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

14results about How to "Preserve accuracy" patented technology

Double-state six-channel polarization multiplexing metasurface design method based on phase change material

The invention discloses a dual-state six-channel polarization multiplexing metasurface design method based on a phase change material. The method comprises the following steps: designing a structural unit based on the phase change material; a double-state six-channel polarization multiplexing metasurface is constructed by using the designed structural unit; the geometric structures and the orientation angles of the structural units are optimized by adopting a screening algorithm, and the optimized double-state six-channel polarization multiplexing metasurface can record six kinds of irrelevant phase information into the same double-state six-channel polarization multiplexing metasurface. The method is based on the metasurface Jones matrix regulation and control theory, comprehensively utilizes the geometric phase and the transmission / resonance phase of the structural unit, is suitable for most types of metasurface unit structures, has wide applicability, can be applied to the fields of metasurface optical encryption, multifunctional wavefront regulation and control and the like, and has important practical value.
Owner:SOUTHEAST UNIV

Method for adaptively relieving MPI noise in IMDD system and IMDD system

The invention discloses a method for adaptively relieving MPI noise in an IMDD system and the IMDD system, and belongs to the technical field of optical communication. The method is applied to a receiving end and comprises the following steps: acquiring a received signal, and calculating probability density distribution of the received signal; according to the probability density distribution of the received signal, obtaining two quantization level values at each trough as a threshold level pair; dividing data points of the received signal into trustworthy points and stray points according to the threshold level pair; after the trustworthy point is judged, a recovery signal corresponding to the trustworthy point is obtained, and a judgment error is calculated; mPI noise estimation and MPI noise compensation are carried out on the spurious points according to the judgment errors of the trustworthy points, and recovery signals corresponding to the spurious points are obtained; and synthesizing the recovery signal corresponding to the reliable point and the recovery signal corresponding to the stray point to obtain a final output signal. The MPI noise of stray points in different intervals in the IMDD system link can be accurately compensated in a self-adaptive mode.
Owner:HUAZHONG UNIV OF SCI & TECH +1

Network security system based on deep learning

The invention relates to the technical field of networks, in particular to a network security system based on deep learning. A network security system based on deep learning comprises a data arrangement module which is used for updating virus data in a database and anti-virus software for processing the corresponding virus data, and obtaining data of each owned server. And the port detection and modification module is used for analyzing the dangerous data in the database by using a fuzzy reasoning algorithm based on the current server condition and the dangerous data in the database, combining the dangerous data with a traditional accurate control method to obtain an analysis result, and modifying the state of each port of the server according to the analysis result. According to the method, the database is updated by adopting big data, analysis is carried out according to the conditions of the servers and the access conditions of the ports, and whether the log data are normal or not is detected and judged by analyzing the log data.
Owner:JIAXING XIUSHUI ECONOMIC & INFORMATION COLLEGE

Wind power cluster wake optimization method and device based on model predictive control

ActiveCN121689299BAccurately describe mutual interference relationshipsImplement real-time scrolling optimizationWind motor controlMachines/enginesTime domainLoop control
This application relates to the field of wind power predictive control technology, and discloses a method and apparatus for optimizing the wake of a wind power cluster based on model predictive control. The method includes: collecting the operating state variables and meteorological environment information of each wind turbine in the entire wind farm; inputting the operating state variables and meteorological environment information into a comprehensive prediction model; and outputting the future wake of each wind turbine in the entire wind farm. N p The power generation sequence at each time point; construct the objective function and constraints, and combine them with future... N p The power generation sequence at each moment determines the future power output of each wind turbine. N p The sequence of yaw angles at each moment; based on the future N p The value of the first moment in the yaw angle sequence controls the wind turbine to perform yaw actions, and forms a rolling closed-loop control according to the rolling time-domain optimization mechanism. In this way, the accuracy of the physical mechanism is preserved, while also possessing the nonlinear fitting capability of deep learning, which can improve the accuracy of overall power prediction and take into account both high-precision physical field description and fast response of cluster control.
Owner:OCEAN UNIV OF CHINA

Live broadcast method and device based on private video transmission protocol

The invention relates to the technical field of communication. The invention provides a live broadcast method and device based on a private video transmission protocol, and the method comprises the steps: receiving a live broadcast request from a client, and initiating a pull stream request according to a terminal mark in the live broadcast request; based on the established data stream transmission channel, pulling protocol video data stream data from the corresponding terminal; the streaming media pushes the private protocol video data stream to the protocol conversion service; and sending the generated video stream address as a back-to-source address to a distribution service to generate a CDN playing address, auditing the live broadcast content based on the generated CDN playing address, and selecting the live broadcast content passing the content auditing by a user through various client applications to watch. According to the invention, the private protocol video data stream is pushed to the protocol conversion service, the video data stream converted by the private protocol can be obtained, the occupied broadband is small, the video data stream is not influenced by the network environment, and the playing fluency is further improved.
Owner:CHINA UNICOM ONLINE INFORMATION TECHNOLOGY CO LTD

A positioning method based on 5G base station

The present invention discloses a positioning method based on 5G base stations, and belongs to the technical field of positioning of 5G base stations. The present invention constructs a signal feature matrix and a signal gap coefficient matrix according to the signal characteristics of each 5G base station. The signal feature matrix reflects the signal characteristics of each 5G base station, and the signal gap coefficient matrix reflects the gap between the signal feature sequences reached in sequence. The difference in signal characteristics at adjacent moments enhances the feature distinction between different 5G base station positions. A multi-fusion positioning network is used to process the signal feature matrix and the signal gap coefficient matrix to obtain an initial position, so that the multi-fusion positioning network combines static characteristics and dynamic change information to accurately estimate the terminal position, while reducing the influence of instantaneous noise and outliers, improving the stability of the positioning result, and then using the terminal positions at multiple historical moments to correct the initial position, further improving the positioning accuracy.
Owner:四川易景智能终端有限公司

Video editing methods, apparatus, electronic devices, and storage media

This application relates to the field of computer vision technology, and provides a video editing method, apparatus, electronic device, and storage medium. The method includes: performing a first inverse denoising process on a first original noise latent variable at the current time step based on original video prompts to obtain a second original noise latent variable, and using the second original noise latent variable as the first original noise latent variable for the next time step; performing a second inverse denoising process on a first target noise latent variable at the current time step based on target video prompts to obtain a second target noise latent variable; and obtaining a third target noise latent variable based on the background data region of the second original noise latent variable and the second target noise latent variable, and using the third target noise latent variable as the first target noise latent variable for the next time step. The video editing method, apparatus, electronic device, and storage medium provided by this application can perform video editing tasks efficiently and at low cost.
Owner:CHINA MOBILE JIUTIAN ARTIFICIAL INTELLIGENCE TECHNOLOGY (BEIJING) CO LTD +1

Fast adaptive loop filtering algorithm based on class merging

The invention relates to the field of video coding, and discloses a fast adaptive loop filtering algorithm based on class merging, which comprises the following steps of: dividing image brightness components into a plurality of filtering classes; based on the rate distortion optimization cost, gradually reducing the number of categories by adopting a staged strategy of coarse screening and fine merging, merging from an initial category to a preset target category, further merging until only one category is left, and recording the merging process; the filter coefficient is updated for the category after each time of merging, matrix calculation is simplified through linearization cost estimation, and the number of iterations is reduced; further selecting an optimal filter set based on a rate distortion optimization cost traversal merging process; in the filtering parameter output stage, the combined filtering categories and the corresponding filter coefficients are packaged into a self-adaptive parameter set, and application of a decoding end is facilitated. According to the algorithm, on the premise that the image quality is kept, the coding efficiency is remarkably improved, and the algorithm is suitable for video coding application scenes with high requirements for real-time performance.
Owner:上海芯开技术有限公司

Method for removing multipath noise from underwater sonar images based on geometric probability

The present invention relates to the technical field of sonar image noise removal, and specifically discloses a method for removing multipath effect noise from underwater sonar images based on geometric probability, the steps of which include: pre-processing and denoising the sonar image, clustering the echo points of the image after pre-processing and denoising, and removing multipath echo effect points from the image after clustering. The method of the present invention is used to detect and remove multipath effect points on a single two-dimensional sonar image, which not only retains the accuracy of the entity pixels, but also plays a role in greatly reducing the multipath effect, achieving a significant noise reduction effect; the image processed by the present invention is used for three-dimensional reconstruction, and the result is greatly improved.
Owner:HARBIN INSTITUTE OF TECHNOLOGY (SHENZHEN) (INSTITUTE OF SCIENCE AND TECHNOLOGY INNOVATION HARBIN INSTITUTE OF TECHNOLOGY SHENZHEN)

Distribution network load transfer control method and system based on neural network decision distillation

The present invention discloses a distribution network load transfer control method and system based on neural network decision distillation, which belongs to the technical field of power grid operation optimization. The method includes collecting distribution network data and constructing a distribution network load transfer control data set; converting the distribution network load transfer control data set into graph data; training a graph neural network model based on the graph data to extract high-dimensional features of the distribution network load transfer control data set, obtain switch state prediction results and output load transfer control strategies; using knowledge distillation technology to use the original input of the graph neural network model and the high-dimensional features and switch state prediction results extracted from the graph neural network model to train a decision tree model respectively, minimizing the difference between the output of the decision tree model and the output of the graph neural network model; using the trained decision tree model to match the specific features and decision results of the power grid state, and providing a decision basis for the distribution network load transfer control. The present invention can more accurately predict the operation state of the power grid.
Owner:CHINA ELECTRIC POWER RESEARCH INSTITUTE CO LTD

Test paper information input method applied to dot matrix pen intelligent homework

InactiveCN119623415AQuick EntryEntry is accurateSemantic analysisQuestion analysisDot matrix
The invention relates to the technical field of smart homework, in particular to a test paper information input method applied to dot-matrix pen smart homework, and the method comprises the steps: carrying out the question segmentation and question structured labeling based on an electronic format test paper, determining a test paper structured analysis result, and then, storing the test paper structured analysis result into a database; based on the test paper structured analysis result and the test paper in the electronic format, question analysis and information input are carried out, online test paper is obtained, and the content of the online test paper is consistent with that of the test paper in the electronic format. And then, based on the test paper structured analysis result and the code laying format, carrying out code laying processing on the test paper in the electronic format to obtain a code laying test paper on which a dot matrix pattern is laid. According to the method, question segmentation and question structured labeling of the electronic-format test paper are realized through an automatic means, question analysis and information input are automatically performed based on segmentation and question structured labeling results, and the online test paper is automatically generated, so that time consumption and error risks caused by manual input are avoided, and the efficiency is improved. Rapid and accurate input of test paper information is realized.
Owner:BEIJING TOGETHER EDUCATION TECH DEV CO LTD

A scratch detection method based on deep convolutional neural network and image segmentation

The present invention discloses a scratch detection method based on deep convolutional neural network and image segmentation. A multi-feature fusion module is designed according to the scratch morphological characteristics, added to the deep convolutional neural network structure, to identify and locate the scratches, and extract the scratch prediction frame image; the principal component points of the scratch prediction frame image are calculated using the principal component analysis method, and the scratch prediction frame image after adaptive mean segmentation is grown in 8 neighborhoods with the principal component points as the growth points, so as to achieve accurate segmentation of scratch pixels. The present invention can identify, locate and segment scratches with high precision, and provide convenience for subsequent precise measurement of scratch size.
Owner:BEIHANG UNIV

Multi-dimensional identity recognition method and device, electronic equipment and program product

ActiveCN116912906BPreserve accuracyimprove accuracyPattern recognitionStreaming data
The application provides a multi-dimensional identity recognition method and device, electronic equipment and program product, and relates to the technical field of artificial intelligence. The method comprises the following steps: collecting video stream data containing a to-be-recognized object, extracting multi-dimensional feature information of the to-be-recognized object based on a first target image in the video stream data, recognizing the identity of the to-be-recognized object according to face information, if the recognition fails, performing target tracking on the to-be-recognized object based on a second target image before the first target image in the video stream data to determine whether a third target image containing the identity recognition result of the to-be-recognized object exists in the second target image, and if not, recognizing the identity of the to-be-recognized object according to body key point information. Through face recognition, target tracking and identity recognition based on body key points, the high accuracy of face recognition is retained, and when face recognition fails, identity recognition can be accurately performed based on body key points, so that the balance between the high accuracy and the easy failure of face recognition is achieved.
Owner:CHINA MOBILE GRP BEIJING +1

Corn kernel quality prediction method and system based on near infrared spectrum and calibration migration

The invention discloses a near infrared spectrum and calibration migration-based corn kernel quality prediction method, which comprises the following steps of: firstly, carrying out de-orthogonalization processing on a source domain spectral data matrix and corresponding chemical component content values by using orthogonal projection O-PLS of a latent structure; through orthogonal projection transformation, non-correlation variables and correlation variables of a spectrum data matrix (an independent variable) and a chemical component content value (a dependent variable) are separated, so that a part irrelevant to the dependent variable in the independent variable is effectively eliminated. And then, projecting the latent structure matrix of the target domain data set to the latent structure matrix of the source domain data set by using a segmented direct standardization PDS method, thereby realizing segmented standardization processing of the whole projection process. And finally, establishing mapping from the latent structure matrix space of the target domain data set to the latent structure matrix space of the source domain data set to obtain a transfer matrix, thereby reducing the complexity of the transformation process method and enhancing the interpretability while keeping the accuracy of the feature space transformation from the target domain spectrum to the source domain spectrum.
Owner:GUANGXI UNIV