spectral data set with the spectral data set after moving and smoothing to serve as a sample data set; performing random sampling on the sample data set by using a Monte Carlo method so as to divide a training set and a first test set, training the discrimination model based on the training set, obtaining a classification result of random sampling, and performing statistics on a classification error rate of each sample; constructing a sample weight coefficient based on the probability density value corresponding to the chemical value of each sample and the classification error rate of each sample; constructing a non-linear discrimination model by using the sample weight coefficient, and optimizing the non-linear discrimination model; according to the near-infrared nonlinear discrimination method, the accuracy and reliability of fruit sorting are improved.
Owner:HEFEI INSTITUTE OF PHYSICAL SCIENCE CHINESE ACADEMY OF SCIENCES +1

An abnormal driving recognition method based on pedal data and facial expression multi-modal

This invention discloses an abnormal driving recognition method based on pedal data and facial expression multimodal data. The process is as follows: Step 1: Collect vehicle driving data and facial video data, and label real facial abnormal expressions to obtain original multimodal data composed of multiple original multimodal samples; Step 2: Train an original abnormal driving model based on the original multimodal dataset; Step 3: Perform noise diffusion on each original multimodal sample to obtain diffused multimodal samples, and select effective multimodal samples from the diffused multimodal samples to construct a multimodal dataset; Step 4: Process the multimodal dataset using the original abnormal driving model, and select prediction-consistent multimodal samples to construct a prediction-consistent multimodal dataset; Step 5: Train the original abnormal driving model again using the prediction-consistent multimodal data to obtain an abnormal driving recognition model; Step 6: Use the abnormal driving recognition model to predict the abnormal driving prediction result.
Owner:HEFEI UNIV OF TECH

Network intrusion detection enhancement method based on diffusion model

According to the network intrusion detection enhancement method based on the diffusion model provided by the invention, noise is gradually added to training data through the diffusion model, and then a denoising process is learned to model data distribution. Based on this, the method can enhance the training sample by synthesizing the virtual sample, especially the category with a small number of enhanced samples, thereby solving the problem of low detection precision caused by sample imbalance. Meanwhile, the method can enhance the adversarial samples, and introduces a certain degree of change to synthesize more diversified adversarial samples while maintaining the core features of the adversarial samples, thereby enhancing the defensiveness of the detection model to unseen adversarial samples. An existing diffusion model is improved to be suitable for intrusion detection data, and the data synthesis effect of the diffusion model is improved through adaptive nonlinear noise. Furthermore, an optimal diffusion model is provided, the random noise of the diffusion model adopts self-adaptive nonlinear noise, and the noise level is dynamically adjusted according to time.
Owner:UNIV OF ELECTRONICS SCI & TECH OF CHINA

A reinforcing cage hoisting device for pile testing of a cast-in-place pile

ActiveCN121778611BTo achieve the purpose of pulling backAddress imbalancesCranesSUSPENDING VEHICLEElectric machinery
The application discloses a reinforcing cage hoisting device for pile testing of a cast-in-place pile and relates to the field of hoisting devices.The reinforcing cage hoisting device comprises a hoisting vehicle, a movable hoisting frame is installed on the hoisting vehicle, and a counterweight is movably installed on the movable hoisting frame; further comprising a counterweight moving device, the counterweight moving device is installed on the hoisting vehicle, the movable hoisting frame and the counterweight are both installed on the counterweight moving device, and the counterweight moving device is used for moving the counterweight.It should be noted that in the embodiment of the application, when the hoisting vehicle is pulled up, the counterweight is automatically extended to the lifting side to form an active balance moment without external power, and meanwhile, the counterweight frame is rotated and unfolded to further shift the center of gravity of the hoisting vehicle; in addition, the earth boring motor drives the drill rod to rotate and drill into the ground to provide additional pulling force, effectively solving the problems of unbalance and overturning of the hoisting vehicle and bringing about remarkable effects of safety, economy and convenient operation.
Owner:KUNMING PROSPECTING DESIGN INSTITUTE OF CHINA NONFERROUS METALS INDUSTRY CO LTD

A matchstick figure posture intelligent monitoring method based on orthogonal amplification

PendingCN122598077AAvoiding the Risk of Privacy Leakageincrease diversity
The application discloses a matchstick person posture intelligent monitoring method based on orthogonal amplification, collects a monitoring scene video, extracts human key points, generates a non-private matchstick person skeleton sample, extracts four types of features of a skeleton structure, motion, support contact and data distribution, adaptively matches a plurality of Hadamard, DCT and Wavelet orthogonal sampling matrices for different skeleton states, realizes sample amplification through point-by-point multiplication of the matrices, and adds a skeleton semantic check to remove invalid samples; and a BiGRU-Attention, TCN or Transformer time sequence model is used to complete fall and long-time lying risk identification. The application solves the problems of traditional monitoring privacy leakage, high-risk posture sample scarcity and random amplification to generate invalid skeleton samples, and is applicable to home, nursing ward and old-age care institution scenes.
Owner:NANCHANG WOMENS HOME CARE CO LTD

Traffic scene image small target detection method and device, equipment and storage medium

The invention relates to the technical field of deep learning, in particular to a traffic scene image small target detection method and device, equipment and a storage medium, and the method comprises the steps: obtaining a traffic scene image and a detection result of a small target in the traffic scene image, generating a training data set according to the traffic scene image and the detection result of the small target in the traffic scene image; constructing a convolutional neural network based on the feature extraction module, the feature fusion module, the sample distribution module and the positioning classification module, and optimizing the sample distribution module and the positioning classification module of the convolutional neural network to obtain an optimized convolutional neural network; and training the optimized convolutional neural network by using the training data set, updating network parameters of the optimized convolutional neural network based on gradient back propagation until a preset training stop condition is met, and detecting a small target of the traffic scene image by using the trained convolutional neural network. Therefore, the problems of small target sample imbalance, poor positioning capability and the like in related technologies are solved.
Owner:WUHAN UNIV

Outbound survey user screening method, device, equipment, medium and program product

The invention relates to the technical field of communication, and provides an outbound survey user screening method and device, equipment, a storage medium and a program product, and the method comprises the steps: obtaining a feature data set, such as a classified feature, a call statistical feature and a voice outbound feature, of a to-be-screened user, inputting the information into a user identification model to identify a target user for outbound investigation; wherein the user identification model is obtained by performing iterative training on a preset TabNet model based on a target sample; the target sample is obtained by performing clustering analysis and random sampling on a sample data set constructed based on the historical feature data of the total user. According to the method, the target user for outbound investigation is identified through the TabNet model, and the sample data set is subjected to clustering analysis and sampling at any time in the model training stage, so that the problem of sample imbalance is solved, the identification accuracy of the outbound investigation user is improved, and the cooperation completion degree of the user for investigation and the outbound investigation efficiency are improved.
Owner:CHINA MOBILE GRP BEIJING +1

Battery energy optimization management method based on artificial intelligence algorithm

The embodiment of the present application provides a battery energy optimization management method based on an artificial intelligence algorithm, which relates to the field of artificial intelligence technology. The method includes: constructing a battery energy optimization management model based on a neural network of logic gates; collecting state variable parameters of each battery cell in the battery system; inputting the state variable parameters of each battery cell into the battery energy optimization management model, and the battery energy optimization management model performs judgment based on a logic formula, and makes charging and discharging decisions through logical operations combined with neuron weights to obtain the optimal charging and discharging decision strategy for each battery cell; executes the optimal charging and discharging decision strategy for each battery cell, and finally outputs the charging current and discharging current of each battery cell to ensure that the battery cell operates within a safe range and the battery power is balanced. Through this solution, the intelligent charging and discharging control strategy optimization of the battery system can be realized, the battery system life can be extended, and the performance of the battery pack can be improved.
Owner:SHAANXI WINDRIDERPOWER CO LTD +2

Cable processing equipment

The present invention relates to the field of cable processing technology, specifically to a cable processing device, which includes a device base, a hot melt mechanism and a molding mechanism; a feed hopper is provided on the upper side of the device base, and a control panel is provided on the front side of the device base; the hot melt mechanism includes a hot melt outer shell and a hot melt inner shell arranged in the hot melt outer shell, the lower end of the feed hopper extends into the hot melt inner shell, the outer wall of the hot melt inner shell is provided with a plurality of main electric heating rings, and an extrusion screw is arranged in the hot melt inner shell; the molding mechanism includes a molding outer shell and a molding assembly, the molding outer shell is located on the right side of the hot melt outer shell, the molding outer shell includes a transverse tube and a longitudinal tube, the transverse tube is assembled and connected to the discharge port of the hot melt inner shell by bolts; the molding assembly is located in the longitudinal tube. The present invention discloses a method for simplifying the steps required for disassembly and assembly of the molding mechanism while ensuring the sealing effect between the hot melt mechanism and the molding mechanism.
Owner:YANGGU XINHUAHUI CABLE CO LTD

Defect detection network construction method, anomaly detection method and system, and storage medium

The present application relates to a defect detection network construction method, anomaly detection method and system, and storage medium, wherein the construction method includes: obtaining a sample image in a preset open data set and a reference image of a standard product of an object to be detected; configuring a convolutional neural network module or multiple convolutional neural network modules of different scales and forming a defect detection network; using the sample image to train the main feature extraction model to obtain the corresponding network parameters, inputting the reference image into the main and slave feature extraction models to obtain the corresponding first feature vector and second feature vector respectively; constructing the loss function of the slave feature extraction model according to the first feature vector and the second feature vector and training and learning to obtain the network parameters of the slave feature extraction model, thereby configuring and forming a defect detection network. The defect detection network can complete the automatic extraction of features according to the input image information, effectively reducing the dependence on the experience of the staff in the process of product defect detection, and has practical value.
Owner:SHENZHEN HUAHAN WEIYE TECH

Dynamic and static potential balance method for reducing primary discharge risk of spacecraft high-voltage solar cell array

PendingCN119093858AImprove reliabilityReduce the worst floating potential differenceCosmonautic power supply systemsPhotovoltaicsCapacitanceGlass cover
The invention discloses a dynamic and static potential balance method for reducing the primary discharge risk of a spacecraft high-voltage solar cell array, and belongs to the technical field of spacecraft power supply system protection. The method comprises the following steps: firstly, obtaining local potential distribution of high-voltage solar cells according to spacecraft surface suspension potential and circuit structure characteristics of a high-voltage solar cell array; then calculating the maximum suspension potential difference among the substrates, the interconnection sheets and the glass cover sheets of different battery strings to obtain a potential difference needing to be balanced; and finally, according to the distributed capacitance and the bias time of the high-voltage solar cell array, calculating to obtain a dynamic bias resistance value, a static bias capacitance and a balanced suspension potential. According to the invention, the worst suspension potential difference of the spacecraft high-voltage solar cell array can be effectively reduced, the risk of primary discharge is greatly reduced, and the power supply reliability of the space high-voltage solar cell array is improved.
Owner:BEIHANG UNIV

De-anonymization method for imbalanced trajectory data based on Granger causality

The embodiment of the present invention provides a method for de-anonymizing unbalanced trajectory data based on Granger causality, which belongs to the field of data processing technology, and specifically includes: step 1, for the problem of unbalanced multi-source trajectory data, data balancing processing is performed on anonymous and non-anonymous trajectory data sets; step 2, the trajectory data set processed by balancing is input into the improved bidirectional long short-term memory model, and anonymous trajectory representation and non-anonymous trajectory representation are obtained respectively; step 3, the non-anonymous trajectory representation is input into the causal learning module, and the causal structure existing in the trajectory data is extracted; step 4, the causal structure is input into the user ID balanced prediction module, and trained under the optimization of the loss function; step 5, the anonymous trajectory representation is input into the trained model, and the user and trajectory link result of the anonymous trajectory is obtained, so as to realize the de-anonymization of the trajectory. Through the scheme of the present invention, the accuracy of the de-anonymization effect is improved.
Owner:湖南工商大学

A method for detecting fake reviews based on big data

This invention proposes a method for detecting fake reviews based on big data, comprising: acquiring a review dataset, wherein the data in the review dataset includes review texts from multiple users and corresponding user behavior data, and performing data preprocessing; extracting review text features from the review dataset, and extracting user behavior features and review behavior features from the user behavior data; defining relationship categories based on the review text and user behavior features, and constructing a multidimensional relationship enhancement graph (MREGC) based on multiple relationships; using a dynamic adaptive feature enhancement graph neural network to learn features on the MREGC relationship graph, obtaining the embedding representation of each node in the multidimensional relationship enhancement graph, and aggregating the embedding representations; inputting the aggregated features into a classifier to determine whether a review is a fake review, and outputting a fake review label if the review is classified as a fake review; this invention, by fusing residual networks and multi-relationship review graphs, fully considers the multidimensional relationships and deep features between reviews, enabling the model to more accurately identify fake reviews, and significantly improving the accuracy and robustness of fake review detection.
Owner:CHONGQING UNIV OF POSTS & TELECOMM

Job Search Data Processing Methods and Systems for Overseas Recruitment

This invention belongs to the field of data processing technology and provides a method and system for processing job application data for overseas recruitment. The method includes: analyzing job request texts uploaded by recruiting users to derive multiple key dimensions for judging job-person suitability, generating a job requirement feature vector; analyzing the relationships between these key dimensions and constructing a dynamic weight allocation network using a graph neural network; parsing each job request text in the set of uploaded job request texts to generate a job text feature vector; based on the dynamic weight allocation network, using semantic bridging to perform multi-dimensional joint similarity calculation between the job text feature vector and the job requirement feature vector to obtain the comprehensive suitability of each job request text; and removing job request texts with a comprehensive suitability below a preset threshold as content that does not belong to the requested business type. This invention enables intelligent and precise filtering of massive amounts of business texts.
Owner:BEIJING YONYOU XINFUSHE CLOUD TECH CO LTD

Training method and device for insurance policy recognition model, insurance policy recognition method and device

The present application provides a training method and device for a policy identification model, and a policy identification method and device, which belong to the field of artificial intelligence technology. The method includes: obtaining first sample data including positive policy samples and negative policy samples; extracting features from the first sample data to obtain target policy features; constructing sample feature data according to the target policy features; training a preset neural network model according to the sample feature data to obtain classification prediction results; performing performance evaluation on the neural network model according to the classification prediction results to obtain performance data; if the performance data is the first data, screening and processing the positive policy samples and the negative policy samples according to the preset quantiles and performance data to obtain second sample data; inputting the second sample data and the classification prediction results into the neural network model for training until the performance data is transformed from the first data to the second data to obtain the policy identification model. The present application can improve the training effect of the model.
Owner:CHINA PING AN LIFE INSURANCE CO LTD

Zone area load balancing power supply system and method based on energy storage unit

The invention provides a transformer area load balancing power supply system and method based on an energy storage unit, and the power supply system is provided with an independent energy storage unit, a shared energy storage unit, a switching network and a transformer area terminal, and is used for regulating and controlling a three-phase load in a power distribution transformer area in real time. When the power distribution terminal calculates that the load of a certain phase in the transformer area is too heavy, the charging and discharging power of the independent energy storage unit is adjusted by controlling the operation state of the independent energy storage unit connected with the phase line, and then the load level of the phase line is changed; when the load balance adjustment requirement exceeds the maximum adjustable range of the independent energy storage unit, the shared energy storage unit needs to be combined for further cooperative work; according to the load balance power supply method, through cooperative regulation and control of the independent energy storage unit and the shared energy storage unit, the depth of treating load imbalance of the transformer area is widened, occurrence of an extreme load of a certain phase in the transformer area can be effectively dealt with in real time, and the power supply reliability is improved.
Owner:SICHUAN PROVINCE AIRPORT GRP CO LTD